Real time object tracking based on segmentation and distance minimization
01 Dec 2010-pp 1-4
TL;DR: A novel algorithm for object tracking from Video images based on segmentation and Kernel based procedure, where the target localization problem is minimized using segmentation technique, instead of using mean shift tracking algorithm.
Abstract: In this paper we propose a novel algorithm for object tracking from Video images based on segmentation and Kernel based procedure. Many Kernel based object tracking algorithms have been developed during last few years. The computational complexity becomes very high in those kernel based techniques. In our proposed method the target localization problem is minimized using segmentation technique, instead of using mean shift tracking algorithm. Following segmentation technique the localization problem of target candidate gets minimized, and then comparing the target candidate with the target model by using Bhattacharya coefficient the object can easily be detected. So, the object can be tracked with less computational burden and more efficiently. The proposed algorithm is validated with an existing video sequence and another with a real time video sequence.
Citations
More filters
06 Nov 2020
TL;DR: In this article, a review of the existing approaches followed in monitoring of crowd behavior and the techniques applied to nab absconding suspects, especially in public places like bus stands, railway stations and airports, is presented.
Abstract: Surveillance automation of public places assumes an important role in proactively detection of possible threat to public and in maintaining law and order. Based on a review of the existing approaches followed in monitoring of crowd behavior and the techniques applied to nab absconding suspects, especially in public places like bus stands, railway stations and airports, the paper proposes surveillance automation i.e. automating the process of detecting, recognizing the suspects and suspicious behavior of the people in the crowd. The process involves not only automatically detecting and recognizing known criminals, but also tracking of movements of persons and objects and notifying the authorities of any suspicious behaviour on the basis of machine learning algorithms. The proposed system, that has automated the surveillance process with multiple cameras was found to be working in simulated environment, can prevent unfortunate incidents in public places.
1 citations
04 Jul 2013
TL;DR: This study presents a hybrid algorithm incorporating Particle Swarm Optimization (PSO) and Particle Filter (PF) for multiple-object tracking based mainly on gray-level histogram model that can be maintained by the hybrid tracker using simple histograms model while circumventing the varying-size problem of the objects during the tracking process.
Abstract: This study presents a hybrid algorithm incorporating Particle Swarm Optimization (PSO) and Particle Filter (PF) for multiple-object tracking based mainly on gray-level histogram model. To start with, the hybrid object tracker uses PSO to search the objects in the beginning, taking advantage of the PSO for global optimization. Once the objects have been successfully found by PSO, the hybrid object tracker then switches to PF to continuously track the objects. To avoid the varying-size problem of the objects, Speeded Up Robust Features (SURF) is used to detect the object around its neighborhood in the video sequence for defining the real image size of the object for remodeling the target object by histogram. As a result, tracking speed can be maintained by the hybrid tracker using simple histogram model while circumventing the varying-size problem of the objects during the tracking process.
1 citations
Cites background from "Real time object tracking based on ..."
...There has been a lot of research dealing with single-object tracking problems with both static and moving cameras [10], [16]....
[...]
References
More filters
TL;DR: This paper presents a method for extracting distinctive invariant features from images that can be used to perform reliable matching between different views of an object or scene and can robustly identify objects among clutter and occlusion while achieving near real-time performance.
Abstract: This paper presents a method for extracting distinctive invariant features from images that can be used to perform reliable matching between different views of an object or scene. The features are invariant to image scale and rotation, and are shown to provide robust matching across a substantial range of affine distortion, change in 3D viewpoint, addition of noise, and change in illumination. The features are highly distinctive, in the sense that a single feature can be correctly matched with high probability against a large database of features from many images. This paper also describes an approach to using these features for object recognition. The recognition proceeds by matching individual features to a database of features from known objects using a fast nearest-neighbor algorithm, followed by a Hough transform to identify clusters belonging to a single object, and finally performing verification through least-squares solution for consistent pose parameters. This approach to recognition can robustly identify objects among clutter and occlusion while achieving near real-time performance.
46,906 citations
01 Jan 2011
TL;DR: The Scale-Invariant Feature Transform (or SIFT) algorithm is a highly robust method to extract and consequently match distinctive invariant features from images that can then be used to reliably match objects in diering images.
Abstract: The Scale-Invariant Feature Transform (or SIFT) algorithm is a highly robust method to extract and consequently match distinctive invariant features from images. These features can then be used to reliably match objects in diering images. The algorithm was rst proposed by Lowe [12] and further developed to increase performance resulting in the classic paper [13] that served as foundation for SIFT which has played an important role in robotic and machine vision in the past decade.
14,708 citations
"Real time object tracking based on ..." refers background in this paper
...Objects can be represented in several ways, sometimes objects are represented integrating the color and shape texture feature [1] and tracked, objects can also be represented in forms of points, that is known as scale invariant features [2], some of the tracking has been performed by estimating the background, and subtracting it from the foreground [3]....
[...]
TL;DR: A new approach toward target representation and localization, the central component in visual tracking of nonrigid objects, is proposed, which employs a metric derived from the Bhattacharyya coefficient as similarity measure, and uses the mean shift procedure to perform the optimization.
Abstract: A new approach toward target representation and localization, the central component in visual tracking of nonrigid objects, is proposed. The feature histogram-based target representations are regularized by spatial masking with an isotropic kernel. The masking induces spatially-smooth similarity functions suitable for gradient-based optimization, hence, the target localization problem can be formulated using the basin of attraction of the local maxima. We employ a metric derived from the Bhattacharyya coefficient as similarity measure, and use the mean shift procedure to perform the optimization. In the presented tracking examples, the new method successfully coped with camera motion, partial occlusions, clutter, and target scale variations. Integration with motion filters and data association techniques is also discussed. We describe only a few of the potential applications: exploitation of background information, Kalman tracking using motion models, and face tracking.
4,996 citations
"Real time object tracking based on ..." refers background or methods in this paper
...) be the associates to the pixel at location xiof its bin [4] in the quantized feature space and be the Kronecker delta function, then the probability of the feature u=1....
[...]
...In kernel based object tracking the objects are tracked based on target representation and localization technique [4], adaptive feature selection procedure was also done in kernel based tracking [6], then, spatiogram instead of histogram was considered recently in kernel based approach [7]....
[...]
01 Feb 1988-Graphical Models \/graphical Models and Image Processing \/computer Vision, Graphics, and Image Processing
TL;DR: This paper presents a survey of thresholding techniques and attempts to evaluate the performance of some automatic global thresholding methods using the criterion functions such as uniformity and shape measures.
Abstract: In digital image processing, thresholding is a well-known technique for image segmentation. Because of its wide applicability to other areas of the digital image processing, quite a number of thresholding methods have been proposed over the years. In this paper, we present a survey of thresholding techniques and update the earlier survey work by Weszka (Comput. Vision Graphics & Image Process 7, 1978 , 259–265) and Fu and Mu (Pattern Recognit. 13, 1981 , 3–16). We attempt to evaluate the performance of some automatic global thresholding methods using the criterion functions such as uniformity and shape measures. The evaluation is based on some real world images.
2,771 citations
"Real time object tracking based on ..." refers background in this paper
...In case of segmentation based approach, [5] the object has to be solely taken out from the background, which can’t be possible in many cases....
[...]
20 Jun 2005
TL;DR: This work shows how to use spatiograms in kernel-based trackers, deriving a mean shift procedure in which individual pixels vote not only for the amount of shift but also for its direction, and shows improved tracking results compared with histograms.
Abstract: We introduce the concept of a spatiogram, which is a generalization of a histogram that includes potentially higher order moments. A histogram is a zeroth-order spatiogram, while second-order spatiograms contain spatial means and covariances for each histogram bin. This spatial information still allows quite general transformations, as in a histogram, but captures a richer description of the target to increase robustness in tracking. We show how to use spatiograms in kernel-based trackers, deriving a mean shift procedure in which individual pixels vote not only for the amount of shift but also for its direction. Experiments show improved tracking results compared with histograms, using both mean shift and exhaustive local search.
460 citations
"Real time object tracking based on ..." refers methods in this paper
...In kernel based object tracking the objects are tracked based on target representation and localization technique [4], adaptive feature selection procedure was also done in kernel based tracking [6], then, spatiogram instead of histogram was considered recently in kernel based approach [7]....
[...]