Search

Graphic Processing Units (GPUs) are becoming very important in the present day. Their high computational capabilities with high speed and accuracy are making them a very strong force in communication engineering. In recent times, their need has increased tremendously due to the increasing range of applications. Video surveillance is an important field where very heavy computations are needed to be done on videos to perfectly detect the motion of an object in suspicious situations. The various analyses on video can be used to extract information and process data to generate actionable intelligent conclusions. However, CPUs fail to deliver real time results when it comes to high-resolution videos from a large number of cameras simultaneously. Thankfully, there is a lot of graphic hardware available nowadays, which comprises powerful hardware processors often intended to process data in parallel and so greatly accelerates the processes being done on them. An accelerated algorithm is required for processing petabytes of data from security cameras and video surveillance satellites and that in real time. In this paper, we propose a method of using GPUs in detecting the motion of an object at different junctions in video surveillance. The results show a great gain in performance when the proposed method runs on GPUs and CPUs in terms of speed as well as accuracy. The new parallel processing approaches are developed on each phase of the algorithm to enhance the efficiency of the system. Proposed algorithm achieved an average speed up of 50.094x for lower resolution video frames (320 × 240,720 × 480,1024 × 768) and 38.012x for higher resolution video frames (1360 × 768,1920 × 1080) on GPU, which is superior to CPU processing. [ABSTRACT FROM AUTHOR]

IntroductionStudying the behavioral responses and movement trajectories of insects under different stimuli is crucial for developing more effective biological control measures. Therefore, accurately obtaining the movement trajectories and behavioral parameters of insects in three-dimensional space is essential.MethodsThis study used the litchi pest Thalassodes immissaria as the research object. A special binocular vision observation system was designed for nighttime movement. A thermal infrared camera was used for video recording of T. immissaria in a lightless environment. Moreover, a multi-object tracking method based on the YOLOX-GMM and SORT-Pest algorithms was proposed for tracking T. immissaria in thermal infrared images. By obtaining the central coordinates of the two T. immissaria in the video, target matching and 3D trajectory reconstruction in the parallel binocular system were achieved.ResultsError analysis of the T. immissaria detection and tracking model, as well as the 3D reconstruction model, showed that the average accuracy of T. immissaria detection reached 89.6%, tracking accuracy was 96.9%, and the average error of the reconstructed 3D spatial coordinates was 15 mm.DiscussionThese results indicate that the method can accurately obtain the 3D trajectory and motion parameters of T. immissaria. Such data can greatly contribute to researchers' comprehensive understanding of insect behavioral patterns and habits, providing important support for more targeted control strategies.

This paper introduces a novel background subtraction method that utilizes texture-level analysis based on the Gabor filter bank and statistical moments. The method addresses the challenge of accurately detecting moving objects that exhibit similar color intensity variability or texture to the surrounding environment, which conventional methods struggle to handle effectively. The proposed method accurately distinguishes between foreground and background objects by capturing different frequency components using the Gabor filter bank and quantifying the texture level through statistical moments. Extensive experimental evaluations use datasets featuring varying lighting conditions, uniform and non-uniform textures, shadows, and dynamic backgrounds. The performance of the proposed method is compared against other existing methods using metrics such as sensitivity, specificity, and false positive rate. The experimental results demonstrate that the proposed method outperforms other methods in accuracy and robustness. It effectively handles scenarios with complex backgrounds, lighting changes, and objects that exhibit similar texture or color intensity as the background. Our method retains object structure while minimizing false detections and noise. This paper provides valuable insights into computer vision and object detection, offering a promising solution for accurate foreground detection in various applications such as video surveillance and motion tracking. [ABSTRACT FROM AUTHOR]

This article presents a background subtraction method to detect moving objects across a stationary camera view. A hybrid pixel representation is presented to minimize the effect of shadow illumination. A non-recursive background model is developed to address the problem with gradual illumination change. Modified Z-score labeling is employed to analyze the sample variation of the temporal sequence to build a multi-modal background. The same measure is further applied to detect the foreground pixels against the stationary background classes. Morphological filtering is employed to suppress the sensor noise as well as to fill the camouflage holes. A decision rule is formulated that considers the period of being stationary of a foreground object and the period of being absence of a background class to tackle the object relocation problem. The proposed approach along with nine other state-of-the-art methods are compared on various image sequences taken from the Wallflower and the I2R datasets in terms of recall, precision, figure of merit, and percentage of correct classification. The tabular results, as well as the obtained figures demonstrate the efficacy of the proposed scheme over its counterparts. [ABSTRACT FROM AUTHOR]

Micro- and macro-expression spotting in an untrimmed video is a challenging task, due to the mass generation of false positive samples. Most existing methods localize higher response areas by extracting hand-crafted features or cropping specific regions from all or some key raw images. However, these methods either neglect the continuous temporal information or model the inherent human motion paradigms (background) as foreground. Consequently, we propose a novel two-stream network, named Local suppression and Global enhancement Spotting Network (LGSNet), which takes segment-level features from optical flow and videos as input. LGSNet adopts anchors to encode expression intervals and selects the encoded deviations as the object of optimization. Furthermore, we introduce a Temporal Multi-Receptive Field Feature Fusion Module (TMRF $^{3}$ 3 M) and a Local Suppression and Global Enhancement Module (LSGEM), which help spot short intervals more precisely and suppress background information. To further highlight the differences between positive and negative samples, we set up a large number of random pseudo ground truth intervals (background clips) on some discarded sliding windows to accomplish background clips modeling to counteract the effect of non-expressive face and head movements. Experimental results show that our proposed network achieves state-of-the-art performance on the CAS(ME) $^{2}$ 2 , CAS(ME) $^{3}$ 3 and SAMM-LV datasets. [ABSTRACT FROM AUTHOR]

This paper introduces a novel background subtraction method that utilizes texture-level analysis based on the Gabor filter bank and statistical moments. The method addresses the challenge of accurately detecting moving objects that exhibit similar color intensity variability or texture to the surrounding environment, which conventional methods struggle to handle effectively. The proposed method accurately distinguishes between foreground and background objects by capturing different frequency components using the Gabor filter bank and quantifying the texture level through statistical moments. Extensive experimental evaluations use datasets featuring varying lighting conditions, uniform and non-uniform textures, shadows, and dynamic backgrounds. The performance of the proposed method is compared against other existing methods using metrics such as sensitivity, specificity, and false positive rate. The experimental results demonstrate that the proposed method outperforms other methods in accuracy and robustness. It effectively handles scenarios with complex backgrounds, lighting changes, and objects that exhibit similar texture or color intensity as the background. Our method retains object structure while minimizing false detections and noise. This paper provides valuable insights into computer vision and object detection, offering a promising solution for accurate foreground detection in various applications such as video surveillance and motion tracking.

To solve the problems that the background reference frames are seriously contaminated by the foreground, and the bit rate increases suddenly incurred by the one-time transmission of the background frames, a progressive background frame generation method with image block as the basic unit was proposed for surveillance video application.An image block codebook model based on clustering was formulated.The image blocks at the same position in the video sequence were effectively clustered by using perceptual hash-based element matching.The background symbol was accurately detected by using the characteristics of the background image area.A complete background frame was produced by extracting the background blocks in different frames based on the codebook model.Experimental results demonstrate that the proposed method achieves 17.89% coding efficiency for luma component compared with standard HM16.20, and can effectively improve the quality of the produced background reference frame.Besides, the proposed method complexity meets the real-time requirements of video applications.

With the development of industrial informatization, video processing technology is receiving more and more attention. Extracting background is a prerequisite for many video processing techniques, so video background modeling technology is becoming highly sought-after. Currently, there are a variety of approaches to estimating background; however, many of these methods have the fault of not being able to accurately distinguish between foreground and background, especially when objects move slowly or remain still for a period of time. In this paper, a novel background modeling scheme is proposed for surveillance video, based on keyframe and particle shape properties. The model consists of three parts: the first part is to reduce running time and eliminate the ghost phenomenon caused by adjacent redundant frames by extracting keyframe and dividing the extracted frames into several groups; the second part includes three steps, computing binarized difference, characterizing the binarized difference and screening the difference where a quadruple, composed of particle shape properties, is designed to quantitatively describe binarized differences; the third part involves generating the temporary background and updating the temporary background according to the similarity of data obtained at the newly proposed locations. Experiment results on SBMnet and SBI datasets and comparisons with some emerging algorithms show that the performance of the proposed model is superior or comparable to the other state-of-the-art methods particularly when dealing with stationary objects. Furthermore, the proposed method ranks as the second for intermittent category video, compared to the other 31 state-of-art methods. Moreover, the speed of the proposed method, 5.38 Frames Per Second (FPS) for SBMnet dataset and 10.94 FPS for SBI dataset, is faster than most public methods.

Copyright of Journal of Donghua University (Natural Science Edition) is the property of Journal of Donghua University (Natural Science) Editorial Office and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

In the dim-small target detection field, background suppression is a key technique for stably extracting the target. In order to effectively suppress the background to enhance the target, this paper presents a novel background modeling algorithm, which constructs base functions for each pixel based on the local region background and models the background of each pixel, named single pixel background modeling (SPB). In SPB, the low-rank blocks of the local backgrounds are first obtained to construct the background base functions of the center pixel. Then, the background of the center pixel is optimally estimated by the background bases. Experiments demonstrate that in the case of extremely low signal-to-noise ratio (SNR < 1.5 dB) and complex motion state of targets, SPB can stably and effectively separate the target from the strongly undulant sky background. The difference image obtained via SPB background modeling has the characters: the non-target residual could be white noise, and the target is significantly enhanced. Compared with the other typical five algorithms, SPB remarkably outperforms other algorithms to detect the target of a low signal-to-noise ratio. [ABSTRACT FROM AUTHOR]

To solve the problems that the background reference frames are seriously contaminated by the foreground, and the bit rate increases suddenly incurred by the one-time transmission of the background frames, a progressive background frame generation method with image block as the basic unit was proposed for surveillance video application. An image block codebook model based on clustering was formulated. The image blocks at the same position in the video sequence were effectively clustered by using perceptual hash-based element matching. The background symbol was accurately detected by using the characteristics of the background image area. A complete background frame was produced by extracting the background blocks in different frames based on the codebook model. Experimental results demonstrate that the proposed method achieves 17.89% coding efficiency for luma component compared with standard HM16.20, and can effectively improve the quality of the produced background reference frame. Besides, the proposed method complexity meets the real-time requirements of video applications. [ABSTRACT FROM AUTHOR]

Foliage penetration is an unsolved important part of border surveillance of remote areas between regular border crossing points. Detecting penetrating objects (e.g., persons and cars) through dense foliage in various climate conditions using visual sensors is prone to high fault rates. Through-foliage scenarios contain an unprecedented amount of occlusion—in fact, they often contain fragmented occlusion (for example, looking through the branches of a tree). Current state-of-the-art detectors based on deep learning perform inadequately under moderate-to-heavy fragmented occlusion. The FOLDOUT project builds a system that combines various sensors and technologies to tackle this problem. Consequently, a hyperspectral sensor was investigated due to its extended spectral bandwidth, beyond the range of typical RGB sensors, where vegetation exhibits pronounced reflectance. Due to the poor performance of deep learning approaches in through-foliage scenarios, a novel background modeling-based detection approach was developed, dedicated to the characteristics of the hyperspectral sensor, namely strong correlations between adjacent spectral bands and high redundancy. The algorithm is based on local dimensional reduction, where the principal subspace of each pixel is maintained and adapted individually over time. The successful application of the proposed algorithm is demonstrated in a through-foliage scenario comprised of heavy fragmented occlusion and a highly dynamical background, where state-of-the-art deep learning detectors perform poorly. [ABSTRACT FROM AUTHOR]

Automatic anomaly activity detection is difficult in video surveillance applications due to variations in size, type, shape, and objects' location. The traditional anomaly detection and classification methods may affect the overall segmentation accuracy. It requires the working groups to judge their constant attention if the captured activities are anomalous or suspicious. Therefore, this defect creates the need to automate this process with high accuracy. In addition to being extraordinary or questionable, the display does not contain the necessary recording frame and activity standard to help the quick judgment of the parts' specialized action. Therefore, to reduce the wastage of time and labor, this work utilizes a Bayesian Feed-Forward Neural Network (BFFN) algorithm to automate the anomaly Recognition System. It aims to detect irregularities, regular signs of aggression, and violence automatically, rather than the conventional methods. This work uses Bayesian Feed Forward Neural Networks to identify and classify frames with high activity levels. From there, the proposed system can issue detection alerts for threats indicating suspicious activity at a particular moment. The simulation results show that the proposed BFFN based anomaly detection responds better than existing methods. The sensitivity, specificity, accuracy, and F1-score of the proposed system are 98.69%, 97.01%, 99.65%, and 0.98%. [ABSTRACT FROM AUTHOR]

Accurate quantitative plankton observation is significant for biogeochemistry and environmental monitoring. However, current observation equipment is mostly shipborne, and there is a lack of long-term, large-scale, and low-cost methods for plankton observation. This paper proposes a solution to investigate plankton using a Seascan holographic camera equipped with a "Petrel-II" underwater glider for a longer time sequence and at a larger scale. Aiming at the new challenges of low efficiency and low accuracy of holographic image processing after integrating holographic imaging systems and underwater gliders, a novel plankton data analysis method applicable to Digital Holographic Underwater Gliders (DHUG) is proposed. The algorithm has the following features: (1) high efficiency: the algorithm breaks the traditional hologram information extraction order, focusing only on the key regions in the hologram and minimizing the redundant computation; (2) high accuracy: applying the Sobel variance algorithm to the plankton in the hologram to focus the plane extraction significantly improves the focus accuracy; and (3) high degree of automation: by integrating a convolutional neural network, the algorithm achieves a fully automated analysis of the observed data. A sea test in the South China Sea verified that the proposed algorithm could greatly improve the problems of severe plankton segmentation and the low focusing accuracy of traditional information extraction algorithms. It also proved that the DHUG plankton survey has great potential. [ABSTRACT FROM AUTHOR]

Introduction: Studying the behavioral responses and movement trajectories of insects under different stimuli is crucial for developing more effective biological control measures. Therefore, accurately obtaining the movement trajectories and behavioral parameters of insects in three-dimensional space is essential., Methods: This study used the litchi pest Thalassodes immissaria as the research object. A special binocular vision observation system was designed for nighttime movement. A thermal infrared camera was used for video recording of T. immissaria in a lightless environment. Moreover, a multi-object tracking method based on the YOLOX-GMM and SORT-Pest algorithms was proposed for tracking T. immissaria in thermal infrared images. By obtaining the central coordinates of the two T. immissaria in the video, target matching and 3D trajectory reconstruction in the parallel binocular system were achieved., Results: Error analysis of the T. immissaria detection and tracking model, as well as the 3D reconstruction model, showed that the average accuracy of T. immissaria detection reached 89.6%, tracking accuracy was 96.9%, and the average error of the reconstructed 3D spatial coordinates was 15 mm., Discussion: These results indicate that the method can accurately obtain the 3D trajectory and motion parameters of T. immissaria . Such data can greatly contribute to researchers' comprehensive understanding of insect behavioral patterns and habits, providing important support for more targeted control strategies., Competing Interests: The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest., (Copyright © 2024 Qiu, Xia, Zeng, Huang, Xin, Jiang, Wu, Ma and Li.)

Detecting moving objects has been widely studied since it plays vital many applications, such as video surveillance and intelligent transportation systems. It is necessary to accurately differentiate the foreground and the background in this technology to analyze object motions in the scene. Conventional detection methods use many reference frames to model the background to detect moving objects; however, the detection is inaccurate when immediate changes occur in the scene because the instant update of the background model is impossible. To be robust illumination changes and dynamic backgrounds, we propose an accurate and instant background modeling (AIBM) method that inpaints the background with superpixels and enhances it in detail with pixel-levels. Unlike the previous approaches, the proposed AIBM method utilizes spatio-temporal information of only two consecutive frames to eliminate the lengthy initialization and update period of the model. In this paper, we illustrate the importance of accurate and instant background modeling in detecting moving objects. The performance of our method is evaluated with three benchmark datasets (CDnet2014, LASIESTA, and SBI). The experimental results show that our AIBM method is robust to sudden changes in the scene and outperforms the other conventional methods with F-measure of 0.8911 and 0.9682 in detection accuracy in CDNet2014 and LASIESTA datasets, respectively. The accuracy of the generated background is also measured on the SBI dataset, which demonstrates the importance of high-quality background modeling. [ABSTRACT FROM AUTHOR]

A large number of modern video background modeling algorithms deal with computational costly minimization problems that often need parameter adjustments. While in most cases spatial and temporal constraints are added artificially to the minimization process, our approach is to exploit Dynamic Mode Decomposition (DMD), a spectral decomposition technique that naturally extracts spatio-temporal patterns from data. Applied to video data, DMD can compute background models. However, the original DMD algorithm for background modeling is neither efficient nor robust. In this paper, we present an equivalent reformulation with constraints leading to a more suitable decomposition into fore- and background. Due to the reformulation, which uses sparse and low-dimensional structures, an efficient and robust algorithm is derived that computes accurate background models. Moreover, we show how our approach can be extended to RGB data, data with periodic parts, and streaming data enabling a versatile use. [ABSTRACT FROM AUTHOR]

Illegal parking is a violation that often occurs in urban areas. Supervision is conducted in places where parking is prohibited to overcome this problem. Manual monitoring requires high staff performance and costs. Therefore, a video-based illegal parking detection is proposed. Previous studies have several weaknesses in videobased illegal parking detection, especially those that use background modeling techniques. Examples of drawbacks include changes in illumination and imperfect segmentation caused by noise, leading to false-positive errors. This study aims to overcome the problem of frequently occurring false-positive errors by using the improved visual background extractor (I-ViBe). This method develops the ViBe approach to detect illegal parking by modifying the background model update mechanism with the static region extraction and vehicle verification processes and integrating tracking processes. Experimental results show that the values of precision, recall, and f-measure are 100%, 60%, and 75% respectively, which means that the proposed system has overcome all false-positive problems. [ABSTRACT FROM AUTHOR]

Copyright of Telecommunication Engineering is the property of Telecommunication Engineering and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Fall detection (FD) system tends to monitor the fall events with restricted movement patterns and provides alerts to detect actions and corresponds to human falls. Based on high-level features, the resultant information often requires well-detected results like activity monitoring, detection, and classification. The objective of the study focuses on the vision-based FD and activity monitoring (AM) methods using different types of cameras and determines the finest method for different backgrounds and complex surroundings in outdoor and indoor scenes. Several works of literature provide various detection algorithms which cannot differentiate the fall from other actions. So, there is a need for efficient detection techniques which can efficiently work on all sorts of fall event images. Also, the AM algorithm lies in different classification techniques but it is not robust to classify the actions being the same speed with the fall such as jumping, bending, etc. In this paper, we view the comparative study of vision-based FD and monitoring techniques such as Inactivity/Body shape change based, Posture based, 3D head motion-based, Spatial–temporal based, Gait based and skeleton tracking techniques based on the source of their techniques, types, description, advantages, and disadvantages. In addition, several performance metrics were used to evaluate the results and compare the resulting study with the previous comparative evaluations. This comparative analysis leads to a deeper understanding of different FD and AM techniques and suggests the possible direction for the researchers to identify a suitable method for their needs. [ABSTRACT FROM AUTHOR]

In a video surveillance system, background modeling is assumed to be a fundamental technique for moving object detection. The surveillance system based on thermal video overcomes many challenges, such as background variations, varying light intensity, external illumination source, and so on. This paper presents a new method for background modeling and background subtraction. The method utilizes the combined approach of Fisher's Linear Discriminant and Relative Entropy for pixel based classification and detection of moving objects in thermal video frames. The experimental results show the higher average value of various performance indicators like Accuracy, ROC, and F-measure. In contrast, the percentage of false classification and total error is minimum and also has lesser execution time. The method outperforms when compared with the other existing methods. [ABSTRACT FROM AUTHOR]

Copyright of Journal of Harbin University of Science & Technology is the property of Journal of Harbin University of Science & Technology and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Background subtraction is commonly employed in foreground object detection in urban traffic scenes. Most of the current color or texture feature-based background subtraction models are easily contaminated by sudden and gradual illumination variations in urban traffic scenes. To resolve this deficiency, an adaptive local median texture feature, which extracts the adaptive distance threshold employing the median information in a predefined local region of a pixel and Weber's law, is introduced. In addition, a sample consensus-based model that evolved from portable visual background extractor is proposed using an adaptive local median texture feature. Then, the foreground is labeled by comparing the input video frames feature with the model. Moreover, to adapt the dynamic background, the random update scheme is used to update the model. Extensive experimental results on the public Change Detection data set of 2014 (CDnet2014) and the real-world urban traffic videos demonstrate that our background subtraction method is superior to the other state-of-the-art texture-feature-based methods. The qualitative and quantitative results show the encouraging efficiency of the proposed technique to deal with sudden and gradual illumination variations in real-world urban traffic scenes.

In the dim-small target detection field, background suppression is a key technique for stably extracting the target. In order to effectively suppress the background to enhance the target, this paper presents a novel background modeling algorithm, which constructs base functions for each pixel based on the local region background and models the background of each pixel, named single pixel background modeling (SPB). In SPB, the low-rank blocks of the local backgrounds are first obtained to construct the background base functions of the center pixel. Then, the background of the center pixel is optimally estimated by the background bases. Experiments demonstrate that in the case of extremely low signal-to-noise ratio (SNR < 1.5 dB) and complex motion state of targets, SPB can stably and effectively separate the target from the strongly undulant sky background. The difference image obtained via SPB background modeling has the characters: the non-target residual could be white noise, and the target is significantly enhanced. Compared with the other typical five algorithms, SPB remarkably outperforms other algorithms to detect the target of a low signal-to-noise ratio.

Background modeling of video frame sequences is a prerequisite for computer vision applications. Robust principal component analysis(RPCA), which aims to recover low rank matrix in applications of data mining and machine learning, has shown improved background modeling performance. Unfortunately, The traditional RPCA method considers the batch recovery of low rank matrix of all samples, which leads to higher storage cost. This paper proposes a novel online motion-aware RPCA algorithm, named OM-RPCAT, which adopt truncated nuclear norm regularization as an approximation method for of low rank constraint. And then, Two methods are employed to obtain the motion estimation matrix, the optical flow and the frame selection, which are merged into the data items to separate the foreground and background. Finally, an efficient alternating optimization algorithm is designed in an online manner. Experimental evaluations of challenging sequences demonstrate promising results over state-of-the-art methods in online application. [ABSTRACT FROM AUTHOR]