Moving Object Detection Research Articles

Human Object Counter Tracking Using Connected Component Labelling On Digital Image Processing

Object tracking is usually used to determine the position of objects and environmental conditions around these objects. Objects that require supervision in order to run properly require a system to monitor the object. Every movement of the object will be known directly through the camera image. So in this study a system was created to detect and track objects. Implementing an object detection system based on image processing using the connected component labeling method. Then displaying the detection results on the divided image frame. Moving object detection and tracking The algorithm used to detect and track moving objects is connected component labeling. Based on the analysis of the workings of the connected component labeling algorithm for detecting and tracking moving objects using video, it can be concluded that the connected component labeling algorithm has succeeded in tracking and labeling each object based on the position and number of objects, even though they look erratic, in human movement.

A gate‐tunable memristor emulator for motion detection

SummaryWith its low power consumption and small size, the memristor has shown great potential for improving data storage density and computing efficiency. Compared to the dual‐port memristor, greater attention should be paid to researching gate‐tunable memristor for image processing to improve the processing speed and reduce hardware resource consumption. Developing gate‐tunable memristor emulators is highly attractive given the immaturity of current fabrication of the gate‐tunable memristor. This work proposes a digital gate‐tunable memristor emulator based on Raspberry Pi, which addresses the non‐reconfigurability and inflexibility issues of the analog emulators. The proposed emulator can match the behavior of different memristor devices by regulating the gate voltage parameter. Additionally, it can operate at a maximum frequency of 500 MHz. To test the functionality of the proposed emulator, a digital implementation of the memristive circuit for motion detection is designed and verified experimentally. Experiments demonstrate that when moving object detection is performed on a 640 × 350 pixel video stream, low power consumption of 53 mW and a delay of 3.52 μs can be achieved.

A Survey Paper on Moving Object Detection Using Deep Learning

Moving object detection in Python using deep learning is a powerful technique for accurately identifying and localizing moving objects in images or videos. By leveraging pre-trained models like YOLO or SSD, developers can implement this task efficiently. The Python implementation allows for customization and extension to handle real-time video streams and complex scenarios. This approach is valuable for researchers, practitioners, and enthusiasts interested in moving object detection using deep learning. Deep Convolution Neural Networks are leveraged to detect more precise coordinates and identify the category of objects. This survey paper provides study of various methodologies for object detection. This paper provides systematic analysis of various existing object detection techniques with precise and arranged representation.

Infrared Moving Small Target Detection Based on Space–Time Combination in Complex Scenes

In the infrared small target images with complex backgrounds, there exist various interferences that share similar characteristics with the target (such as building edges). The accurate detection of small targets is crucial in applications involving infrared search and tracking. However, traditional detection methods based on small target feature detection in a single frame image may result in higher error rates due to insufficient features. Therefore, in this paper, we propose an infrared moving object detection method that integrates spatio-temporal information. To address the limitations of single-frame detection, we introduce a temporal sequence of images to suppress false alarms caused by single-frame detection through analyzing motion features within the sequence. Firstly, based on spatial feature detection, we propose a multi-scale layered contrast feature (MLCF) filtering for preliminary target extraction. Secondly, we utilize the spatio-temporal context (STC) as a feature to track the image sequence point by point, obtaining global motion features. Statistical characteristics are calculated to obtain motion vector data that correspond to abnormal motion, enabling the accurate localization of moving targets. Finally, by combining spatial and temporal features, we determine the precise positions of the targets. The effectiveness of our method is evaluated using a real infrared dataset. Through analysis of the experimental results, our approach demonstrates stronger background suppression capabilities and lower false alarm rates compared to other existing methods. Moreover, our detection rate is similar or even superior to these algorithms, providing further evidence of the efficacy of our algorithm.

MOD-IR: moving objects detection from UAV-captured video sequences based on image registration

MOD-IR: moving objects detection from UAV-captured video sequences based on image registration

Moving object detection using modified GMM based background subtraction

Academics have become increasingly interested in creating cutting-edge technologies to enhance Intelligent Video Surveillance (IVS) performance in terms of accuracy, speed, complexity, and deployment. It has been noted that precise object detection is the only way for IVS to function well in higher level applications including event interpretation, tracking, classification, and activity recognition. Through the use of cutting-edge techniques, the current study seeks to improve the performance accuracy of object detection techniques based on Gaussian Mixture Models (GMM). It is achieved by developing crucial phases in the object detecting process. In this study, it is discussed how to model each pixel as a mixture of Gaussians and how to update the model using an online k-means approximation. The adaptive mixture model's Gaussian distributions are then analyzed to identify which ones are more likely to be the product of a background process. Each pixel is categorized according to whether the background model is thought to include the Gaussian distribution that best depicts it.

FenceTalk: Exploring False Negatives in Moving Object Detection

Deep learning models are often trained with a large amount of labeled data to improve the accuracy for moving object detection in new fields. However, the model may not be robust enough due to insufficient training data in the new field, resulting in some moving objects not being successfully detected. Training with data that is not successfully detected by the pre-trained deep learning model can effectively improve the accuracy for the new field, but it is costly to retrieve the image data containing the moving objects from millions of images per day to train the model. Therefore, we propose FenceTalk, a moving object detection system, which compares the difference between the current frame and the background image based on the structural similarity index measure (SSIM). FenceTalk automatically selects suspicious images with moving objects that are not successfully detected by the Yolo model, so that the training data can be selected at a lower labor cost. FenceTalk can effectively define and update the background image in the field, reducing the misjudgment caused by changes in light and shadow, and selecting images containing moving objects with an optimal threshold. Our study has demonstrated its performance and generality using real data from different fields. For example, compared with the pre-trained Yolo model using the MS COCO dataset, the overall recall of FenceTalk increased from 72.36% to 98.39% for the model trained with the data picked out by SSIM. The recall of FenceTalk, combined with Yolo and SSIM, can reach more than 99%.

DeepFTSG: Multi-stream Asymmetric USE-Net Trellis Encoders with Shared Decoder Feature Fusion Architecture for Video Motion Segmentation

Discriminating salient moving objects against complex, cluttered backgrounds, with occlusions and challenging environmental conditions like weather and illumination, is essential for stateful scene perception in autonomous systems. We propose a novel deep architecture, named DeepFTSG, for robust moving object detection that incorporates single and multi-stream multi-channel USE-Net trellis asymmetric encoders extending U-Net with squeeze and excitation (SE) blocks and a single shared decoder network for fusing multiple motion and appearance cues. DeepFTSG is a deep learning based approach that builds upon our previous hand-engineered flux tensor split Gaussian (FTSG) change detection video analysis algorithm which won the CDNet CVPR Change Detection Workshop challenge competition. DeepFTSG generalizes much better than top-performing motion detection deep networks, such as the scene-dependent ensemble-based FgSegNet_v2, while using an order of magnitude fewer weights. Short-term motion and longer-term change cues are estimated using general-purpose unsupervised methods—flux tensor and multi-modal background subtraction, respectively. DeepFTSG was evaluated using the CDnet-2014 change detection challenge dataset, the largest change detection video sequence benchmark with 12.3 billion labeled pixels, and had an overall F-measure of 97%. We also evaluated the cross-dataset generalization capability of DeepFTSG trained solely on CDnet-2014 short video segments and then evaluated on unseen SBI-2015, LASIESTA and LaSOT benchmark videos. On the unseen SBI-2015 dataset, DeepFTSG had an F-measure accuracy of 87%, more than 30% higher compared to the top-performing deep network FgSegNet_v2 and outperforms the recently proposed KimHa method by 17%. On the unseen LASIESTA, DeepFTSG had an F-measure of 88% and outperformed the best recent deep learning method BSUV-Net2.0 by 3%. On the unseen LaSOT with axis-aligned bounding box ground-truth, network segmentation masks were converted to bounding boxes for evaluation, DeepFTSG had an F-Measure of 55%, outperforming KimHa method by 14% and FgSegNet_v2 by almost 1.5%. When a customized single DeepFTSG model is trained in a scene-dependent manner for comparison with state-of-the-art approaches, then DeepFTSG performs significantly better, reaching an F-Measure of 97% on SBI-2015 (+ 10%) and 99% on LASIESTA (+ 11%). The source code, pre-trained weights, and video demo for DeepFTSG are available at https://github.com/CIVA-Lab/DeepFTSG.

DCS-CTN: Subtle Gesture Recognition Based on TD-CNN-Transformer via Millimeter-Wave Radar

Gesture recognition has been a hot research topic in human-computer interaction, since contactless gesture recognition will provide increasing applications in many fields. Millimeter-wave (mmWave) radar well serves this technology because of its high accuracy, easy integration, and strong anti-jamming ability in moving object detection. However, it is still challenging to meet the requirement of high precision in subtle gesture recognition based on traditional methods via point cloud or Range-Doppler heat map of millimeter-wave radar. Considering the raw data from millimeter-wave radar with more information such as phase, we propose a system that uses the constructed millimeter-wave radar data cube sequence and Timedistributed-CNN-Transformer network (CTN), called DCS-CTN system, to get higher hand gesture recognition accuracy. In this system, we introduce a time-distributed wrapper (TD) and convolution neural network (CNN) to extract local features of the data cube sequence, a position encoder to retain time information of the sequence, and a transformer network to get global features of the sequence. The experiments results show that this system can achieve hand gesture recognition accuracy of 99.75%, which is significantly higher than the other traditional approaches.

Design of a Temperature Control and Object Motion Detection System in the Server Room Using IOT-Based Wemos D1

Servers play an important role in managing information and managing network traffic within an agency. It contains various types of network devices that must be kept on every day to provide the best service. This leads to the importance of maintaining server presence, such as paying attention to room humidity, regulating temperature and monitoring server room movements to keep it under control. The air temperature in the server room is set within 20 - 25 degrees Celsius with a relative humidity of 40 - 55%. In this context, the concept of the Internet of Things (IoT) emerges as a potential solution by connecting physical objects via the internet. This research designs and builds an IoT-based temperature control and object motion detection system using a DHT22 sensor to measure the temperature in the room and a PIR sensor to detect human movement. Hardware components such as Wemos D1, Buzzer, Relay 2 channel 5v, and others are used to control the system via a smartphone. The software used includes the Arduino IDE. This system aims to monitor and control room temperature and object movement detection systems intelligently, prevent overheating, reduce the risk of loss and combine the advantages of IoT technology to create a temperature control and motion detection system that is responsive to room temperature and the movement of objects in the room.

Saliency-Induced Moving Object Detection for Robust RGB-D Vision Navigation Under Complex Dynamic Environments

Localization in unknown environments is an essential requirement for vision navigation of robotic vehicles in intelligent transportation systems. However, moving objects in dynamic scenarios usually bring about great difficulty for robot localization, because motion estimation of robotic vehicles is disturbed by increasing feature outliers caused by moving objects. In order to improve the accuracy and robustness of robot localization, a novel saliency-induced moving object detection (SMOD) approach is proposed to filter out feature outliers for RGB-D-based simultaneous localization and mapping (SLAM) in complex dynamic workspaces. Firstly, three complementary motion saliency potentials, including motion energy (ME), spatiotemporal objectness (STO), and dynamic superpixels (DS), are modeled by fully analyzing spatial, temporal, appearance and depth cues in RGB-D inputs. They can be used to identify the dynamic objects effectively from diversely changing backgrounds. Then, a superpixel-level graph-based motion saliency (MS) measure is proposed to generate the MS map for reliable localization of the moving objects. The edge weights and background nodes on the graph are determined reasonably by fusing ME, STO, and DS, which is not vulnerable to background interferences. Furthermore, the SMOD approach is embedded into the front-end of ORB-SLAM3 as a pre-processing stage, in order to filter out feature outliers associated with the moving objects. Finally, the extensive experiments are performed to verify the accuracy and robustness of the proposed approach on the public dynamic datasets. The experimental results show that the SMOD method can detect the moving objects effectively in a variety of challenging dynamic environments, and separate the dynamic regions reliably from the irrelevant background. The data comparison demonstrates that the SMOD-SLAM navigation system can outperform other state-of-the-art dynamic visual SLAM (vSLAM) systems.

EMOS: Enhanced moving object detection and classification via sensor fusion and noise filtering

AbstractDynamic object detection is essential for ensuring safe and reliable autonomous driving. Recently, light detection and ranging (LiDAR)‐based object detection has been introduced and shown excellent performance on various benchmarks. Although LiDAR sensors have excellent accuracy in estimating distance, they lack texture or color information and have a lower resolution than conventional cameras. In addition, performance degradation occurs when a LiDAR‐based object detection model is applied to different driving environments or when sensors from different LiDAR manufacturers are utilized owing to the domain gap phenomenon. To address these issues, a sensor‐fusion‐based object detection and classification method is proposed. The proposed method operates in real time, making it suitable for integration into autonomous vehicles. It performs well on our custom dataset and on publicly available datasets, demonstrating its effectiveness in real‐world road environments. In addition, we will make available a novel three‐dimensional moving object detection dataset called ETRI 3D MOD.

Background memory‐assisted zero‐shot video object segmentation for unmanned aerial and ground vehicles

AbstractUnmanned aerial vehicles (UAV) and ground vehicles (UGV) require advanced video analytics for various tasks, such as moving object detection and segmentation; this has led to increasing demands for these methods. We propose a zero‐shot video object segmentation method specifically designed for UAV and UGV applications that focuses on the discovery of moving objects in challenging scenarios. This method employs a background memory model that enables training from sparse annotations along the time axis, utilizing temporal modeling of the background to detect moving objects effectively. The proposed method addresses the limitations of the existing state‐of‐the‐art methods for detecting salient objects within images, regardless of their movements. In particular, our method achieved mean and values of 82.7 and 81.2 on the DAVIS'16, respectively. We also conducted extensive ablation studies that highlighted the contributions of various input compositions and combinations of datasets used for training. In future developments, we will integrate the proposed method with additional systems, such as tracking and obstacle avoidance functionalities.

Dual-Frequency Signal Enhancement Method of Moving Target Echoes for GNSS-S Radar

The GNSS-S radar utilizes the signals of a global navigation satellite system (GNSS) to carry out target detection. Due to the very low power of GNSS signals, long-term accumulation is needed to improve the gain of the echo signals. However, when it is used for moving object detection, the random movement of the target will cause residual Doppler frequency after the echoes are correlated and compressed through the direct signal. The residual Doppler frequency will cause two problems: on the one hand, the signal coherence will deteriorate, affecting the coherent accumulation gain; on the other hand, the amplitude of the signal after compression will decrease due to the sensitivity of GNSS signals to Doppler frequency. Therefore, how to increase the signal amplitude and eliminate the phase fluctuation caused by the Doppler frequency shift in the GNSS echoes of moving targets is an important issue for GNSS-S radar to detect moving targets. This paper proposes a dual-frequency GNSS echo enhancement method that uses the dual-frequency signals transmitted by the GNSS satellites to enhance and regularize the target echo. First, the phase relationship model of the GNSS dual-frequency echo is constructed, and the phase difference is made to the compressed dual-frequency echo signal to obtain the differential phase without fluctuation; then, the amplitudes of the dual-frequency echo signals are added together; and finally, a new signal with enhanced amplitude and consistent phase is constructed by using the dual-frequency additive amplitude and differential phase, and the long-term coherent accumulation of the signal is carried out, which can improve the processing gain of the weak echo signal of the moving target. The simulation and field experiments show that this method makes full use of the energy of the GNSS dual-frequency signal and eliminates the phase fluctuation in the echo signal of the moving target so that the compressed signal energy remains consistent in the slow-time dimension. After long-term coherent accumulation, the echo SNR was greatly improved, which enabled the detection of two high-speed cars by GNSS-S radar in the experiment.

A novel deep convolutional encoder–decoder network: application to moving object detection in videos

A novel deep convolutional encoder–decoder network: application to moving object detection in videos

Nonconvex $$\gamma $$-norm and Laplacian scale mixture with salient map for moving object detection

Nonconvex $$\gamma $$-norm and Laplacian scale mixture with salient map for moving object detection

Nearly Panoramic Neuromorphic Vision with Transparent Photosynapses.

Neuromorphic vision based on photonic synapses has the ability to mimic sensitivity, adaptivity, and sophistication of bio-visual systems. Significant advances in artificial photosynapses are achieved recently. However, conventional photosyanptic devices normally employ opaque metal conductors and vertical device configuration, performing a limited hemispherical field of view. Here, a transparent planar photonic synapse (TPPS) is presented that offers dual-side photosensitive capability for nearly panoramic neuromorphic vision. The TPPS consisting of all two dimensional (2D) carbon-based derivatives exhibits ultra-broadband photodetecting (365-970nm) and ≈360° omnidirectional viewing angle. With its intrinsic persistent photoconductivity effect, the detector possesses bio-synaptic behaviors such as short/long-term memory, experience learning, light adaptation, and a 171% pair-pulse-facilitation index, enabling the synapse array to achieve image recognition enhancement (92%) and moving object detection.

Integrated and Deep Learning Based Social Surveillance System: A Novel Approach

In industry and research, big data applications are gaining a lot of traction and space. Surveillance videos contribute significantly to big unlabelled data. The aim of visual surveillance is to understand and determine object behaviour. It includes static and moving object detection, as well as video tracking to comprehend scene events. Object detection algorithms may be used to identify items in any video scene. Any video surveillance system faces a significant challenge in detecting moving objects and differentiating between objects with the same shapes or features. The primary goal of this work is to provide an integrated framework for a quick overview of video analysis utilizing deep learning algorithms to detect suspicious activity. In greater applications, the detection method is utilized to determine the region where items are available and the form of objects in each frame. This video analysis also aids in the attainment of security. Security may be characterized in a variety of ways, such as identifying theft or violation of covid protocols. The obtained results are encouraging and superior to existing solutions with 97% accuracy.

Motion and geometry-related information fusion through a framework for object identification from a moving camera in urban driving scenarios

Accurate and robust semantic scene understanding for urban driving is challenging due to the complex object types, object motion, and as well as ego-motion. Typical approaches to this problem use the fusion of multiple sensors such as camera, IMU, LiDAR, and radar to identify the state of the surrounding objects, including distance, direction, position, and velocity. However, such sensor modalities are very much complex and costly. This paper proposes a new framework for object identification (FOI) from a moving camera in a complex driving environment utilizing only camera sensor data. The framework is capable of detecting objects, extracting their behavioral features in terms of motion, position, velocity, and distance. All of this information (referred to as object-wise semantic information) are fused in order to acquire a better understanding of the driving scenario. The work addresses ego-motion compensation and extraction of accurate motion information of moving objects from a moving camera using image registration and optical flow estimation. A moving object detection model is designed within the framework by integrating an encoder–decoder network with a semantic segmentation network. The approach involves two mutual tasks: semantic segmentation of objects into specific classes and binary pixel classification to predict whether the detected object is moving or static based on temporal information. The work also contributes a novel dataset for moving object detection that covers all types of dynamic objects. The evaluation of FOI has been performed on different sequences of KITTI, EU-life long, and the proposed datasets. The experimental results show that the proposed framework provides accurate object-wise semantic information.

Stereo SLAM in Dynamic Environments Using Semantic Segmentation

As we all know, many dynamic objects appear almost continuously in the real world that are immensely capable of impairing the performance of the majority of vision-based SLAM systems based on the static-world assumption. In order to improve the robustness and accuracy of visual SLAM in high-dynamic environments, a real-time and robust stereo SLAM system for dynamic scenes was proposed. To weaken the influence of dynamic content, the moving-object detection method was put forward in our visual odometry, and then the semantic segmentation network was combined in our stereo SLAM to extract pixel-level contours of dynamic objects. Then, the influences of dynamic objects were significantly weakened and the performance of our system increased markedly in dynamic, complex, and crowed city spaces. Following experiments with both the KITTI Odometry dataset and in a real-life scene, the results showed that our method could dramatically decrease the tracking error or drift, and improve the robustness and stability of our stereo SLAM in high dynamic outdoor scenarios.