Publications
2022
|
Human visual Attention modelling is a persistent interdisciplinary research challenge, gaining new interest in recent years mainly due to the latest developments in deep learning. That is particularly evident in saliency benchmarks. Novel deep learning-based visual saliency models show promising results in capturing high-level (top-down) human visual attention processes. Therefore, they strongly differ from the earlier approaches, mainly characterised by low-level (bottom-up) visual features. These developments account for innate human selectivity mechanisms that are reliant on both high- and low-level factors. Moreover, the two factors interact with each other. Motivated by the importance of these interactions, in this project, we tackle visual saliency modelling holistically, examining if we could consider both high- and low-level features that govern human attention. Specifically, we propose a novel method SAtSal … |
|
Classifying and analyzing human cells is a lengthy procedure, often involving a trained professional. In an attempt to expedite this process, an active area of research involves automating cell classification through use of deep learning-based techniques. In practice, a large amount of data is required to accurately train these deep learning models. However, due to the sparse human cell datasets currently available, the performance of these models is typically low. This study investigates the feasibility of using few-shot learning-based techniques to mitigate the data requirements for accurate training. The study is comprised of three parts: First, current state-of-the-art few-shot learning techniques are evaluated on human cell classification. The selected techniques are trained on a non-medical dataset and then tested on two out-of-domain, human cell datasets. The results indicate that, overall, the test accuracy of state-of … |
|
Code Clone Detection (CCD) is a software engineering task that is used for plagiarism detection, code search, and code comprehension. Recently, deep learning-based models have achieved an F1 score (a metric used to assess classifiers) of 95\% on the CodeXGLUE benchmark. These models require many training data, mainly fine-tuned on Java or C++ datasets. However, no previous study evaluates the generalizability of these models where a limited amount of annotated data is available. |
|
We propose NullSpaceNet, a novel network that maps from the pixel-level image to a joint-nullspace, as opposed to the traditional feature space. The features in the proposed learned joint-nullspace have clearer interpretation and are more separable. NullSpaceNet ensures that all input images that belong to the same class are collapsed into one point in this new joint-nullspace, and the input images of different classes are collapsed into different points with high separation margins. Moreover, a novel differentiable loss function is proposed that has a closed-form solution with no free parameters. NullSpaceNet architecture consists of two components; 1) a feature extractor backbone (i.e., the convolution and pooling layers), which is used to extract features from the input, and 2) a nullspace layer, which maps from the pixel-level image to the joint-nullspace. This novel architecture and formulation results in a … |
2021
|
Background/Foreground separation is a fundamental and challenging task of many computer vision applications. The F-measure performance of state-of-the-art models is limited due to the lack of fine details in the predicted output (ie, the foreground object) and the limited labeled data. In this paper, we propose a background/foreground separation model based on a transformer that has a higher learning capacity than the convolutional neural networks. The model is trained using self-supervised learning to leverage the limited data and learn a strong object representation that is invariant to changes. The proposed method, dubbed TransBlast, reformulates the background/foreground separation problem in self-supervised learning using the augmented subspace loss function. The augmented subspace loss function consists of two components: 1) the cross-entropy loss function and 2) the subspace that depends on Singular Value Decomposition (SVD). The proposed model is evaluated using three benchmarks, namely CDNet, DAVIS, and SegTrackV2. The performance of TransBlast outperforms state-of-the-art background/foreground separation models in terms of F-measure. |
|
The Visual Object Tracking challenge VOT2021 is the ninth annual tracker benchmarking activity organized by the VOT initiative. Results of 71 trackers are presented; many are state-of-the-art trackers published at major com-puter vision conferences or in journals in recent years. The VOT2021 challenge was composed of four sub-challenges focusing on different tracking domains:(i) VOT-ST2021 challenge focused on short-term tracking in RGB,(ii) VOT-RT2021 challenge focused on" real-time" short-term track-ing in RGB,(iii) VOT-LT2021 focused on long-term track-ing, namely coping with target disappearance and reap-pearance and (iv) VOT-RGBD2021 challenge focused on long-term tracking in RGB and depth imagery. The VOT-ST2021 dataset was refreshed, while VOT-RGBD2021 in-troduces a training dataset and sequestered dataset for win-ner identification. The source code for most of the trackers, the datasets, the evaluation kit and the results along with the source code for most trackers are publicly available at the challenge website. |
|
The COVID-19 pandemic has been deemed a global health pandemic. The early detection of COVID-19 is key to combating its outbreak and could help bring this pandemic to an end. One of the biggest challenges in combating COVID-19 is accurate testing for the disease. Utilizing the power of Convolutional Neural Networks (CNNs) to detect COVID-19 from chest X-ray images can help radiologists compare and validate their results with an automated system. In this paper, we propose a carefully designed network, dubbed CORONA-Net, that can accurately detect COVID-19 from chest X-ray images. CORONA-Net is divided into two phases: (1) The reinitialization phase and (2) the classification phase. In the reinitialization phase, the network consists of encoder and decoder networks. The objective of this phase is to train and initialize the encoder and decoder networks by a distribution that comes out of medical images. In the classification phase, the decoder network is removed from CORONA-Net, and the encoder network acts as a backbone network to fine-tune the classification phase based on the learned weights from the reinitialization phase. Extensive experiments were performed on a publicly available dataset, COVIDx, and the results show that CORONA-Net significantly outperforms the current state-of-the-art networks with an overall accuracy of 95.84%. |
|
Recently, the COVID-19 pandemic has affected the world and spread in the majority of the countries. To decrease the number of infections, experts suggested people practice social distancing by maintaining a distance of six feet apart. It is hard to monitor this restriction by only a traditional surveillance system. Existing methods used deep learning to tackle this problem by designing a Deep Convolutional Neural Network (DCNN). However, these methods do not accommodate for low-power systems such as Internet-of-Things-based devices. In this paper, we propose SocialNet, a novel network design that can detect violations of social distancing in public crowd scene. SocialNet is composed of two components, (1) The detector backbone and (2) The Autoencoder. In the detector backbone, the network generates the bounding boxes of the human/person category. In the Autoencoder, the network learns to predict a … |
|
Vehicle object detection is a fundamental task in computer vision. Most modern classifiers and trackers are built upon the object detectors. For example, self-driving cars use object detection on low-power devices to capture the information from the surrounding environment. Currently, object detection uses a huge amount of labelled data to train the detector. Moreover, these detectors are designed for high-end hardware (i.e., GPUs) and cannot be used on low-power devices. In this paper, we propose DETECTren, a novel object detector that uses self-supervised learning to leverage both the limited labelled data and the huge amount of unlabelled data. DETECTren learns to accurately detect the vehicle and its bounding box. DETECTren is divided into two tasks, (1) The pretext task and (2) The downstream task. In the pretext task, DETECTren uses an autoencoder with ResNet50 as a backbone sub-network to learn … |
|
Detecting the moving objects in a video is a challenging task. Most of the currently existing methods work on a pixel level. Working on a pixel level is usually inaccurate because the pixel intensity level may change due to illumination, dynamic background, camera movement, and many other challenges. This is why, in this article, a deep network is proposed, which learns to extract features that are invariant to these challenges and then detect moving objects based on the extracted features instead of pixel intensities. The proposed network MOD-YNet consists of two main parts: an encoder and a decoder. The encoder learns to extract suitable features from both the background frame and the current frame. Then, the decoder uses the extracted features from both frames and learns to extract the objects that exist in the current frame and does not exist in the background frame. MODY-Net is evaluated on a benchmark … |
|
Foreground segmentation of moving objects is widely used in different computer vision applications; however, existing deep learning-based methods generally suffer from overall degraded F-measure performance. The two main sources that degrade the F-measure are under-segmentation and catastrophic forgetting. Under-segmentation is the problem of misdetecting objects' fine details. The catastrophic forgetting problem occurs when training on a large number of video sequences that leads to forgetting information learned from early video sequences. This paper proposes a novel multi-scale region and edges fusion network with task-based parameter isolation (REFNet-TBPI) to overcome these two problems. The proposed method consists of a novel multi-scale region and edges fusion network (REFNet) to capture the moving objects' boundary details by extracting regions and boundary edges of each object at … |
|
Numerical modelings for industrial trials, which are used to predict expensive defects, are unrealistic. As a result, since generating annotated data is very expensive in the industry, a model is needed that can be adapted to real-world parameters with a handful of real experiments. Thus, this paper investigates applying the model-agnostic meta-learning algorithm, which is one of the state-of-the-art models in few-shot learning, to create a machine learning model for predicting bunching defects in manufacturing industrial hoses. This model is expected to extract the knowledge from simulated data from the bunching defects and adapt this knowledge for the real world using a few examples collected from real experiments. While the accuracy of the proposed algorithm is less than 15 percent, this paper shows a promising result if better-simulated data can be used. |
|
COVID-19 affects everyone on a daily-basis causing adjustments in which society functions. One of these major adjustments is the need to measure how well people distance from each other, that is referred to as social distancing. Previous work to automate social distancing violations does not take into consideration the exceptions to minimum distance guidelines. In this paper, we propose GroupNet, a novel multi-object tracking social distancing violation detector through the addition of group detection to reduce the number of false positives that are currently missed in existing literature. We define the social distancing violation occurs when two individuals are within a specified Euclidean distance of two meters. GroupNet leverages the contextual information learned by group detection. Moreover, GroupNet uses a Joint Detection and Embedding (JDE) multi-object tracker as a backbone network for group detection … |
|
Salient object detection (SOD) is the operation of detecting and segmenting a salient object in a natural scene. Several studies have examined various state-of-the-art machine learning approaches for SOD. In particular, deep convolutional neural networks (CNNs) are commonly applied for SOD because of their powerful feature extraction abilities. In this paper, we investigate the semantic segmentation capability of several well-known pre-trained models, including FCNs, VGGs, ResNets, MobileNet-v2, Xception and InceptionResNet-v2. These models have been trained over an ImageNet dataset, fine-tuned on a MSRA-10K dataset and evaluated using other public datasets, such as ECSSD, MSRA-B, DUTS and THUR15k. The results illustrate the superiority of ResNet50 and ResNet18, which have mean absolute errors (MAE) of approximately 0.93 and 0.92, respectively, compared to other well-known FCN models … |
2020
|
Recently, human computer interaction (HCI) technology is becoming more important and usable technology in different fields such as medicine and computer games. There are many new possible applications based on this technology. Fully body skeleton, head and facial features, and hand gestures based algorithms are well stabilised in literature. In this paper, real time applications for hand detection, tracking and recognition are surveyed and classified. The details of a system that responds to real time hand activity from the detection algorithm, which uses various key point textures to distinguish the hand and to track it is presented. Offline input or database feeds the system by a training input hand model to simulate the real time hand pose. The purpose of this survey is to obtain and demonstrate a suitable approach for the particular requirements of an application. |
|
In this letter, we analytically derive the probability distribution of the vehicle-to-drone packet delay on a bi-directional highway. The model on which the analysis is based considers the wireless communication range of the vehicles and the cluster length. In addition, the proposed analysis finds that the same calculation in related work underestimates the maximum inter-drone distance, stochastically limiting the vehicle-to-drone packet delay using the drone active service (DAS). Simulations are used to validate the proposed analysis. |
|
Human Visual System (HVS) has the ability to focus on specific parts of the scene, rather than the whole scene. This phenomenon is one of the most active research topics in the computer vision and neuroscience fields. Recently, deep learning models have been used for visual saliency prediction. In this paper, we investigate the performance of five state-of-the-art deep neural networks (VGG-16, ResNet-50, Xception, InceptionResNet-v2, and MobileNet-v2) for the task of visual saliency prediction. In this paper, we train five deep learning models over the SALICON dataset and then use the trained models to predict visual saliency maps using four standard datasets, namely: TORONTO, MIT300, MIT1003, and DUT-OMRON. The results indicate that the ResNet-50 model outperforms the other four and provides a visual saliency map that is very close to human performance. |
|
We propose NullSpaceNet, a novel network that maps from the pixel level input to a joint-nullspace (as opposed to the traditional feature space), where the newly learned joint-nullspace features have clearer interpretation and are more separable. NullSpaceNet ensures that all inputs from the same class are collapsed into one point in this new joint-nullspace, and the different classes are collapsed into different points with high separation margins. Moreover, a novel differentiable loss function is proposed that has a closed-form solution with no free-parameters. NullSpaceNet exhibits superior performance when tested against VGG16 with fully-connected layer over 4 different datasets, with accuracy gain of up to 4.55%, a reduction in learnable parameters from 135M to 19M, and reduction in inference time of 99% in favor of NullSpaceNet. This means that NullSpaceNet needs less than 1% of the time it takes a traditional CNN to classify a batch of images with better accuracy. |
|
A human Visual System (HVS) has the ability to pay visual attention, which is one of the many functions of the HVS. Despite the many advancements being made in visual saliency prediction, there continues to be room for improvement. Deep learning has recently been used to deal with this task. This study proposes a novel deep learning model based on a Fully Convolutional Network (FCN) architecture. The proposed model is trained in an end-to-end style and designed to predict visual saliency. The entire proposed model is fully training style from scratch to extract distinguishing features. The proposed model is evaluated using several benchmark datasets, such as MIT300, MIT1003, TORONTO, and DUT-OMRON. The quantitative and qualitative experiment analyses demonstrate that the proposed model achieves superior performance for predicting visual saliency. |
|
Moving object detection is a challenging task in computer vision. A class agnostic model is learned to detect moving objects in a video despite their category. This is done using the proposed MODSiam that takes a single background image of the scene and the current frame as input, then the model extracts features from both inputs and merges then to output the foreground objects. A comparison of using this model with three different backbone convolutional neural networks is presented. The evaluation is done using the metrics precision, recall, F1-measure, false-positive rate, false-negative rate, specificity, accuracy, and the number of frames per second. All models are tested on the benchmark dataset CDNet, which is a dataset of videos for moving objects under different conditions like low frame rate, shadows, and dynamic background. The results show that using ResNet as a backbone produced promising … |
|
Welcome to a Special Issue of the IEEE Canadian Journal of Electrical and Computer Engineering, which presents some of the top articles (in extended form) from the Canadian Conference of Electrical and Computer Engineering (CCECE 2019) held in Edmonton, AB, Canada, May 5–8, 2019. This conference was hosted by the Northern Canada Section, with four technical sponsors (IEEE Canada, IEEE Northern Canada Section, IEEE Southern Alberta Section, and the IAS/PES Northern Canada Chapter). |
|
Recently, our world witnessed major events that attracted a lot of attention towards the importance of automatic crowd scene analysis. For example, the COVID-19 breakout and public events require an automatic system to manage, count, secure, and track a crowd that shares the same area. However, analyzing crowd scenes is very challenging due to heavy occlusion, complex behaviors, and posture changes. This paper surveys deep learning-based methods for analyzing crowded scenes. The reviewed methods are categorized as (1) crowd counting and (2) crowd actions recognition. Moreover, crowd scene datasets are surveyed. In additional to the above surveys, this paper proposes an evaluation metric for crowd scene analysis methods. This metric estimates the difference between calculated crowed count and actual count in crowd scene videos. |
2019
|
The Visual Object Tracking challenge VOT2018 is the sixth annual tracker benchmarking activity organized by the VOT initiative. Results of over eighty trackers are presented; many are state-of-the-art trackers published at major computer vision conferences or in journals in the recent years. The evaluation included the standard VOT and other popular methodologies for short-term tracking analysis and a" real-time" experiment simulating a situation where a tracker processes images as if provided by a continuously running sensor. A long-term tracking subchallenge has been introduced to the set of standard VOT sub-challenges. The new subchallenge focuses on long-term tracking properties, namely coping with target disappearance and reappearance. A new dataset has been compiled and a performance evaluation methodology that focuses on long-term tracking capabilities has been adopted. The VOT toolkit has been updated to support both standard short-term and the new longterm tracking subchallenges. Performance of the tested trackers typically by far exceeds standard baselines. The source code for most of the trackers is publicly available from the VOT page. The dataset, the evaluation kit and the results are publicly available at the challenge website. |
|
The threat of small self-propelled semi-submersible vessels cannot be understated; payloads from drugs to weapons of mass destruction could be housed in these small, inconspicuous vessels. With a current apprehension rate of approximately 10%, a method resulting in increased interdiction of this illegal traffic is required for national security both in the ports along the coastlines of Canada, as well as the rest of North America. A smart, autonomous payload containing an infrared imaging device, designed for use in small unmanned aircraft systems for the specific mission of detecting self-propelled semi-submersibles over the vast ocean coastline will address the current security needs. Thermal imagery of the disturbed colder water layers, driven to the surface by the vessel will allow for the detection of this traffic using long wave infrared technology. Infrared signatures of ship wakes are highly variable in both persistence and temperature contrast as compared to the surrounding surface water, thus infrared imaging devices with a high resolution, a high responsivity, and a very low minimum resolvable temperature will be required to provide high quality imagery for airborne detection of the thermal wake. A theoretical understanding of the physics associated with the energy collected by the infrared sensor and the resulting infrared images is provided. Explanation of the factors affecting the resulting image with respect to the camera properties are detailed. A variety of examples of airborne thermal images are presented, with detailed explanations of the imaged scenes based on theory and sensor characteristics provided in the previous sections … |
|
Acutely ill patients presenting with conditions such as sepsis, trauma, and congestive heart failure require judicious resuscitation in order to achieve and maintain optimal circulating blood volume. Increasingly, emergency and critical care physicians are using portable ultrasound to approximate the temporal changes of the anterior–posterior (AP)-diameter of the inferior vena cava (IVC) in order to guide fluid administration or removal. This paper proposes semi-automatic active ellipse and rectangle algorithms capable of improved and quantified measurement of the AP-diameter. The proposed algorithms are compared to manual measurement and a previously published active circle model. Results demonstrate that the rectangle model outperforms both active circle and ellipse irrespective of IVC shape and closely approximates tedious expert assessment. |
|
Objective To evaluate the evidence about the safety and efficacy of tramadol in pain relief during diagnostic outpatient hysteroscopy (OH). Design Systematic review and meta-analysis of randomized controlled trials. Setting Not applicable. Patient(s) Patients undergoing diagnostic OH received tramadol versus those who were administered placebo. Intervention(s) Electronic databases were searched using the following MeSH terms (tramadol OR opioids OR opioid OR narcotic OR narcotics) AND (hysteroscopy OR hysteroscopic). Main Outcome Measure(s) Pain assessed by visual analogue scale (VAS) during OH, immediately and 30 minutes after the procedure. Result(s) One hundred thirteen studies were identified of which four randomized clinical trials were deemed eligible for this review (tramadol: n = 209; placebo: n = 209). The pooled estimate showed that tramadol significantly reduced VAS during procedure than … |
|
There has been much increased interest in the academic and industrial research communities on vehicular ad-hoc networks (VANETs). In this paper, we present an analytical model to study the end-to-end delay in a one-way VANET. This paper proposes an analytical formula for the end-to-end delay probability distribution. Using the derived probability distribution, the probability that the end-to-end delay is lower than a given threshold may be calculated. In addition, one can straightforwardly study the impact of parameters such as wireless communication range, vehicular densities, distance between source the destination, and minimum and maximum vehicle speeds on the end-to-end delay. This can help to better understand data dissemination in VANETs. Moreover, closed forms for lower and upper bounds on the end-to-end delay probability distribution are obtained. Extensive simulation results demonstrate the … |
|
Portable ultrasound is increasingly used to assess jugular venous pressure (JVP) to approximate volume status in patients with congestive heart failure (CHF). Traditionally, increases in jugular venous pressure height signify increasing circulating blood volume. Emerging evidence, suggests that JVP correlates well with sonographic images of the internal jugular vein (IJV). This paper represents a preliminary investigation on the ability of cross-sectional area (CSA) of the IJV to measure relative changes in circulating blood volume. Fourteen healthy subjects had serial transverse ultrasound videos of their IJV captured while lying at five angles designed to simulate relative changes in blood volume. Ultrasound videos of the IJV were both manually and semi-automatically segmented, the CSA was measured, outliers were detected and removed, and Rotation Forest classifier was used to classify the data. By limiting the … |
|
The detection of ground-moving objects in aerial videos has evolved over the years to handle more challenges such as large camera motion, the small size of the objects, and occlusion. Recently, aerial detection has been attempted using principal component pursuit (PCP) due to its superiority in detecting small moving objects. However, PCP-based detection methods generally suffer from high-false detections as well as high-computational loads. This paper presents a novel PCP-based detection method called kinematic regularization with local null space pursuit (KRLNSP) that drastically reduces false detections and the computational loads. KRLNSP models the background in an aerial video as a subspace that spans a low-dimension subspace while it models the moving objects as moving sparse. Accordingly, the detection is achieved by using multiple local null spaces and enhanced kinematic regularization … |
|
Controversy exists regarding the use of cement for hemiarthroplasty to treat displaced intracapsular hip fractures. The aim of this systematic review and meta-analysis was to compare the clinical outcomes between contemporary cemented and contemporary uncemented hemiarthroplasty for the treatment of displaced femoral neck fractures. |
|
Human eye movement is one of the most important functions for understanding our surroundings. When a human eye processes a scene, it quickly focuses on dominant parts of the scene, commonly known as a visual saliency detection or visual attention prediction. Recently, neural networks have been used to predict visual saliency. This paper proposes a deep learning encoder-decoder architecture, based on a transfer learning technique, to predict visual saliency. In the proposed model, visual features are extracted through convolutional layers from raw images to predict visual saliency. In addition, the proposed model uses the VGG-16 network for semantic segmentation, which uses a pixel classification layer to predict the categorical label for every pixel in an input image. The proposed model is applied to several datasets, including TORONTO, MIT300, MIT1003, and DUT-OMRON, to illustrate its efficiency. The results of the proposed model are quantitatively and qualitatively compared to classic and state-of-the-art deep learning models. Using the proposed deep learning model, a global accuracy of up to 96.22% is achieved for the prediction of visual saliency. |
|
The problem of forged images has become a global phenomenon that is spreading mainly through social media. New technologies have provided both the means and the support for this phenomenon, but they are also enabling a targeted response to overcome it. Deep convolution learning algorithms are one such solution. These have been shown to be highly effective in dealing with image forgery derived from generative adversarial networks (GANs). In this type of algorithm, the image is altered such that it appears identical to the original image and is nearly undetectable to the unaided human eye as a forgery. The present paper investigates copy-move forgery detection using a fusion processing model comprising a deep convolutional model and an adversarial model. Four datasets are used. Our results indicate a significantly high detection accuracy performance (~95%) exhibited by the deep learning CNN and discriminator forgery detectors. Consequently, an end-to-end trainable deep neural network approach to forgery detection appears to be the optimal strategy. The network is developed based on two-branch architecture and a fusion module. The two branches are used to localize and identify copy-move forgery regions through CNN and GAN. |
|
The recent digital revolution has sparked a growing interest in applying convolutional neural networks (CNNs) and deep learning to the field of image forensics. The proposed methods aim to train algorithms for solving a range of predetermined tasks. However, training a model that has been randomly initialized requires extensive time for computation as well as an enormous pool of training data to draw from. Moreover, such a model needs to be developed and redeveloped from the ground up if there are any alterations to the feature-space distribution. In addressing these problems, the present paper proposes a novel approach to training image forgery detection models. The method applies prior knowledge that has been transferred to the new model from previous steganalysis models. Additionally, because CNN models generally perform badly when transferred to other databases, transfer learning accomplished through knowledge transfer allows the model to be easily trained for other databases. The various models are then evaluated using image forgery techniques such as shearing, rotating, and scaling images. The experimental results, which show an image manipulation detection has validation accuracy of over 94.89%, indicate that the proposed transfer learning approach successfully accelerates CNN model convergence but does not improve image quality. |
|
The recent digital revolution has sparked a growing interest in applying convolutional neural networks (CNNs) and deep learning to the field of image forensics. The proposed methods aim to train algorithms for solving a range of predetermined tasks. However, training a model that has been randomly initialized requires extensive time for computation as well as an enormous pool of training data to draw from. Moreover, such a model needs to be developed and redeveloped from the ground up if there are any alterations to the feature-space distribution. In addressing these problems, the present paper proposes a novel approach to training image forgery detection models. The method applies prior knowledge that has been transferred to the new model from previous steganalysis models. Additionally, because CNN models generally perform badly when transferred to other databases, transfer learning accomplished through knowledge transfer allows the model to be easily trained for other databases. The various models are then evaluated using image forgery techniques such as shearing, rotating, and scaling images. The experimental results, which show an image manipulation detection has validation accuracy of over 94.89%, indicate that the proposed transfer learning approach successfully accelerates CNN model convergence but does not improve image quality. |
|
Balancing the trade-off between real-time performance and accuracy in object tracking is a major challenge. In this paper, a novel dynamic policy gradient Agent-Environment architecture with Siamese network (DP-Siam) is proposed to train the tracker to increase the accuracy and the expected average overlap while performing in real-time. DP-Siam is trained offline with reinforcement learning to produce a continuous action that predicts the optimal object location. DP-Siam has a novel architecture that consists of three networks: an Agent network to predict the optimal state (bounding box) of the object being tracked, an Environment network to get the Q-value during the offline training phase to minimize the error of the loss function, and a Siamese network to produce a heat-map. During online tracking, the Environment network acts as a verifier to the Agent network action. Extensive experiments are performed on … |
|
Purpose Both sliding hip screws (SHS) and cancellous screws are used in the surgical management of intracapsular femoral neck fracture. However, there is paucity of information as to which is the superior treatment modality. We performed this systematic review and meta-analysis study to compare the clinical outcomes of SHS and cancellous screws for the treatment of femoral neck fractures in adult patients. Methods We searched PubMed, Scopus, Web of Science, and Cochrane CENTRAL, up to December 2017. Randomized controlled trials (RCTs) directly comparing the clinical outcomes of SHS and cancellous screws for femoral neck fractures were retrieved with no language or publication year restrictions. Data retrieved included operative details, nonunion rate, avascular necrosis, reoperation, infection and mortality, hip pain, functional hip scores, and medical complications. These were pooled as risk … |
|
Visual object tracking is a fundamental task in the field of computer vision. Recently, Siamese trackers have achieved state-of-the-art performance on recent benchmarks. However, Siamese trackers do not fully utilize semantic and objectness information from pre-trained networks that have been trained on image classification task. Furthermore, the pre-trained Siamese architecture is sparsely activated by the category label, which leads to unnecessary calculations and overfitting. In this paper, we propose to learn a Domain-Aware that fully utilizes semantic and objectness information while producing a class-agnostic using a ridge regression network. Moreover, to reduce the sparsity problem, we solve the ridge regression problem with a differentiable weighted-dynamic loss function. Our tracker, dubbed DomainSiam, improves the feature learning in the training phase and generalization capability to other … |
|
Digital image forgery is a growing problem due to the increase in readily-available technology that makes the process relatively easy. In response, several approaches have been developed for detecting digital forgeries. This paper proposes a novel scheme based on neural networks and deep learning, focusing on the convolutional neural network (CNN) architecture approach to enhance a copy-move forgery detection. The proposed approach employs a CNN architecture that incorporates pre-processing layers to give satisfactory results. In addition, the possibility of using this model for various copy-move forgery techniques is explained. The experiments show that the overall validation accuracy is 90%, with a set iteration limit. |
|
Recently, accurate detection of moving objects has achieved via principal component pursuit (PCP). However, in the case of aerial imagery, existing PCP-based detection methods suffer from low accuracy and/or high computational loads. This paper presents a novel S-PCP method, called local null space pursuit (LNSP), which achieves a high detection accuracy and real-time performance on aerial images. LNSP models the background as a subspace that lies in a low-dimensional subspace, while the moving objects are modelled as sparse. Based on these two models, LNSP proposes a new formulation for the detection problem by using multiple local null spaces and -norm. The performance of LNSP is evaluated on challenging aerial datasets and then compared the results with relevant current state-of-the-art methods. |
2018
|
Fall detection is of interest to health care providers and researchers in recent and past decades since it can be used to reduce emergency response time. In fact this is can be used to reduce health care cost. Extensive research has been done to detect fall in all possible conditions. The research has generated many different algorithms and application to automatized fast alarm to reduce the consequences of the fall. This article gives an inclusive review of different research of fall detection systems, identify the existing approaches and principles methods used to detect the fall. Fall detection categories can be scattered into the following: wearable device based, vision based and ambience device based. These categories were analyzed and compared for each published work. At the end of this work we proposed some ways to improve the presented systems and some future work. |
|
Image forgery detection approaches are varied and serve same objectives. However, the difference in image properties causes some limitations of most of these approaches. Integrate multiple forensic approaches to increase the efficiency of detecting and localize the forgery was proposed based on the same image input source. In this paper, we propose a new detector algorithm based on different image source format. We propose a fusion approach to detect a copy-move forgery based on PatchMatch enhanced by the dense field technique, and sensor pattern noise based on photo response non-uniformity (PRNU). The F-measure score used same evaluation function to make the system more robust. The output result shows high efficiency of detecting and localizing the forgery in different image formats, for both passive and active forgery detection. |
|
Detection of relative changes in circulating blood volume is important to guide resuscitation and manage a variety of medical conditions, including sepsis, trauma, dialysis, and congestive heart failure. Recent studies have shown that estimates of circulating blood volume can be obtained from the cross-sectional area of the internal jugular vein (IJV) from ultrasound images. However, accurate segmentation and tracking of the IJV in ultrasound imaging is a challenging task and is significantly influenced by a number of parameters, such as the image quality, shape, and temporal variation. In this paper, we propose a novel adaptive polar active contour (Ad-PAC) algorithm for the segmentation and tracking of the IJV in ultrasound videos. In the proposed algorithm, the parameters of the Ad-PAC algorithm are adapted based on the results of segmentation in previous frames. The Ad-PAC algorithm is applied to 65 … |
|
Wireless communications between vehicles are a focus of research in both the academic research community and automobile industry. Using unmanned aerial vehicles or drones in wireless communications and vehicular ad hoc networks (VANETs) have started to attract attention. This paper proposes a routing protocol that uses the infrastructure drones for boosting VANET communications to achieve a minimum vehicle-to-drone packet delivery delay. This paper also proposes a closed-form expression for the probability distribution of the vehicle-to-drone packet delivery delay on a two-way highway. In addition, based on that closed-form expression, we can calculate the minimum drone density (maximum separation distance between two adjacent drones) that stochastically limits the worst case of the vehicle-to-drone packet delivery delay. Moreover, this paper proposes a drones-active service that is added to the … |
|
Detecting moving objects has been well studied in the past due to its importance in computer vision applications. Nevertheless, in aerial imagery, the small sizes of moving objects and the camera motion present challenges to existing well-known detection methods. Most moving object detection methods have reported either high true detection rates associated with high false-detection rates, or low false-detection rates at the expense of lowering true detection rates. This paper proposes a novel method, Kinematic Regularization and Matrix Rank Optimization (KRMARO), to achieve high true-detection rates and reduce false-detection rates significantly. KRMARO introduces a formulation of the moving objects detection problem that integrates a novel kinematic regularization into the principal component pursuit. This formulation models moving objects as sparse, which is located in regions exhibiting unique kinematic … |
|
Medical research suggests that the anterior-posterior (AP)-diameter of the inferior vena cava (IVC) and its associated temporal variation as imaged by bedside ultrasound is useful in guiding fluid resuscitation of the critically-ill patient. Unfortunately, indistinct edges and gaps in vessel walls are frequently present which impede accurate estimation of the IVC AP-diameter for both human operators and segmentation algorithms. The majority of research involving use of the IVC to guide fluid resuscitation involves manual measurement of the maximum and minimum AP-diameter as it varies over time. This effort proposes using a time-varying circle fitted inside the typically ellipsoid IVC as an efficient, consistent and novel approach to tracking and approximating the AP-diameter even in the context of poor image quality. In this active-circle algorithm, a novel evolution functional is proposed and shown to be a useful tool for … |
|
In this paper, facial recognition has been widely studied due to its importance in many applications in the civilian and military domains. Although this computer vision problem was initially challenging due to the dynamic nature of the human face and the different poses it can take, however, the research conducted over the last two decades made huge advances with many algorithms reporting high accuracy in the published literature. However, this accuracy is usually reduced in real-life usage especially in the presence of different types of noise. In this paper, six different facial recognition algorithms are evaluated and compared, namely, principle component analysis (PCA), two-dimensional PCA (2D-PCA), linear discriminant analysis (LDA), discrete cosine transform (DCT), support Vector Machines (SVM) and independent component analysis (ICA). The effect of the presence of Gaussian and salt and Pepper noises are also considered during the evaluation of these algorithms. The results show that the best performance was obtained using the DCT algorithm with 92% dominant eigenvalues and 95.25% accuracy which makes it the best choice under different noise conditions. |
|
This letter proposes a closed-form expression for the probability distribution of the re-healing delay (time taken in the store-and-forward strategy to send a packet from a cluster head to the tail of the next cluster) conditioned on the gap distance between those two clusters on a one-way highway. Moreover, a closed-form expression is derived for the unconditional probability distribution of the re-healing delay. Using the derived probability distribution, one can straightforwardly study the impact of VANET parameters on the re-healing delay. Also, the probability distribution of the end-to-end delay in VANETs can be derived from the results in this letter. The accuracy of the proposed analysis is validated using simulations. |
|
Visual tracking is a difficult and challenging problem, for numerous reasons such as small object size, pose angle variations, occlusion, and camera motion. Object tracking has many real-world applications such as surveillance systems, moving organs in medical imaging, and robotics. Traditional tracking methods lack a recovery mechanism that can be used in situations when the tracked objects drift away from ground truth. In this paper, we propose a novel framework for tracking moving objects based on a composite framework and a reporter mechanism. The composite framework tracks moving objects using different trackers and produces pairs of forward/backward tracklets. A robustness score is then calculated for each tracker using its forward/backward tracklet pair to find the most reliable moving object trajectory. The reporter serves as the recovery mechanism to correct the moving object trajectory when the … |
|
There are currently various classification algorithms, each with its own advantages and limitations. It is expected that fusing different classifiers in a way that the advantages of each are selected can boost the accuracy in the classification of complex land covers, such as wetlands, compared to using a single classifier. Classification of wetlands using remote-sensing methods is a challenging task because of considerable similarities between wetland classes. This fact is more important when utilizing synthetic aperture radar (SAR) data, which contain speckle noise. Consequently, discriminating wetland classes using only SAR data is generally not as accurate as using some other satellite data, such as optical imagery. In this study, a new Multiple Classifier System (MCS), which combines five different algorithms, was proposed to improve the classification accuracy of similar land covers. This system was then applied … |
|
Convolutional Siamese neural networks have been recently used to track objects using deep features. Siamese architecture can achieve real time speed, however it is still difficult to find a Siamese architecture that maintains the generalization capability, high accuracy and speed while decreasing the number of shared parameters especially when it is very deep. Furthermore, a conventional Siamese architecture usually processes one local neighborhood at a time, which makes the appearance model local and non-robust to appearance changes. To overcome these two problems, this paper proposes DensSiam, a novel convolutional Siamese architecture, which uses the concept of dense layers and connects each dense layer to all layers in a feed-forward fashion with a similarity-learning function. DensSiam also includes a Self-Attention mechanism to force the network to pay more attention to … |
2017
|
Image forgery detection approaches are varied and serve same objectives. However, the difference in image properties causes some limitations of most of these approaches. Integrate multiple forensic approaches to increase the efficiency of detecting and localize the forgery was proposed based on the same image input source. In this paper, we propose a new detector algorithm based on different image source format. We propose approach to detect a copy-move forgery based on PatchMatch enhanced by the dense field technique. The F-measure score used same evaluation function to make the system more robust. The output result shows high efficiency of detecting and localizing the forgery in different image formats, for passive forgery detection. |
|
Fast and robust image matching is a very important task with various applications in computer vision and robotics. In this paper, we compare the performance of three different image matching techniques, i.e., SIFT, SURF, and ORB, against different kinds of transformations and deformations such as scaling, rotation, noise, fish eye distortion, and shearing. For this purpose, we manually apply different types of transformations on original images and compute the matching evaluation parameters such as the number of key points in images, the matching rate, and the execution time required for each algorithm and we will show that which algorithm is the best more robust against each kind of distortion. Index Terms-Image matching, scale invariant feature transform (SIFT), speed up robust feature (SURF), robust independent elementary features (BRIEF), oriented FAST, rotated BRIEF (ORB). |
|
In this paper, we propose a new routing protocol called multi-copy intersection-based routing (MCIR) for vehicular ad-hoc networks (VANETs) in urban areas. MCIR is an intersection-based routing protocol that forwards multiple copies of the packets in different road segments. Moreover, it is a beacon-less routing protocol with a carry-and-forward strategy. We show via simulation that MCIR protocol is superior to other existing routing protocols, especially in low vehicular density scenarios. The results show that MCIR achieves a shorter endto-end delay and a higher packet delivery ratio in urban VANET communications. |
|
Structural health monitoring (SHM) using wireless sensor networks (WSNs) has gained research interest due to its ability to reduce the costs associated with the installation and maintenance of SHM systems. SHM systems have been used to monitor critical infrastructure such as bridges, high-rise buildings, and stadiums and has the potential to improve structure lifespan and improve public safety. The high data collection rate of WSNs for SHM pose unique network design challenges. This paper presents a comprehensive survey of SHM using WSNs outlining the algorithms used in damage detection and localization, outlining network design challenges, and future research directions. Solutions to network design problems such as scalability, time synchronization, sensor placement, and data processing are compared and discussed. This survey also provides an overview of testbeds and real-world deployments of … |
|
Evaluation of an image processing algorithm (IPA) task is tedious and full of complexity. A lot of effort is spent to compare a new IPA with benchmark IPAs. Comparing with benchmark IPAs requires either implementing them from scratch or a lot of configurations. Also, setting up datasets for the evaluation consumes additional effort. Therefore, the need for a system to overcome the former overhead is imperative. In this paper, a design of a novel automated evaluation system, AEIPA, is proposed. AEIPA allows automatic evaluation of different IPAs using different datasets regardless the programming language of the IPAs. Also, automatic reporting module is provided to compare different IPAs results. The proposed system applies openness principal to enrich AEIPA with IPAs (using plugin-based concepts in Docker containers), and datasets. |
|
Detection of relative changes in circulating blood volume is important to guide resuscitation and manage variety of medical conditions including sepsis, trauma, dialysis and congestive heart failure. Recent studies have shown that estimates of circulating blood volume can be obtained from ultrasound imagery of the of the internal jugular vein (IJV). However, segmentation and tracking of the IJV is significantly influenced by speckle noise and shadowing which introduce uncertainty in the boundaries of the vessel. In this paper, we investigate the use of optical flow algorithms for segmentation and tracking of the IJV and show that the classical Lucas-Kanade (LK) algorithm provides the best performance among well-known flow tracking algorithms. |
|
An ice detection system is designed using an imaging sensor which has a forward field of a view direction. The proposed system solves the problem of detecting the Black Ice on various surfaces using a depth imaging sensor based on Kinect device. In some places where ice cannot be seen easily from a distance, especially in low visibility, by a human eye, it would be highly dangerous and causes slip. The designed system can detect the forming ice at a distance from 82cm up to 1.52m from the camera. Ice detection system has been tested in five different backgrounds: wood, glass, ceramic, plastic and concrete, and indeed it showed high proficiency. |
|
Various denoising algorithms exist in the literature, however, no studies have ever been made to measure the impact of denoising algorithms on the quality of the video produced by a stabilization algorithm. In this paper, the impact of state of the art denoising algorithms on a feature-based video stabilization is measured and evaluated. Also, a quantitative measure is proposed which can give more insight on the impact of the chosen denoising algorithm on stabilization. The results show that the denoising algorithm can drastically affect the quality of stabilization results and choosing the latest denoising algorithm does not always guarantee the best stabilization results. |
|
Fast and robust image matching is a very important task with various applications in computer vision and robotics. In this paper, we compare the performance of three different image matching techniques, i.e., SIFT, SURF, and ORB, against different kinds of transformations and deformations such as scaling, rotation, noise, fish eye distortion, and shearing. For this purpose, we manually apply different types of transformations on original images and compute the matching evaluation parameters such as the number of key points in images, the matching rate, and the execution time required for each algorithm and we will show that which algorithm is the best more robust against each kind of distortion. Index Terms-Image matching, scale invariant feature transform (SIFT), speed up robust feature (SURF), robust independent elementary features (BRIEF), oriented FAST, rotated BRIEF (ORB). |
|
Image identification is one of the most challenging tasks in different areas of computer vision. Scale-invariant feature transform is an algorithm to detect and describe local features in images to further use them as an image matching criteria. In this paper, the performance of the SIFT matching algorithm against various image distortions such as rotation, scaling, fisheye and motion distortion are evaluated and false and true positive rates for a large number of image pairs are calculated and presented. We also evaluate the distribution of the matched keypoint orientation difference for each image deformation. |
|
Using Unmanned Aerial Vehicles (UAVs) or drones in Vehicular Ad-hoc Networks (VANETs) has started to attract attention. This paper proposes a mathematical framework to determine the minimum drone density (maximum separation distance between two adjacent drones) that stochastically limits the worst case for the vehicle-to-drone packet delivery delay. In addition, it proposes a drones-active service (DAS) that is added to the location service in a VANET to obtain the required number of active drones based on the current vehicular density while satisfying a probabilistic requirement for vehicle-to-drone packet delivery delay. Our goal is boosting VANET communications using infrastructure drones to achieve the minimum vehicle-to-drone packet delivery delay. We are interested in two-way highway VANET networks with low vehicular density. The simulation results show the accuracy of our mathematical … |
|
Image forgery detection approaches are varied and serve same objectives. However, the difference in image properties causes some limitations of most of these approaches. Integrate multiple forensic approaches to increase the efficiency of detecting and localize the forgery was proposed based on the same image input source. In this paper, we propose a new detector algorithm based on different image source format. We propose approach to detect a copy-move forgery based on PatchMatch enhanced by the dense field technique. The F-measure score used same evaluation function to make the system more robust. The output result shows high efficiency of detecting and localizing the forgery in different image formats, for passive forgery detection. |
|
Medical research suggests that the area of the IVC and its temporal variation imaged by bedside ultrasound is useful in guiding resuscitation of the critically-ill. Unfortunately, gaps in the vessel wall and intraliminal artifact represents a challenge for both manual and existing algorithm-based segmentation techniques. In this paper, a novel polar active contour algorithm based on the third image moment is proposed and used for segmentation and tracking of the IVC in ultrasound images. To validate the proposed research in this paper, we compare the proposed algorithm with manual segmentation and three state-of-the-art relevant algorithms. It is shown that the algorithm outperforms other techniques and in some scenarios appear to have advantages over manual segmentation creating the potential to improve medical management of the critically-ill patients. |
|
Moving objects detection from aerial camera platforms is a very challenging problem due to the small-size of the moving objects and the false motion of the static background elements. Although many methods have been proposed in this domain, they always have a trade-off between true detections and false detections. This paper proposes a novel solution called matrix rank optimization method (MARO) to achieve high true detections with low false detections. In MARO, the detection problem is formulated as a principal component pursuit with a transformation domain. The novelty of MARO is that it solves this problem by using the inexact Newton method and a backtracking behaviour in inexact augmented Lagrange multiplier. MARO has been extensively evaluated using DARPA VIVID, UCF aerial action, and VIRAT aerial datasets. The results show that MARO outperforms current-state-of-the-art methods … |
2016
|
A new method presented in this paper to describe the movement of a camera mounted on a mobile platform using local Maximally Stable Extremal Region detector (MSER) algorithm. In fact, it was established that MSER tracker method is quicker than other methods and it needs less resources. Furthermore, MSER outperforms other interest points detection and matching algorithms. Using Motion Estimation (ME) technique to detect the direction and speed of the camera during a video recording demonstrates the reliability of the algorithm. Results indicate over 93 percent tracking accuracy in variant environment. The paper presents the details of algorithm, implementation setup and results for a set of recorded videos. |
|
Point of Care Ultrasound is being used in many different disciplines of medicine due to its availability, low cost and lack of ionizing radiation. It has the ability to diagnose internal bleeding in trauma, rule out ectoptic pregnancy, assess for gallstones and provide needle guidance during invasive procedures. External processing of ultrasound imagery allows it to serve as a low-cost non-invasive monitoring solution. A range of physiologic parameters can be extracted. Heart rate and respiratory rate can be estimated from the power spectral density of the cross sectional area of the internal jugular vein (IJV). Circulating blood volume can be estimated through ridge regression of features extracted from videos of the IJV. This paper provides preliminary proof-of-concept for the extraction of these signals from ultrasound imagery of the internal jugular vein. |
|
Information regarding a patient’s health status can be extracted from ultrasound imagery of the arterial and venous vasculature. This paper investigates the use of Haralick features and edge features to distinguish euvolemia from hypovolemia. Transverse ultrasound videos of the internal jugular vein were collected from a set of healthy subjects using a simulation to generate different volume states. Features were extracted from each frame and assessed using common feature selection methods. These features provided a reasonable classification accuracy of 88.6% and worked best when considering texture on a small scale. |
|
The assessment of the blood volume is crucial for the management of many acute and chronic diseases. Recent studies have shown that circulating blood volume correlates with the cross-sectional area (CSA) of the internal jugular vein (IJV) estimated from ultrasound imagery. In this paper, a semi-automatic segmentation algorithm is proposed using a combination of region growing and active contour techniques to provide fast and accurate segmentation of IJV ultrasound videos. The algorithm is applied to track and segment the IJV across a range of image qualities, shapes and temporal variation. The experimental results show that the algorithm performs well compared to expert manual segmentation and outperforms several published algorithms incorporating speckle tracking. |
|
With the widespread use of handheld devices and unmanned aerial vehicles (UAVs) that has the ability to record video sequences. Digital video stabilization becomes more important as these sequences are usually shaky undermining the visual quality of the video. Digital video stabilization has been studied for decades yielding an extensive amount of literature in the field. However, most of them are highly sequential. In this paper, we present a new parallel technique that exploits the parallel architecture found in modern day devices. The algorithm divides the frame into blocks and estimates a camera path for each block to better enhance the estimation of the transformation needed to adjust for the shakiness of the video. |
|
Aerial imagery applications have gained a great interest especially in the area of comprehensive ground activities analysis. One of the key tasks in such applications is moving objects segmentation. Although many efforts have been presented in the literature that claim high true object detection rates, they still suffer from high false positive rates. This paper focuses on maintaining a high true positive detection rates while significantly reducing the false positive detection rates. To achieve this goal, this paper proposes a novel method that integrates matrix recovery concept with physical spring model to drastically reduce false detections. The proposed method segment all candidate moving objects by recovering the low rank matrix, which normally results high false positive detection. To reject false detections, each candidate moving object is modelled as a mass suspended by system of springs, such that the forces of … |
|
Digital video stabilization is crucial in many applications such as object detection and tracking. It has been studied for decades yielding an extensive amount of literature in the field, however, current approaches suffer from either being computationally expensive or under-performing in terms of visual quality . In this paper, we present StableFlow, a novel real-time method that was inspired by the mass-spring-damper physical model. In StableFlow, a video frame is modelled as a mass suspended in each direction by a critically dampened spring and damper which can be fine-tuned to adapt with different shaking patterns. The proposed method is tested on video sequences that have different types of shakiness and diverse video contents. The obtained results are then compared to current state-of-the-art stabilization algorithms including Youtube stabilization and it is found that the proposed method significantly … |
|
Portable ultrasound is commonly used to image blood vessels such as the Inferior Vena Cava or Internal Jugular Vein (IJV) in the attempt to estimate patient intravascular volume status. A large number of features can be extracted from a vessel’s cross section. This paper examines the role of shape factors and statistical moment descriptors to classify healthy subjects enrolled in a simulation modeling relative changes in volume status. Features were evaluated using a range of selection methods and tested with a variety of classifiers. It was determined that a subset of features derived from moments are the most appropriate for this task. |
|
This paper presents a comprehensive evaluation of the performance of common 3D keypoint detectors and descriptors currently available in the Point Cloud Library (PCL) to recover the transformation of 300 real objects. Current research on keypoints detectors and descriptors considers their performance individually in terms of their repeatability or descriptiveness, rather than on their overall performance at multi-sensor alignment or recovery. We present the data on the performance of each pair under all transformations independently: translations and rotations in and around each of the x-, y- and z-axis respectively. We provide insight into the implementation of the detectors and descriptors in PCL leading to abnormal or unexpected performance. The obtained results show that the ISS/SHOT and ISS/SHOTColor detector/descriptor pair works best at 3D recovery under various transformations. |
|
Traditional methods of capturing vital signs by monitoring electrical impulses are quite effective however this data has the potential to be extracted from alternative technology. Non-invasive monitoring using low-cost ultrasound imaging of arterial and venous vasculature has the potential to detect standard vital signs such as heart and respiratory rate as well as additional parameters such as relative changes in circulating blood volume. This paper explores the feasibility of using ultrasound to monitor these signals by detecting spatial and temporal changes in the internal jugular vein (IJV). Ultrasound videos of the jugular in the transverse plane were collected from a subset of healthy subjects. Frame-by-frame segmentation of the IJV demonstrates frequency characteristics similar to certain physiological systems. Heart and respiratory rate appear to be present in IJV cross-sectional area variations in select … |
2015
|
This paper presents a new method to describe the movement of a camera mounted on a mobile platform using local Maximally Stable Extremal Region detector (MSER) algorithm. In this work, it was established that MSER tracker is quicker than other methods and it needs less resources. Furthermore, MSER outperforms other interest points detection and matching algorithms. Using a Motion Estimation (ME) technique to detect the direction and speed of the camera movement demonstrates the reliability of the algorithm. Results indicate over 93% tracking accuracy in different environments. The paper presents the details of the algorithm, implementation setup and results for a set of recorded videos. |
|
Estimation of circulating blood volume is important for the management of acute and chronic diseases. Both excessive and insufficient fluid administration have been shown to result increased morbidity and mortality. Studies have shown that the circulating blood volume is correlated to the cross-sectional area (CSA) of the internal jugular vein (IJV) which can be estimated from its ultrasound images. In this paper, an efficient active contour-based algorithm is proposed for segmentation of the IJV. In the proposed algorithm, region growing is used to obtain an initial contour for each frame and then an active contour is used for its fine adjustments. The proposed algorithm was applied for segmentation and tracking of the cross-sectional area of the IJV in ultrasound videos with different image quality and different IJV shapes and quality and results of the validation experiments showed that the algorithm performs very accurately compared to expert manual segmentation, as considered as ground truth, and efficiently tracks the changes in the cross-sectional area of the IJV. The results from the proposed algorithm are also compared to current state of the art algorithms. |
|
Abstract Background: Hepatitis C virus (HCV) infection is an important public health problem in Egypt. Several reports have documented the association of chronic HCV infection with many extra-hepatic manifestations. Interstitial lung involvement has also been integrated into the list of these manifestations [1-21]. Aim of this study is to evaluate the interstitial pulmonary fibrotic changes via high resolution computed tomography (HRCT) of the chest and pulmonary function tests (PFTs) among Egyptian patients with chronic HCV infection. Patients: After departmental ethics committee approval and patient consent were obtained, 20 patients with chronic HCV infection without any previous pulmonary diseases were enrolled in this study. Methods: All patients were subjected to: History taking and clinical examination, diagnosis of hepatitis-C-virus (HCV) infection by third generation ELISA test for detection of HCV antibodies, PCR for HCV RNA, liver function tests, renal function tests, Complete blood count, Arterial Blood Gases (ABGs), serum cryoglobulins, Anti-Nuclear-Antibody (ANA), chest x-ray, abdominal ultrasonography, Pulmonary Function Testes (PFTs), and High Resolution Computed Tomography (HRCT) of the chest. Results: The patients were classified into 2 groups according to High Resolution Computed Tomography (HRCT) results: Group I: included 11 patients with chronic liver disease due to HCV who were positive for interstitial pulmonary fibrotic changes (IPF) in HRCT. Group II: included 9 patients with chronic liver disease due to HCV who were negative for interstitial pulmonary fibrotic changes (IPF) in HRCT. Interstitial pulmonary … |
|
Moving object detection is the first key step for many automated vision analysis applications. One of the major challenges to achieve accurate moving object detection is detecting moving objects in videos captured by moving camera platforms, also called active cameras, where both interest objects and background elements are moving. This paper presents a novel algorithm for moving objects detection from active cameras. The proposed method decomposes a video from an active camera into three components: background, moving objects, and transformation matrix between consecutive frames. The proposed method formulates the problem as a robust principle component analysis (PCA) problem (low rank matrix optimization problem) and solves it using inexact augmented Lagrange multiplier (IALM). In the proposed method, the background represents the low rank matrix, and the moving objects and … |
|
Accurate resuscitation of the critically-ill patient using intravenous fluids and blood products is a challenging, time sensitive task. Ultrasound of the inferior vena cava (IVC) is a non-invasive technique currently used to guide fluid administration, though multiple factors such as variable image quality, time, and operator skill challenge mainstream acceptance. This study represents a first attempt to develop and validate an algorithm capable of automatically tracking and measuring the IVC, compared to human operators across a diverse range of image quality. Minimal tracking failures and high levels of agreement between manual and algorithm measurements were demonstrated on good quality videos. Addressing problems such as gaps in the vessel wall and intra-lumen speckle should result in improved performance in average and poor quality videos. Semi-automated measurement of the IVC for the purposes of … |
|
Development and evaluation of Point of Care Ultrasound (PoCUS) skill is a resource intensive undertaking. Current practice involves expert supervision of trainees performing numerous practice scans in the clinical setting. Computer augmented training represents the potential for significant reduction in resources utilization. Multiple ultrasound training simulators exist however of unclear value in teaching image generation skills. This paper describes the concept of using a standard ultrasound machine and human subjects, combined with image processing and depth sensing technologies, to develop a realistic PoCUS training tool. In addition to the concept, we describe the initial data collection experiment and preliminary work integrating ultrasound imagery and probe movement. Operator assessment metrics explored in this paper include image quality, stability, and acquisition time, demonstrate good potential to … |
|
Aerial imagery is widely used in many civilian and military applications, as it provides a comprehensive view and real-time surveillance. Automated analysis is an essential task of aerial imagery to detect moving objects, however, the shakiness of these images and the small size of the moving objects are major challenges facing such task. This paper proposes UT-MARO, a novel moving object detection technique. UT-MARO achieves high accurate detection of small-size moving objects in shaky aerial images with low computation complexity and is composed of two phases: (1) UT-alignment and (2) MARO-extraction. UT-alignment utilizes unscented transformation to first align shaky images, then in the second phase, MARO-extraction detects small moving objects by extracting the background using low rank matrix optimization. The robustness of the proposed technique is tested on DARPA and UCF aerial … |
2014
|
Recent statistics show that the population of elderly people who live alone is increasing at a fast rate. Automatic monitoring systems for elderly people can help automatically detect falls and speed emergency response authorities and help reduce the number of injuries, and subsequent fatalities that occur as a result of falls. Most current vision-based approaches extract geometric information on the moving human from a live video stream, such as orientation, centroid, velocity, and other dynamic object features. These techniques suffer from a high false detection rate in specific postures. This paper presents an algorithm for fall detection that has high accuracy of fall detection in all postures. |
2013
|
In this paper, we proposed a discrete cosine transform (DCT)-based attnuation and accentuation method to remove lighting effects on face images for faciliating face recognition task under varying lighting conditions. In the proposed method, logorithm transform is first used to convert a face image into logarithm domain. Then discrete cosine transform is applied to obtain DCT coefficients. The low-frequency DCT coefficients are attenuated since illumination variations mainly concentrate on the low-frequency band. The high-frequency coefficients are accentuated since when under poor illuminations, the high-frequency features become more important in recognition. The reconstructed log image by inverse DCT of the modified coefficients is used for the final recognition. Experiments are conducted on the Yale B database, the combination of Yale B and Extended Yale B databases and the CMU-PIE database. The proposed method does not require modeling and model fitting steps. It can be directly applied to single face image, without any prior information of 3D shape or light sources. |
|
The threat of small self-propelled semisubmersible vessels cannot be understated; payloads from drugs to weapons of mass destruction could be housed in these small, inconspicuous vessels. With a current apprehension rate of approximately 10%[1], a method resulting in an increased interdiction of this illegal traffic is required for national security both in the ports along the coastlines of Canada, as well as the rest of North America. An unmanned aircraft with an integrated payload allowing the autonomous detection of illegal traffic will not only increase the number of detained vessels, but also reduce the resources required to find these vessels. Thermal imagery of the disturbed colder water layers, driven to the surface by the vessel will allow for the detection of this traffic using long wave infrared technology. Infrared signatures of ship wakes are highly variable in both persistence and temperature contrast as compared to the surrounding surface water, thus infrared imaging devices with a high resolution, a high responsivity, and a very low minimum resolvable temperature will be required to provide high quality imagery for airborne detection of the thermal wake. Infrared images taken over the Pacific ocean from manned aircraft platforms are presented. Temperature measurements taken using Vemco Minilog II temperature loggers confirmed the thermal stratification of the upper 5 meters of the water. Thermal scarring due to upwelled colder water to the surface was noted, with temperature difference found to be consistent with temperature profile. A proposed system architecture is presented towards the development of a real-time, autonomous … |
2012
|
Varying illumination conditions affect the appearance of face images significantly. Thus, it severely degrades image-based face recognition performance. This paper presents a novel face image pre-processing approach that deals with the illumination problem to make face recognition robust to illumination variations. In the proposed method, logarithm transform is first used to convert a face image into logarithm domain. Then discrete cosine transform (DCT) coefficients of it are modified to remove illumination variations. The reconstructed log image by inverse DCT of the modified coefficients is used for the final recognition. We achieved 100% face recognition rate on Yale face database B. The proposed method requires no assumption on the light source and any prior information on 3-D face geometry. |
|
Automatic license plate recognition (ALPR) is the extraction of vehicle license plate information from an image or a sequence of images. The extracted information can be used with or without a database in many applications, such as electronic payment systems (toll payment, parking fee payment), and freeway and arterial monitoring systems for traffic surveillance. The ALPR uses either a color, black and white, or infrared camera to take images. The quality of the acquired images is a major factor in the success of the ALPR. ALPR as a real-life application has to quickly and successfully process license plates under different environmental conditions, such as indoors, outdoors, day or night time. It should also be generalized to process license plates from different nations, provinces, or states. These plates usually contain different colors, are written in different languages, and use different fonts; some plates may have … |
2011
|
Typical video surveillance systems are very demanding in terms of infrastructure required to deploy them and also in terms of human resources required to operate them on a continuous basis. In this paper, we propose a system architecture that aims at addressing both issues. The system is composed of multiple intelligent nodes that acquire, process, and archive data/video at a remote site and then automatically generate either alerts or summary reports that are sent to a station at the central operations office of the customer. The intelligent nodes are capable of analyzing multiple types of input data, including video, and take actions ranging from communicating alerts back to the human operator to automatic shutdown of a complete facility. These intelligent nodes serve as the middleware devices in this distributed architecture. We present a case study in the pipelining industry in Canada. |
|
Human's safety in construction areas is vital. A hard hat is required to enter a construction area. Stopping a person who is not wearing a hard hat entering a construction area is very important. Video-based surveillance to detect hard hat is a new solution to this safety problem. This paper brings different video processing techniques together to construct a framework for fast and robust hard hat detection in construction areas. The proposed system can detect face and hard hat in real time. |
2010
|
The new CSA Standard, Security Management for Petroleum & Natural Gas Industry Systems, is changing the operational landscape throughout the oil and gas industry. This document focuses on an innovation that will help pipeline operators meet the new recommendations for monitoring and managing their remote assets as outlined in the new CSA standard. This paper includes an analysis of the current monitoring architecture that can be used for compliance with the new regulation as well as a detailed comparison of different architectures. New video surveillance architecture developments are also reviewed. The IntelliView technology uses software that turns passive cameras into video sensors capable of reporting video-based behavior exceptions based on user-defined rules. A hardware device known as a SmrtDVR sits on site and records the video in the highest quality (H.264) to ensure the images are … |
2009
|
Video-based Automatic Incident Detection (AID) systems are widely deployed in many cities for detecting traffic incidents to provide smoother, safer and congestion free traffic flow. However, the accuracy of an AID system operating in an outdoor environment suffers from high false alarm rates due to environmental factors. These factors include snow movement, static shadow and static glare on the roads. In this paper, a robust real-time algorithm is proposed to detect snow movement in video streams to improve the rate of detection. This is done by having the AID system reducing its sensitivity in the areas that have snow movements. The feasibility of the proposed algorithm has been evaluated using traffic videos captured from several cameras at the City of Calgary. |
2008
|
Video-based automatic incident detection (AID) systems are increasingly being used in intelligent transportation systems (ITS). Video-based AID is a promising method of incident detection. However, the accuracy of video-based AID is heavily affected by environmental factors such as shadows, snow, rain, and glare. This paper presents a review of the different work done in the literature to detect outdoor environmental factors, namely, static shadows, snow, rain, and glare. Once these environmental conditions are detected, they can be compensated for, and hence, the accuracy of alarms detected by video-based AID systems will be enhanced. Based on the presented review, this paper will highlight potential research directions to address gaps that currently exist in detecting outdoor environmental conditions. This will lead to an overall enhancement in the reliability of video-based AID systems and, hence, pave the … |
2007
|
This paper presents an interaction taxonomy for classifying and identifying requirement interactions in software systems. The proposed taxonomy is in the form of a four-layered pyramid that defines 6 Main Interaction Categories in the first layer, 17 Interaction Subcategories in the second layer, 29 Interaction Types in the third layer, and 29 Interaction Scenarios in the fourth layer. Each interaction scenario has a corresponding interaction detection guideline that describes how the interaction can be detected. The proposed interaction taxonomy was compared to other existing taxonomies in the literature and was not only able to address all the issues in those taxonomies, but also contained many other interaction types. The proposed interaction taxonomy serves as the first domain-independent requirement interactions taxonomy. It provides a detailed description of when two requirements interact. |
|
The water quality of Lake Manzala was studied seasonallyJ during the period from autumn 2000 to summer 2001. The results of physical parameters revealed that the values of transparency at the southern region are relatively low and reflect the type of effluents, characterized by high amounts of floating materials, which decrease the water transparency. Moreover, the electrical conductivity at these stations were somewhat high as a result of sewage and industrial wastes at that region. On the other hand, the chemical analysis of water showed high values of chemical oxygen demand (COD) and biological oxygen demand (BOD) but low levels of dissolved oxygen (DO) especially at the southern region opposite to Hadous and Bahr El-Baqar drains, in addition to high levels of trace metals. |
|
Feature Interaction is a problem mostly considered in the telecommunications domain. Many solutions for detecting interactions between telephony features have been reported. In this paper, we investigate the feature interaction problem beyond the traditional telecommunications domain and look at interactions between policies in other domains. We propose the use of semi-formal methods for detecting interactions between policies in the smart homes domain. The novelty of this research is threefold: firstly, a six step semi-formal approach, called IRIS (Identifying Requirements Interactions using Semi-formal methods), for detecting interactions is presented. A major component within IRIS, which is an interaction taxonomy, is also presented. Secondly, we extend the scope of the problem of feature interactions beyond telecommunication features and investigate interactions between policies in the smart homes … |
|
Smart homes have enjoyed increasing popularity in recent years. In order for them to further expand their market share, users need to be able to fully control devices. Policies are one way for users to achieve such flexible control of devices. However, user policies often tend to interact in unwanted ways leading to unexpected behavior of devices. This paper describes the design of a run-time policy interaction management module (PIMM) that serves as manager for detecting and resolving interactions between user policies in KNX-based smart homes. This module extends the traditional KNX networking system with the ability to manage policy interactions. The module operates in the run-time S-mode of the KNX network and works as part of the engineering tool software (ETS) used to configure and control the operation of the KNX network in smart homes. The proposed module serves as the first of its kind that can … |
|
Many ITS systems suffer from high false alarm rates due to static glare. Static glare appears in video scenes when the road is wet and light originating from static light source, such as road lamp, is reflected on the road. Often an Automatic Incidents Detection (AID) system will detect occurrences of static glare as incidents, causing the generation of false alarms. In this paper, an effective real-time algorithm is proposed to detect static glare. This algorithm generates a static glare map that can be used by an AID system to avoid static glare false alarms. |
|
In the process of selecting Commercial Off-The-Shelf (COTS) products, goals may be used to define the selection criteria. Goals can range from high-level strategic objectives to low-level technical objectives. During the process of defining these goals, high-level objectives are decomposed into more refined objectives at lower levels. However, it is important not only to define the decomposition links, but also to define how different technical goals interact. Defining such interaction is important to avoid unexpected system behaviour when a negative interaction scenario occurs between two goals. Moreover, defining these interactions at an early stage will prevent high repair costs at later stages. In this paper, we describe the application of an approach called IRIS to identify interactions among goals during COTS selection. As a proof of concept, we present an example of an e commerce system to illustrate the proposed work. |
2006
|
This paper presents an adaptive empirical algorithm which identifies static shadows within video sequences and produces static shadow maps that are used to improve the performance of video based automatic incident detection (AID) systems. The algorithm distinguishes between static shadows and other objects using background generation, motion detection, and three static shadow filters. The proposed algorithm has been tested on streams from 9 cameras to demonstrate its detection accuracy and robustness in varying lighting conditions |
|
One of the key components of an intelligent transportation systems is video-based automatic incident detection (AID). An AID system is able to detect incidents that require operator intervention. However, the accuracy of an AID system operating during the winter suffers from high false alarm rates due to the movement of snow on the roads. In this paper, a robust algorithm is proposed to detect moving snow in video streams and improve the rate of detection by having the AID system to reduce its sensitivity in the area that has snow movement. The proposed algorithm conducts glare processing, background generation & differencing, snow sample correlation and final snow map generation. The feasibility of the proposed algorithm has been evaluated using traffic videos captured from several cameras from the City of Calgary. This algorithm demonstrates accurate and real-time detection of moving snow |
|
This paper investigates the problem of requirement interactions which occurs due to negative relationships between requirements when developing software systems. This paper presents IRIS-TS (Requirements Interactions using Semi-formal methods-Tool Support) which identifies and detects requirement interactions using semi-formal methods in any software domain. IRIS-TS is implemented as an independent add-on module that can be added to DOORS (which is one of the most famous and commonly used requirements management tools). This paper presents also a case study in which the proposed IRIS-TS approach was successfully used as an add-on module in DOORS to detect interactions between smart homes requirements which represent a new application domain for interaction detection. The presented case study is the first comprehensive effort to fully detect interactions in the smart homes domain. |
2005
|
Finding ways of detecting interactions between requirements is essential in order to develop a set of clear requirements, which serves as a foundation for successful software development. Detecting requirements interactions as early as possible helps avoid high repair costs. This thesis presents IRIS, Identifying Requirements Interactions usingSemi-formal methods, which is a semi-formal approach for detecting requirements interactions. IRIS is a systematic six step approach that uses tables, graphs, interaction detection scenarios, and subjective judgment to detect requirements interactions in software systems. IRIS has the advantage of not only being domain independent but also customizable towards a specific domain in order to enhance its performance. IRIS helps reduce the number of necessary pair-wise comparisons between requirements that have to be performed informally by a human expert. This reduction is achieved by discarding irrelevant comparisons that will not lead to interactions. A general requirements interaction taxonomy was developed for identifying when two requirements are considered interacting. This requirements interaction taxonomy provides interaction detection scenarios that are used within IRIS for detecting interactions. |
2004
|
Requirements interactions are a big challenge in any software development methodology. This paper describes a three-level framework that can be customized for any domain and used to detect requirements interactions at different levels of cost and complexity. A definition of the different possible scenarios in which interactions between requirements will occur is presented with the goal of creating a general requirements interaction taxonomy. The paper then focuses on the semi-formal appraoch contained in level 2 of the framework and applies it to eight telecommunication features presented in the second feature interaction contest held in 2000. Finally, the results obtained are compared with those reported in literature. |
|
One of the most challenging tasks in requirements engineering is the establishment of guidelines of when and how to use certain requirements engineering processes within the project context. Requirements interaction detection, as a requirements engineering activity, still lacks the definition of precise guidelines for when and how to apply the different detection techniques. The novelties described in this paper are twofold: first, this paper presents an overall framework for requirements interaction detection along with guidelines on what techniques to use for a given project. Second, this paper explicitly links project characteristics with the practices and techniques of requirements interaction detection. The aim of this paper is to show how requirements interaction detection needs to be an integral part of RE process development. |
|
Requirements engineering is considered a critical phase of the software development life cycle. However, because of the complexity of today's projects, requirements often have a negative impact on each other. Requirements interaction detection is an important activity for the discovery of such unwanted interactions. Commonly used detection processes are oriented towards the telecommunication domain and are done using either human experts or formal approaches. This paper presents IRIS, which stands for identifying requirements interactions using semiformal methods. The novelty of IRIS is threefold: first, IRIS uses semiformal methods for the detection of interactions between requirements. This helps to fill in the gap between the commonly used informal and formal approaches. Secondly, IRIS is a domain independent approach, which means that it is not limited to the telecommunications domain but can be … |
2003
|
Increased pressure to reduce time-to-market has resulted in an increased effort to reuse previously developed software components when developing new software systems. However, in numerous cases, reuse has resulted in interactions between features and/or requirements. As the amount of reuse increases, interaction detection becomes a more and more challenging subject. Although feature interaction is not a new problem and has already been researched especially in the telecommunications domain where new features are added to large-scale base systems, relatively little research has been done outside of telecommunications. The aim of this paper is to introduce a systematic approach for the detection of interactions based on requirements attributes. This systematic approach is part of a three-level framework that offers several approaches to feature-interaction detection in any domain at different levels … |
|
This paper deals with the problem of requirements interaction. We introduce a three level framework to detect requirements interactions at different levels of cost, time, and complexity. Level 2 where we use semiformal methods to detect interactions contains the main contribution of the research. Also we combine existing approaches (e.g. informal and formal) with our semiformal approach to provide a comprehensive framework for developers to use. The approach is illustrated using two case studies, one from the telecommunications domain and the other one being a lift control system. The results obtained are very encouraging with regards to the time and effort spent on requirements interaction detection. |
2002
|
ion and a low reuse potential; while another requirement can be very general, with a of abstraction and thus a higher reuse potential. We identify the following abstraction specific requirements, generic requirements, and domain-requirements frameworks. Domain-specific requirements are requirements that are derived from a certain dom applications and are concerned only with this domain. Therefore, they they cannot be reused in any other domains and applications (e.g. system-specific requirem General requirements are requirements that are with some variations common to different ap one can replace the differentiation part in the requirement with a variable making requirement (e.g., The system shall support saving of email addresses up to X entries.) A domain requirements framework is a framework that provides guidance on how to ge requirements specification document with hookups to facilitate development The first part of our process for requirements reuse starts with the analysis of a group of sy systems that are part of a product line) in a certain domain that have numerous commo them, and thus it is p actual building of a ne of the common platform to build the new system, along with introducing the new require specific system needs. The third part to this process is interaction management of the requirements. This general process framework can be seen in figure (1) and is described the following subsections. 3.1 Phase 1: Elicitation The aim of phase 1 is to create a database of reusable requirements that can be used later a new system. Phase 1 starts with domain analysis during which a team of developers already develo that can be used t analysis is a set of reusable requirements (kernel requirements and/or parameterized req optional requirements) along with a set of rules of usage that the developer has to follo these requirements (e.g. requirements A and B are mutually exclusive). The next step i group of similar requirements into one component (e.g. all reusable requirements re |
Other
|
Video-based Automatic Incident Detection (AID) systems are widely used for the detection of traffic incidents. Unfortunately, the accuracy of AID is influenced by environmental factors such as glare, shadows, snow, and rain. This paper presents a method of glare detection that improves the performance and reliability of video-based AID systems. |
|
Accurate resuscitation of the critically-ill patient using intravenous fluids and blood products is a challenging, time sensitive task. Ultrasound of the inferior vena cava (IVC) is a non-invasive technique currently used to guide fluid administration, though multiple factors such as variable image quality, time, and operator skill challenge mainstream acceptance. This study represents a first attempt to develop and validate an algorithm capable of automatically tracking and measuring the IVC compared to human operators across a diverse range of image quality. Minimal tracking failures and high levels of agreement between manual and algorithm measurements were demonstrated on good quality videos. Addressing problems such as gaps in the vessel wall and intra-lumen speckle should result in improved performance in average and poor quality videos. Semi-automated measurement of the IVC for the purposes of non … |
|
A system for detecting a class of objects at a location, for example humans on a conveyor belt. A thermal camera may be used to detect objects and to detect the variance of the heat distribution of objects to classify them. Objects detected in an image from one camera may be detected in an image from another camera using geometric correction. A color camera may be used to detect the number of edges and the number of colors of an object to classify it. A color camera may be used with an upright human body classifier to detect humans in an area, and blobs corresponding to the detected humans may be tracked in a thermal or color camera image to detect if a human enters an adjacent forbidden area such as a conveyor belt. |