Robust Visual Tracking in Cluttered Environment Using an Active Contour Method

Robust visual tracking is the important stage in the computer vision applications such as robotics, man-free control systems, and the visual surveillance. Accurate motion states estimation and the target representation in visual tracking system are based on the appearances of the target. The factor affects the learning of target representation is the accumulated error due to the pose, illumination changes, and the uneven background. The presence of dynamic background and the shadowing effects causes the visual drift and destructive information. Besides, the misclassification of target region induces the false detection of moving objects. The K-means and Fuzzy-C-means clustering algorithms are available to segment the foreground/background and suppress the shadow region on the basis of the non-changing background of the surveillance area. This paper proposes the novel background normalization technique with textural pattern analysis to suppress the shadow region. The Neighborhood Chain Prediction (NCP) algorithm is used to cluster the uneven background and the Differential Boundary Pattern (DBP) extracts the texture of the video frame to suppress the shadow pixels present in the frame. The lower intensity estimation and the prediction of the area around the lower intensity in proposed work enhance the pixels for shadow removal. The shadow-free frame split up into several grids and the histograms of features are extracted from the grid formatted frame. Finally, the Machine Level Classification (MLC) finds the matching grid corresponds to the tracking region and provides the binary labeling to separate the background and foreground. The proposed DBP-based visual tracking system is high robustness over the sudden illumination changes and the dynamic background due to the texture pattern analysis. The comparison of proposed NCP-DBP combination with the existing segmentation techniques regarding the accuracy, precision, recall, F-measure, success and error rate assured the effectiveness in visual tracking applications.

The goal of our project is to devise a robust method to track and compensate patient motion by combining an emission data based approach with a visual tracking system (VTS) that provides an independent estimate of motion. In a previous study, we used the Polaris infra-red system (NDI Inc., Ontario, Canada) as the gold standard to show that the VTS can accurately track motion without having the limitations of the Polaris. In the present work, we present the latest hardware configuration of the VTS and our solution for temporal synchronization between the SPECT and the optical acquisitions. The current version of the VTS includes stereo imaging with sets of optical network cameras with lighting, a SPECT/VTS calibration phantom, a black stretchable garment with reflective spheres to track chest motion, and a computer to control the cameras. The computer also stores the JPEG files generated by the optical cameras with timing synchronization to the list-mode acquisition of events on our SPECT system. Five Axis PTZ 2130 network cameras (Axis Communications AB, Lund, Sweden) were used to track motion of spheres with a highly retro-reflective coating using stereo methods. The calibration phantom is comprised of seven reflective spheres, and radioactivity can be added to the tip of the mounts holding the spheres. This phantom is used to determine the transformation to be applied to convert the motion detected by the VTS into the SPECT coordinates system. Synchronization is assessed in two ways. First, optical cameras are aimed at a digital clock and the elapsed time estimated by the cameras is compared to the actual time shown by the clock in the images. Synchronization is also assessed by moving a radioactive and reflective sphere three times during concurrent VTS and SPECT acquisitions and comparing the time at which motion occurred in the optical and SPECT images. The results show that optical and SPECT images stay synchronized within a 150 ms range. The 100Mbit network load is less than 10%, and the computer's CPU load is between 15 and 25%; thus, the VTS can be improved by adding more cameras or by increasing the image size and/or resolution while keeping an acquisition rate of 30 images per second per camera

In this paper we study a new estimation method for the maximum production rate in a high quality relay production system and an optimum operation method yieling the production rate. A relay is generally composed of two main parts (Base and Armature) with machining errors occurred in machining processes, so a 'Matching method' must be applied to the system to produce high quality relays. In the 'Matching method' one part is reprocessed using a tool selected according to an assembling error to reduce the assembling error, but the correct tool can not always be selected because the assembling error, which is calculated by machining errors, contains measuring errors. Measuring errors are occurred in measuring processes of machining errors. At first, we propose the estimation method for the maximum production rate and the optimum operation method under two conditions that tools are already set and that tools are not set yet. Next, the effects of machining and measuring accuracies on the maximum production rate and the optimum operation method are investigated. Results show that the proposed method can estimate the maximum production rate and the optimum operation method and that the machining and measuring accuracies both affect the production rate and the operation method.

Object tracking is an important technology in video surveillance. The main approach is Mean Shift algorithm and its improved version. Studies show that the traditional Mean Shift algorithm adopts a fixed searching window in the tracking process, which cannot adjust the template adaptively. The improved algorithm, CamShift, overcomes this problem with an adaptively changing searching window. However, these algorithms are both based on color tracking, which requires that the colors of the foreground targets are unique. If the color of the target is similar to the color of the background, tracking errors will occur or tracking targets will be lost. In this study, we developed an adaptive gradient enhanced texture based tracking algorithm for traffic monitoring applications. This algorithm combines the characteristics of the color and texture of objects. The algorithm builds a joint histogram template of color and texture for targeting, which solves the problems of tracking targets losing when the color of the object is similar to the color of the background. The experiments show that the algorithm can improve the accuracy and robustness of object tracking.

In the classical kernel correlated target tracking algorithms, most of them initialize the target detector by importing the manually marked targets. Moreover, the anti-occlusion ability of the correlated filtering algorithm is poor, which leads to the model drift in complex scenes. This work combined the image matching technology with the target tracking technology. In the case of a certain similarity between the input image and the target, the improved SIFT algorithm SIFT-BRISK is used to complete the initialization of the target detector, which can realize the target tracking in more common cases. And then APCE confidence level is introduced to judge whether the target tracking performance is well, adapt the update strategy of keeping track in the high confidence, dynamically adjusting the tracking model in the low confidence, and improve the details of the kernel correlated target tracking algorithm. Thus, the drift of the training model is suppressed to a certain extent. Finally, the algorithm is tested under the OTB-2015 dataset and the result proves that the improved algorithm has better accuracy and higher speed than traditional target tracking algorithms.

Recovering people contours from partial occlusion is a challenging problem in a visual tracking system. Partial occlusions would bring about unreasonable contour changes of the target object. In this paper, a novel method is presented to detect partial occlusion on people contours and recover occluded portions. Unlike other occlusion detection methods, the proposed method is only based on contours, which makes itself more flexible to be extended for further applications. Experiments with synthetic images demonstrate the accuracy of the method for detecting partial occlusions, and experiments on real-world video sequence are also carried out to prove that the method is also good enough to be used to recover target contours.

Here, the authors propose a novel tracking algorithm that can automatically modify the initial configuration of a target to improve the tracking accuracy in subsequent frames. To achieve this goal, the authors’ method analyses the likelihood landscape (LL) for the image patch described by the initial configuration. A good configuration has a unimodal distribution with a steep shape in the LL. Using the LL analysis, the authors’ method improves the initial configuration, resulting in more accurate tracking results. The authors improve the conventional LL analysis based on two ideas. First, the authors’ method analyses the LL in the RGB space rather than the grey space. Second, the method introduces an additional criterion for a good configuration: a high likelihood value at the mode. The authors further enhance their method through post‐processing of the visual tracking results at each frame, where the estimated bounding boxes are modified by the LL analysis. The experimental results demonstrate that the authors’ advanced LL analysis helps improve the tracking accuracy of several baseline trackers on a visual tracking benchmark data set. In addition, the authors’ simple post‐processing technique significantly enhances the visual tracking performance in terms of precision and success rate.

Respiratory motion induces artifacts in reconstructed cardiac perfusion SPECT images. Correction for respiratory motion often relies on a respiratory signal describing the heart displacements during breathing. However, using external tracking devices to estimate respiratory signals can add cost and operational complications in a clinical setting. We aim to develop a deep learning (DL) approach that uses only SPECT projection data for respiratory signal estimation. A modified U-Net was implemented that takes temporally finely sampled SPECT sub-projection data (100ms) as input. These sub-projections are obtained by reframing the 20-s list-mode data, resulting in 200 sub-projections, at each projection angle for each SPECT camera head. The network outputs a 200-time-point motion signal for each projection angle, which was later aggregated over all angles to give a full respiratory signal. The target signal for DL model training was from an external stereo-camera visual tracking system (VTS). In addition to comparing DL and VTS, we also included a data-driven approach based on the center-of-mass (CoM) strategy. This CoM method estimates respiratory signals by monitoring the axial changes of CoM for counts in the heart region of the sub-projections. We utilized 900 subjects with stress cardiac perfusion SPECT studies, with 302 subjects for testing and the remaining 598 subjects for training and validation. The Pearson's correlation coefficient between the DL respiratory signal and the reference VTS signal was 0.90, compared to 0.70 between the CoM signal and the reference. For respiratory motion correction on SPECT images, all VTS, DL, and CoM approaches partially de-blured the heart wall, resulting in a thinner wall thickness and increased recovered maximal image intensity within the wall, with VTS reducing blurring the most followed by the DL approach. Uptake quantification for the combined anterior and inferior segments of polar maps showed a mean absolute difference from the reference VTS of 1.7% for the DL method for patients with motion >12mm, compared to 2.6% for the CoM method and 8.5% for no correction. We demonstrate the capability of a DL approach to estimate respiratory signal from SPECT projection data for cardiac perfusion imaging. Our results show that the DL based respiratory motion correction reduces artefacts and achieves similar regional quantification to that obtained using the stereo-camera VTS signals. This may enable fully automatic data-driven respiratory motion correction without relying on external motion tracking devices.

Patient motion is inevitable in SPECT and PET due to the lengthy period of time patients are imaged and patient motion can degrade diagnostic accuracy. The goal of our studies is to perfect a methodology for tracking and correcting patient motion when it occurs. In this paper we report on enhancements to the calibration, camera stability, accuracy of motion tracking, and temporal synchronization of a low-cost visual tracking system (VTS) we are developing. The purpose of the VTS is to track the motion of retro-reflective markers on stretchy bands wrapped about the chest and abdomen of patients. We have improved the accuracy of 3D spatial calibration by using a MATLAB optical camera calibration package with a planar calibration pattern. This allowed us to determine the intrinsic and extrinsic parameters for stereo-imaging with our CCD cameras. Locations in the VTS coordinate system are transformed to the SPECT coordinate system by a VTS/SPECT mapping using a phantom of 7 retro-reflective spheres each filled with a drop of Tc(99m). We switched from pan, tilt and zoom (PTZ) network cameras to fixed network cameras to reduce the amount of camera drift. The improved stability was verified by tracking the positions of fixed retro-reflective markers on a wall. The ability of our VTS to track movement, on average, with sub-millimeter and sub-degree accuracy was established with the 7-sphere phantom for 1 cm vertical and axial steps as well as for an arbitrary rotation and translation. The difference in the time of optical image acquisition as decoded from the image headers relative to synchronization signals sent to the SPECT system was used to establish temporal synchrony between optical and list-mode SPECT acquisition. Two experiments showed better than 100 ms agreement between VTS and SPECT observed motion for three axial translations. We were able to track 3 reflective markers on an anthropomorphic phantom with a precision that allowed us to correct motion such that no loss in visual quality was noted in motion corrected slices relative to motion free slices.

Adversarial attacks on visual object tracking aim to degrade tracking accuracy by introducing imperceptible perturbations into video frames, exploiting vulnerabilities in neural networks. In real-world symmetrical double-blind engagements, both attackers and defenders operate with mutual unawareness of strategic parameters or initiation timing. Black-box attacks based on iterative optimization show excellent applicability in this scenario. However, existing state-of-the-art adversarial attacks based on iterative optimization suffer from high computational costs and limited effectiveness. To address these challenges, this paper proposes the Universal Low-frequency Noise black-box attack method (ULN), which generates perturbations through discrete cosine transform to disrupt structural features critical for tracking while mimicking compression artifacts. Extensive experimentation on four state-of-the-art trackers, including transformer-based models, demonstrates the method’s severe degradation effects. GRM’s expected average overlap drops by 97.77% on VOT2018, while SiamRPN++’s AUC and Precision on OTB100 decline by 76.55% and 78.9%, respectively. The attack achieves real-time performance with a computational cost reduction of over 50% compared to iterative methods, operating efficiently on embedded devices such as Raspberry Pi 4B. By maintaining a structural similarity index measure above 0.84, the perturbations blend seamlessly with common compression artifacts, evading traditional spatial filtering defenses. Cross-platform experiments validate its consistent threat across diverse hardware environments, with attack success rates exceeding 40% even under resource constraints. These results underscore the dual capability of ULN as both a stealthy and practical attack vector, and emphasize the urgent need for robust defenses in safety-critical applications such as autonomous driving and aerial surveillance. The efficiency of the method, when combined with its ability to exploit low-frequency vulnerabilities across architectures, establishes a new benchmark for adversarial robustness in visual tracking systems.

In this paper, a visual tracking system is proposed. The whole system is divided into two parts. The first part is the detection of the target and the second part is the trajectory filtering. To detect the target in the image, we use the method based on the distribution of edges. In this method, the template is edge based, and the distribution of edges of the image is extracted, so that the two are convolved to find the location of the target. The method is more reliable than the traditional template matching in the complicated scene. After finding the target in the image, we will control the camera platform to keep the target at the center of the image. The location of the target in the world coordinate frame is first transformed to the orientation relative to the camera whereby a tracking algorithm based on the IMM (interacting multiple model) algorithm is developed. The algorithm estimates the angular velocity and the angular acceleration of the target relative to the camera, which along with some error information in the images are used for control of the camera platform. Finally, a number of experiments are conducted to validate the developed visual tracker.

Target detection and tracking is a key technology in intelligent monitoring system. However, it is still difficult to design a robust, accurate and real-time target detection and tracking algorithm. Taking target Tracking as the research focus, this paper focuses on one of the current mainstream algorithms, namely, TLD (tracking-learning-detection) target Tracking algorithm. Based on the theory of TLD target tracking and corner detection, a new target tracking method was proposed to solve the existing problems. A classifier is built online to track and learn the target in real time, and a target contour detector is added in the system to complement the random forest detector to solve the problem of tracking failure. Experimental results show that the proposed algorithm can effectively overcome the influence of fast change or depth occlusion, and improve the tracking efficiency of moving targets.

An image-based visual seam tracking system is designed for butt weld of thin plate in this paper. The torch deviation from the joint in the horizontal direction is measured from the seam image. Since the weld seam is so narrow, its image is based on natural lighting. Due to the hardness of measuring depth information using monocular vision, structured laser light is used for detecting the torch deviation in the vertical direction. For steady seam tracking, a robust and fast feature extraction algorithm is developed. Moreover, a reliable controller is designed with error filtering and output pulse verification. Finally, experiments are conducted to verify the proposed seam tracking system.

Convolutional neural networks are potent models that yield hierarchies of features and have drawn increasing interest in the visual tracking field. In the paper, we design an end-to-end trainable tracking framework based on Siamese network, which proposes to learn the low-level fine-grained and high-level semantic representations simultaneously with the aim of mutual benefit. Due to the distinct and complementary characteristics of the feature hierarchies, different tracking mechanisms are adopted for different feature layers. The low-level features are exploited and updated with a correlation filter layer for adaptive tracking and the high-level features are compared through cross-correlation directly for robust tracking. The two-level features are jointly trained with a multi-task loss function end-to-end. The proposed tracker takes full advantage of the adaptability of the low-level features and the generalization ability of the high-level features. Extensive experimental tracking results on the widely used OTB and TC128 benchmarks demonstrate the superiority of our tracker. Meanwhile, our proposed tracker can achieve a real-time tracking speed.

Intelligent video surveillance is currently one of the most active research topics in computer vision, especially when facing the explosion of video data captured by a large number of surveillance cameras. As a key step of an intelligent surveillance system, robust visual tracking is very challenging for computer vision. However, it is a basic functionality of the human visual system (HVS). Psychophysical findings have shown that the receptive fields of simple cells in the visual cortex can be characterized as being spatially localized, oriented, and bandpass, and it forms a sparse, distributed representation of natural images. In this article, motivated by these findings, we propose an effective appearance model based on sparse coding and apply it in visual tracking. Specifically, we consider the responses of general basis functions extracted by independent component analysis on a large set of natural image patches as features and model the appearance of the tracked target as the probability distribution of these features. In order to make the tracker more robust to partial occlusion, camouflage environments, pose changes, and illumination changes, we further select features that are related to the target based on an entropy-gain criterion and ignore those that are not. The target is finally represented by the probability distribution of those related features. The target search is performed by minimizing the Matusita distance between the distributions of the target model and a candidate using Newton-style iterations. The experimental results validate that the proposed method is more robust and effective than three state-of-the-art methods.