Image Object Tracking Research Articles

Real-Time Tracking Target System Based on Kernelized Correlation Filter in Complicated Areas.

The achievement of rapid and reliable image object tracking has long been crucial and challenging for the advancement of image-guided technology. This study investigates real-time object tracking by offering an image target based on nuclear correlation tracking and detection methods to address the challenge of real-time target tracking in complicated environments. In the tracking process, the nuclear-related tracking algorithm can effectively balance the tracking performance and running speed. However, the target tracking process also faces challenges such as model drift, the inability to handle target scale transformation, and target length. In order to propose a solution, this work is organized around the following main points: this study dedicates its first part to the research on kernelized correlation filters (KCFs), encompassing model training, object identification, and a dense sampling strategy based on a circulant matrix. This work developed a scale pyramid searching approach to address the shortcoming that a KCF cannot forecast the target scale. The tracker was expanded in two stages: the first stage output the target's two-dimensional coordinate location, and the second stage created the scale pyramid to identify the optimal target scale. Experiments show that this approach is capable of resolving the target size variation problem. The second part improved the KCF in two ways to meet the demands of a long-term object tracking task. This article introduces the initial object model, which effectively suppresses model drift. Secondly, an object detection module is implemented, and if the tracking module fails, the algorithm is redirected to the object detection module. The target detection module utilizes two detectors, a variance classifier and a KCF. Finally, this work includes trials on object tracking experiments and subsequent analysis of the results. Initially, this research provides a tracking algorithm assessment system, including an assessment methodology and the collection of test videos, which helped us to determine that the suggested technique outperforms the KCF tracking method. Additionally, the implementation of an evaluation system allows for an objective comparison of the proposed algorithm with other prominent tracking methods. We found that the suggested method outperforms others in terms of its accuracy and resilience.

OBJECT TRACKING CONTROL USING A GIMBAL MECHANISM

Abstract. This work describes a control solution for real time object tracking in images acquired for a RPAS on an object inspection environment. This, controlling a 3-axis gimbal mechanism to control a camera orientation embedded to a RPAS, using its image processed for feedback. The objective of control is to maintain the target of interest at the center of the image plane. The proposed solution uses a YOLOv3 object detection model in order to detect the target object and determine, thru rotation matrices, the new desired angles to converge the object’s position to the center of the image. To compare results of the proposed control, a linear control was tuned using a linear PI algorithm. Simulation and practice experiments successfully tracked the desired object in real time using YOLOv3 in both control approaches presented.

Implementation of Offline Consumer Behavior Tracking

Due to declining trading volume growth in e-commerce platforms, physical channels have attracted considerable investments from various large international companies (e.g. Alibaba, JD, Walmart, Wanda, and Wuzhou International). However, e-commerce platforms can track consumers’ behaviors (attraction to landing page design, clicks on certain products, consumer behavior trajectory tracking, clicks on advertisements, and internal link optimization of product pages), a feat unachievable in current physical channels. Consequently, this study attempted to apply the characteristics of online channels in a physical channel by using image object tracking and image detection techniques. Through this inclusion, physical channels are capable of providing consumers with more favorable experience and interaction, and brick-and-mortar store owners can obtain a more accurate understanding of consumer behaviors of store consumers. Information acquired through this system can be provided to store owners to serve as reference for merchandise placement, arrangement of display shelves, and consumer circulation path planning. This study used the technique of image processing to locate the Region of Interest and applied object tracking to get the consumer’s trajectory which successfully implemented the consumer-tracking characteristics of online platforms in a physical channel while retaining the unique experience of the physical channel. This results in a win–win scenario for businesses and consumers.

A high-speed tracking algorithm for dense granular media

Many fields of study, including medical imaging, granular physics, colloidal physics, and active matter, require the precise identification and tracking of particle-like objects in images. While many algorithms exist to track particles in diffuse conditions, these often perform poorly when particles are densely packed together—as in, for example, solid-like systems of granular materials. Incorrect particle identification can have significant effects on the calculation of physical quantities, which makes the development of more precise and faster tracking algorithms a worthwhile endeavor. In this work, we present a new tracking algorithm to identify particles in dense systems that is both highly accurate and fast. We demonstrate the efficacy of our approach by analyzing images of dense, solid-state granular media, where we achieve an identification error of 5% in the worst evaluated cases. Going further, we propose a parallelization strategy for our algorithm using a GPU, which results in a speedup of up to 10× when compared to a sequential CPU implementation in C and up to 40× when compared to the reference MATLAB library widely used for particle tracking. Our results extend the capabilities of state-of-the-art particle tracking methods by allowing fast, high-fidelity detection in dense media at high resolutions.

A Multiple Random Feature Extraction Algorithm for Image Object Tracking

This paper proposes an object-tracking algorithm with multiple randomly-generated features. We mainly improve the tracking performance which is sometimes good and sometimes bad in compressive tracking. In compressive tracking, the image features are generated by random projection. The resulting image features are affected by the random numbers so that the results of each execution are different. If the obvious features of the target are not captured, the tracker is likely to fail. Therefore the tracking results are inconsistent for each execution. The proposed algorithm uses a number of different image features to track, and chooses the best tracking result by measuring the similarity with the target model. It reduces the chances to determine the target location by the poor image features. In this paper, we use the Bhattacharyya coefficient to choose the best tracking result. The experimental results show that the proposed tracking algorithm can greatly reduce the tracking errors. The best performance improvements in terms of center location error, bounding box overlap ratio and success rate are from 63.62 pixels to 15.45 pixels, from 31.75% to 64.48% and from 38.51% to 82.58%, respectively.

An object tracking algorithm based on optical flow and temporal–spatial context

Image object tracking, as one of the hot spots in computer vision, has made great progress recently. Nevertheless, there has been no algorithm that could show good robustness against all kinds of challenging video scenes. The tracking algorithm of temporal–spatial context effectively took advantage of the information contained in the background and the appearance of the object. By adopting this algorithm, good tracking effects has been achieved. However, such algorithm could easily lead to tracking failure in case of the object moving too fast or the object location changing too much. With Harris corner point adopted as the feature point, this paper corrected the tracking result of the STC tracking algorithm by using the L–K optical flow method as an auxiliary technique. Consequently, better tracking effects were achieved under the premise of preserving the excellent performance of the STC algorithm.

Image object tracking based on temporal context and MOSSE

Designing an image target tracking algorithm which is suitable for all occasions is a hotspot in visual field. The MOSSE algorithm based on correlation filter achieves good tracking effect, but it has the disadvantage of poor anti-drift ability. Based on the MOSSE algorithm, multi-frame historical images are used as the input samples of AdaBoost, the classification effect of the weak classifier is measured by the response to the peak coordinate distance, the weights of the training samples are updated according to the classification effect, and multiple weak classifiers and then weighed the weak classifiers according to the accuracy of the tracking target, then we get the final strong filter. The algorithm makes full use of the historical appearance information of the target, which can improve the robustness of the system effectively, while still maintaining the real-time of the related filter algorithm.

Video Image Object Tracking Algorithm based on Improved Principal Component Analysis

Since the existing object tracking algorithms are very difficult to adapt the object appearance changes caused by illumination changes, large pose variations, and partial or full occlusions, an object tracking algorithm based on two-dimensional principal component analysis (2DPCA) and sparse-representation is proposed in this paper. The tracking algorithm adopts 2DPCA and sparse-representation to establish object appearance model. In order to reduce dimension of object template, incremental subspace updating algorithm is introduced to online update the object template, reduce the requirement of memory space and enhance the precision of object appearance description. Experimental results show the proposed algorithm is robust for image illumination variance and object partial occlusion.

Pose determination of parameterized object models from a monocular image

Abstract Pose determination of parameterized object models plays an important role in verification of model based recognition systems and real-time tracking of objects in images. Described herein is a data structure that models 3D parametric objects. Algorithms are presented for obtaining analytical partial derivatives of distance functions from projected model edges and endpoints to corresponding image segments, with respect to changes in model and camera parameters. These parameters include rotational, translational, scale and dilation in camera and object model. Solving for camera and model parameters in a 2D image (pose determination) is a nonlinear least squares paradigm. Weights are considered for line-to-line and point-to-point matchings to allow for the applicability of the methods to images with noisy data. A number of complex models are tested, and convergence to proper values occurs from a wide range of initial error, using advanced numerical analysis techniques. The experimentation suggests that by treating the model and camera parameters uniformly, correct pose determination can be obtained from minimal matching information relative to the number of matches required per component individually. The robust verification of model parameters thus simplifies associated object recognition problems.