Object Tracking Applications Research Articles

OptiRet-Net: An Optimized Low-Light Image Enhancement Technique for CV-based Applications in Resource-Constrained Environments

The illumination of images can significantly impact computer-vision applications such as image classification, multiple object detection, and tracking, leading to a significant decline in detection and tracking accuracy. Recent advancements in deep learning techniques have been applied to low-light image enhancement (LLIE) to combat this issue. Retinex theory-based methods following a decomposition-adjustment pipeline for LLIE have performed well in various aspects. Despite their success, current research on Retinex-based deep learning still needs to improve in terms of optimization techniques and complicated convolution connections, which can be computationally intensive for end-device deployment. We propose an Optimized Retinex-based CNN (OptiRet-Net) deep-learning framework to address these challenges for the low-light image enhancement problem. Our results demonstrate that the proposed method outperforms existing state-of-the-art models in terms of full reference metrics with a PSNR of 21.87, SSIM of 0.80, LPIPS of 0.16, and zero reference metrics with a NIQE of 3.4 and PIQE of 56.6. Additionally, we validate our approach using a comprehensive evaluation comprising five datasets and nine prior methods. Furthermore, we assess the efficacy of our proposed model combining low-light multiple object tracking applications using YOLOX and ByteTrack in versatile video coding (VVC/H.266) across various quantization parameters. Our findings reveal that LLIE-enhanced frames surpass their tracking results with a MOTA of 80.6% and a remarkable precision rate of 96%. Our model also achieves minimal file sizes by effectively compressing the enhanced low-light images while maintaining their quality, making it suitable for resource-constrained environments where storage or bandwidth limitations are a concern.

Optical analog computing for salient object detection in complex scenes via dielectric metasurface

Optical analog computing overcomes the throughput and computational speed limitations of traditional digital computing, facilitating salient object detection on platforms that rapidly process massive image data, such as optical remote sensing. Further device miniaturization and integration can be achieved by replacing filters in optical analog computing with metasurface. Based on the combination of optical analog computing and dielectric metasurface, we proposed an optical analog cross-correlation operation operating in the mid-infrared wavelength band to achieve the real-time and compact salient object detection. Even in extremely complex scenes, this cross-correlation operation can be used to identify and segment target patterns based on several target features accurately. The proposed optical analog cross-correlation operation may find applications in object tracking, medical image segmentation, and person re-identification.

OLOD: a new UAV dataset and benchmark for single tiny object tracking

ABSTRACT The integration of visual data obtained from unmanned aerial vehicles (UAVs) has ushered in an era of computer vision, greatly expanding the possibilities for object tracking applications. Nevertheless, existing UAV datasets predominantly focus on large-scale objects characterized by distinct contours, overlooking single tiny objects encountered in real-world flight scenarios. Extracting appearance information from these diminutive objects poses a considerable challenge for object tracking. To rectify this imbalance in data distribution, we proposed a UAV dataset called Overhead Look Of Drones (OLOD), encompassing 70 sequences meticulously designed to address tiny object tracking. It contains over 55k frames and provides supplementary information about altitude and flight attitude. Additionally, we incorporated 11 challenging attributes to enhance the complexity of the scenes, thereby establishing a comprehensive benchmark for single object tracking. OLOD serves as a valuable tool for evaluating the tracking capabilities of various algorithms when it comes to tiny objects. Subsequently, through experimental results, we shed light on the limitations of existing methods for tracking tiny objects on this benchmark, underscoring the necessity for further research in this field. Our dataset and evaluation code will be released at https://github.com/yuymf/OLOD.

Real time object tracking using deep learning and OpenCV

This article summarizes the background and application fields of Open-Source Computer Vision Library (OpenCV) and deep learning and conducts research based on their object detection and tracking applications. The model algorithm selected in this article is the Convolutional Neural Networks (CNN) algorithm. CNN algorithm can be used for object detection in real-time scenarios. Moreover, the CNN algorithm has good credibility. The Python program developed based on CNN algorithm can effectively achieve real-time object tracking. The model shows good detection and tracking performance for trained targets and can be further applied in more specific scenarios in the future. Because deep learning can process large-scale data and recognize complex patterns, it can automatically learn and extract advanced features. Combining the two can achieve faster and more accurate detection and tracking of target objects. After having a large enough training sample size, it can detect and track the specified object more accurately. However, it is precisely due to the enormous sample size required for deep learning that there are still some difficulties in applying it to real-time object tracking. This article first discusses the use of supervised learning methods for deep understanding. When the input sample capacity is large enough, the machine can better reflect the real-time detection and tracking of the target object. But the time required will significantly increase. This issue still needs to be addressed in the subsequent research process.

Cell Multi-Bernoulli (Cell-MB) Sensor Control for Multi-Object Search-While-Tracking (SWT).

Information-driven control can be used to develop intelligent sensors that can optimize their measurement value based on environmental feedback. In object tracking applications, sensor actions are chosen based on the expected reduction in uncertainty also known as information gain. Random finite set (RFS) theory provides a formalism for quantifying and estimating information gain in multi-object tracking problems. However, estimating information gain in these applications remains computationally challenging. This paper presents a new tractable approximation of the RFS expected information gain applicable to sensor control for multi-object search and tracking. Unlike existing RFS approaches, the information gain approximation presented in this paper considers the contributions of non-ideal noisy measurements, missed detections, false alarms, and object appearance/disappearance. The effectiveness of the information-driven sensor control is demonstrated through two multi-vehicle search-while-tracking experiments using real video data from remote terrestrial and satellite sensors.

Improved Real Time Target Tracking System Based on Cam-shift and Kalman Filtering Techniques

This paper presents an improved real time target tracking system based on Cam-Shift and advanced Kalman filter algorithms. The paper, proposes a simple low implementation cost scheme while keeping the high tracking performance. Two degrees of freedom target tracking system based on two hybrid stepper motors with their drive systems is fully designed and implemented to improve the tracking accuracy while keeping low-cost. The potential of the proposed algorithm lies in its ability to accurately detect, track and produce firing commands on the designated target while utilizing low cost on the shelf embedded Microcontroller boards. A low cost Universal Serial Port (USB) camera is used and integrated with the proposed algorithm in order to detect and track maneuvering targets. A series of experiments for stationary and moving objects are carried out for testing the proposed system efficiency. The proposed system enables accurate performance for tracking static and moving objects. Hence it provides a simple and low implementation cost scheme for real time object tracking applications.

Siamese Cross-Domain Tracker Design for Seamless Tracking of Targets in RGB and Thermal Videos

Multimodal (RGB and thermal) applications are swiftly gaining importance in the computer vision community with advancements in self-driving cars, robotics, Internet of Things, and surveillance applications. Both the modalities have complementary performance depending on illumination constraints. Hence, a judicious combination of both modalities will result in robust RGBT systems capable of all-day all-weather applications. Several studies have been proposed in the literature for integrating the multimodal sensor data for object tracking applications. Most of the proposed networks try to delineate the information into modality-specific and modality shared features and attempt to exploit the modality shared features in enhancing the modality specific information. In this work, we propose a novel perspective to this problem using a Siamese inspired network architecture. We design a custom Siamese cross-domain tracker architecture and fuse it with a mean shift tracker to drastically reduce the computational complexity. We also propose a constant false alarm rate inspired coasting architecture to cater for real-time track loss scenarios. The proposed method presents a complete and robust solution for object tracking across domains with seamless track handover for all-day all-weather operation. The algorithm is successfully implemented on a Jetson-Nano, the smallest graphics processing unit (GPU) board offered by NVIDIA Corporation.

Fundamental limits to depth imaging with single-photon detector array sensors

Single-Photon Avalanche Detector (SPAD) arrays are a rapidly emerging technology. These multi-pixel sensors have single-photon sensitivities and pico-second temporal resolutions thus they can rapidly generate depth images with millimeter precision. Such sensors are a key enabling technology for future autonomous systems as they provide guidance and situational awareness. However, to fully exploit the capabilities of SPAD array sensors, it is crucial to establish the quality of depth images they are able to generate in a wide range of scenarios. Given a particular optical system and a finite image acquisition time, what is the best-case depth resolution and what are realistic images generated by SPAD arrays? In this work, we establish a robust yet simple numerical procedure that rapidly establishes the fundamental limits to depth imaging with SPAD arrays under real world conditions. Our approach accurately generates realistic depth images in a wide range of scenarios, allowing the performance of an optical depth imaging system to be established without the need for costly and laborious field testing. This procedure has applications in object detection and tracking for autonomous systems and could be easily extended to systems for underwater imaging or for imaging around corners.

A Visual Target Representation using Saliency Detection Approach

The task of saliency detection is to identify the most important and informative part of a scene. Saliency detection is broadly applied to numerous vision problems, including image segmentation, object recognition, image compression, content-based image retrieval, and moving object detection. Existing saliency detection methods suffer a low accuracy rate because of missing components of saliency regions. This study proposes a visual saliency detection method for the target representation to represent targets more accurately. The proposed method consists of five modules. In the first module, the salient region is extracted through manifold ranking on a graph, which incorporates local grouping cues and boundary priors. Secondly, using a region of interest (ROI) algorithm and the subtraction of the salient region from the original image, other parts of the image, either related or nonrelated to the interested target, are segmented. Lastly, those related and non-related regions are classified and distinguished using our proposed algorithm. Experimental result shows that proposed salient region accurately represent the interested target which can be used for object detection and tracking applications.

Hybridization of Deep Convolutional Neural Network for Underwater Object Detection and Tracking Model

In this present work, underwater object detection and tracking was studied using the efficient Hybridization of Deep Convolutional Neural Network for Underwater Object Detection and Tracking (HDCNN-UODT) model for three bench mark data sets namely UOT32, brackish, and URPC 2020 datasets. The HDCNN-UODT technique primarily employs data augmentation process for increasing the size of the training dataset to improve the detection and average precision. Besides, a hybridization of two Deep Learning (DL) models namely RetinaNet and EfficientNet models was applied as feature extractors. In addition, the bounding box prediction process takes place via support vector regression (SVR) followed by kernel extreme learning machine (KELM) model. The novelty of the present work was demonstrated by the design of SVR based bounding box regression and fusion based feature extractions. The average precision examination (APE), average success rate(ASR) & average frame per second(AFPS) analysis of the HDCNN-UODT model using UOT32 dataset showed better APE (51.27%), ASR (43.19) & AFPS (310.25) over other reported techniques. Additionally, the detection accuracy of the HDCNN-UODT model for brackish and URPC datasets showed improved accuracies for different objects over YOLO v4, T-YOLO v4, and AFFM-YOLO v4 techniques. Moreover, using brackish dataset, HDCNN-UODT model showed highest accuracy of 94.85% for ‘Crab’ object and for URPC dataset, a highest accuracy of 88.34% for ‘Scallop’ object was obtained. Hence based on our outcome, HDCNN-UODT technique might be better suited for the object detection and tracking application.

Robust object tracking via deformation samples generator

In object tracking applications, it is common for trackers to experience drift problems when the object of interest becomes deformed, which compromises the ability of the tracker to track the object. It is therefore desirable to develop a learning tracker classifier that is robust to deformations. The performance of existing trackers that employ deep classification networks degrades when the amount of training data is limited and does not cover all possible scenarios. While these limitations can be mitigated in part by using larger training datasets, these datasets may still not cover all situations and the positive samples are still monotonous. To overcome this problem, we propose a novel deformation samples generator that generates samples that would normally be difficult for the tracker to classify. In the proposed framework, both the classifier and deformation samples generator learn in a joint manner. Our experiments show that the proposed approach outperforms state-of-the-art methods in both quantitative and qualitative evaluations for the visual object tracking task.

Energy Efficient Node Deployment Technique for Heterogeneous Wireless Sensor Network based Object Detection

Lifetime of the network and the quality of operation are the two important issues in a wireless sensor network system meant for object detection and tracking application. At the same time, there should be a tradeoff between network cost and the quality of operation as high cost of the network limits its real-time usability. Heterogeneous wireless sensor networks promises the prolonged network lifetime as well as enhances network reliability as they contain a mixture of nodes with different characteristics. Further prolongation of network lifetime can be achieved by managing the available node energy in a proper way, i.e, by minimizing number of communication, minimizing node density, minimizing overhead information generated during operation etc. Proper node deployment scheme not only helps to enhance the lifetime of the network but also helps in reducing deployment cost while maintaining the quality of operation in terms of object detection accuracy. This paper focuses on the energy efficient node deployment in heterogeneous wireless sensor network system with the features of maximum network coverage, optimum node density and optimum network cost. This paper proposes a novel energy efficient node deployment algorithm that determines the number of static and mobile nodes required for deployment and then relocates the mobile nodes to cover up the coverage hole using 8-neighbourhood and Particle Swarm Optimization (PSO) algorithm. The performance of the proposed algorithm is compared with corresponding model of Harmony Search Algorithm (HSA) and PSO based node deployment and it is seen that the proposed model outperforms better in comparison to them.

Recent advancements in privacy-aware protocols of source location privacy in wireless sensor networks: A survey

This review article summarises the protocols proposed in recent researches to secure location information in Wireless Sensor Networks (WSNs). Due to their lightweightness and easy to deploy properties, WSNs are widely used in numerous object tracking and monitoring applications. Due to such, source location privacy attracts the researchers and hence continuously enhances its improvement. Though, this privacy breach is not acceptable for WSNs, as it may reveal some critical information that is harmful. The SLP issue on WSN attracted researchers a lot, and hence a number of solutions are provided for it. However, an up-to-date survey does not exist for the same. To fill this gap, in this article, we summarize different approaches proposed in the last years to preserve location privacy. We first discuss the different privacy characteristics in WSNs, a detailed overview of the proposed protocols and their limitations, and discussions of solutions for the adversaries? capabilities in WSNs. Then the future research directions in this area are discussed. This review work may support researchers identifying the new research area in location privacy of wireless sensor networks.

Configuration-Adaptive Wireless Visual Sensing System with Deep Reinforcement Learning

Visual sensing has been increasingly employed in various industrial applications including manufacturing process monitoring and worker safety monitoring. This paper presents the design and implementation of a wireless camera system, namely, EFCam, which uses low-power wireless communications and edge-fog computing to achieve cordless and energy-efficient visual sensing. The camera performs image pre-processing and offloads the data to a resourceful fog node for advanced processing using deep models. EFCam admits dynamic configurations of several parameters that form a configuration space. It aims to adapt the configuration to maintain desired visual sensing performance of the deep model at the fog node with minimum energy consumption of the camera in image capture, pre-processing, and data communications, under dynamic variations of the monitored process, the application requirement, and wireless channel conditions. However, the adaptation is challenging due to the complex relationships among the involved factors. To address the complexity, we apply deep reinforcement learning to learn the optimal adaptation policy when a fog node supports one or more wireless cameras. Extensive evaluation based on trace-driven simulations and experiments show that EFCam complies with the accuracy and latency requirements with lower energy consumption for a real industrial product object tracking application, compared with five baseline approaches incorporating hysteresis-based and event-triggered adaptation.

Multiphase Energetic Experiments: Application of Multiple Object Tracking

The process of using computer vision for multiple-objects tracking is incredibly complex. Thus, simulated data was created to mimic the complexities of more realistic data. These test cases would isolate a few of the inaccuracies of real data and allow the researchers to determine what factor of said data is the most detrimental to the object-tracking process. Due to the large quantity of factors at play, Cotter’s method was used to analyze the significance of each factor. The number of detections and the number of centroids were the main dependent results that were utilized to analyze the data. The overall results show that the most significant detriment of successful object tracking is a lack of data in quantity. Additionally, the results show that if particles move too fast for proper imaging, the resultant data is inaccurate. In the future other methods of particle detection should be explored, as currently Kalman filtering is not a viable option for multiphase energetic experiments.

Actor–Critic Reinforcement Learning and Application in Developing Computer-Vision-Based Interface Tracking

This paper synchronizes control theory with computer vision by formalizing object tracking as a sequential decision-making process. A reinforcement learning (RL) agent successfully tracks an interface between two liquids, which is often a critical variable to track in many chemical, petrochemical, metallurgical, and oil industries. This method utilizes less than 100 images for creating an environment, from which the agent generates its own data without the need for expert knowledge. Unlike supervised learning (SL) methods that rely on a huge number of parameters, this approach requires far fewer parameters, which naturally reduces its maintenance cost. Besides its frugal nature, the agent is robust to environmental uncertainties such as occlusion, intensity changes, and excessive noise. From a closed-loop control context, an interface location-based deviation is chosen as the optimization goal during training. The methodology showcases RL for real-time object-tracking applications in the oil sands industry. Along with a presentation of the interface tracking problem, this paper provides a detailed review of one of the most effective RL methodologies: actor–critic policy.

UAV Object Tracking Application Based on Patch Color Group Feature on Embedded System

The discriminative object tracking system for unmanned aerial vehicles (UAVs) is widely used in numerous applications. While an ample amount of research has been carried out in this domain, implementing a low computational cost algorithm on a UAV onboard embedded system is still challenging. To address this issue, we propose a low computational complexity discriminative object tracking system for UAVs approach using the patch color group feature (PCGF) framework in this work. The tracking object is separated into several non-overlapping local image patches then the features are extracted into the PCGFs, which consist of the Gaussian mixture model (GMM). The object location is calculated by the similar PCGFs comparison from the previous frame and current frame. The background PCGFs of the object are removed by four directions feature scanning and dynamic threshold comparison, which improve the performance accuracy. In the terms of speed execution, the proposed algorithm accomplished 32.5 frames per second (FPS) on the x64 CPU platform without a GPU accelerator and 17 FPS in Raspberry Pi 4. Therefore, this work could be considered as a good solution for achieving a low computational complexity PCGF algorithm on a UAV onboard embedded system to improve flight times.

Under water motion tracking and monitoring using wireless sensor network and Machine learning

Wireless Sensor Networks (WSNs) plays an indispensable role in different application and development in new generation technology. In 5G technology, one of the essential features is the connectivity with registered and unregistered networks to promote remote automation. By which the technology can be harnessed at its best and scaled in multi-dimensional applications. These applications are also enabled for diverse motion detection and object tracking applications. In this presented work, an underwater object tracking technique has been proposed, which enable the prediction of the mobile sensor node. The method involves a simulation which demonstrates a 2D-UASN architecture of underwater scenario and in this network using a different number of nodes using random positioning static sensor nodes are deployed. Finally, a mobile sensor node with random mobility is introduced into the network. The approach follows statistical analysis to compute the next possible position of the node. Based on experiments, the accuracy of the predictive methodology is measured in two different scenarios. In the first scenario, the number of nodes is variable, and the simulation area kept fixed size, i.e., 1000X1000. Additionally, in the second scenario, the number of nodes remains fixed, and the simulation area is variable. The results demonstrate that when we deploy sensor nodes randomly in a large area and network node density is less, the detection accuracy significantly affects and demonstrates low accuracy compared to dense area networks. Secondly, the prediction trend from dense to less dense area demonstrates the reducing accuracy.

Visual object tracking using Fourier domain phase information

In this article, phase of the Fourier transform (FT), which has observed to be a crucial component in image representation, is utilized for visual target tracking. The main aim of the proposed scheme is to reduce the computational complexity of cross-correlation-based matching frameworks. Normalized cross-correlation (NCC) function-based object tracker is converted to a phase minimization problem under the following assumption: In visual object tracking applications, if the frame rate is high, the moving object can be considered to have translational shifts in image domain in a small time window. Since the proposed tracking framework works in the Fourier domain, the translational shifts in the image space are converted to phase variations in the Fourier domain due to the “translational invariance” property of the FT. The proposed algorithm estimates the spatial target position based on the phase information of the target region. The proposed framework uses the \(\ell _1\)-norm and provides a computationally efficient solution for the tracking problem. Experimental studies indicate that the proposed phase-based technique obtain comparable results with baseline tracking algorithms which are computationally more complex.

A Mirror-Based Active Vision System for Underwater Robots: From the Design to Active Object Tracking Application.

A mirror-based active system capable of changing the view’s direction of a pre-existing fixed camera is presented. The aim of this research work is to extend the perceptual tracking capabilities of an underwater robot without altering its structure. The ability to control the view’s direction allows the robot to explore its entire surroundings without any actual displacement, which can be useful for more effective motion planning and for different navigation strategies, such as object tracking and/or obstacle evasion, which are of great importance for natural preservation in environments as complex and fragile as coral reefs. Active vision systems based on mirrors had been used mainly in terrestrial platforms to capture the motion of fast projectiles using high-speed cameras of considerable size and weight, but they had not been used on underwater platforms. In this sense, our approach incorporates a lightweight design adapted to an underwater robot using affordable and easy-access technology (i.e., 3D printing). Our active system consists of two arranged mirrors, one of which remains static in front of the robot’s camera, while the orientation of the second mirror is controlled by two servomotors. Object tracking is performed by using only the pixels contained on the homography of a defined area in the active mirror. HSV color space is used to reduce lighting change effects. Since color and geometry information of the tracking object are previously known, a window filter is applied over the H-channel for color blobs detection, then, noise is filtered and the object’s centroid is estimated. If the object is lost, a Kalman filter is applied to predict its position. Finally, with this information, an image PD controller computes the servomotor articular values. We have carried out experiments in real environments, testing our active vision system in an object-tracking application where an artificial object is manually displaced on the periphery of the robot and the mirror system is automatically reconfigured to keep such object focused by the camera, having satisfactory results in real time for detecting objects of low complexity and in poor lighting conditions.