Performance evaluation of object detection and tracking method under illumination variation

doi:10.1109/INDICON.2014.7030369

Home
/
Papers
/
Performance evaluation of object detection and tracking method under illumination variation

Proceedings Article•DOI•

Performance evaluation of object detection and tracking method under illumination variation

Naimish Kasundra¹, Krishna K. Warhade¹•Institutions (1)

College of Engineering, Pune¹

01 Dec 2014-pp 1-6

TL;DR: Experiments shows that proposed discrete wavelet transform based method has a high capability to detect and track non-rigid moving object, even when light intensities change abruptly.

read less

Abstract: A robust, meticulous and high performance approach is still a great challenge in tracking approach. There are various difficulties in object tracking like noise in scene, illumination changes, occlusion effect, and pose variation into the scene. As an object moves, it changes its orientation relative to the light sources which illuminate it. An illumination variation causes tracking algorithm to lose the target in the scene. This paper presents a discrete wavelet transform based method of detecting and tracking moving object under varying illumination condition with a stationary camera. Discrete wavelet transform provides illumination invariant feature extraction method using gaussian smoothing function and thresholding. We have tested tracking results, on number of video sequences with indoor and outdoor environments and demonstrated the effectiveness of our proposed method. Experiments shows that proposed method has a high capability to detect and track non-rigid moving object, even when light intensities change abruptly.

...read moreread less

Citations

PDF

Open Access

More filters

Journal Article•DOI•

Development of deep learning-based equipment heat load detection for energy demand estimation and investigation of the impact of illumination

[...]

Shuangyu Wei¹, John Kaiser Calautit¹•Institutions (1)

University of Nottingham¹

01 Apr 2021-International Journal of Energy Research

9 citations

Cites background from "Performance evaluation of object de..."

...On the other hand, the change of the position of targets is another factor influencing the lighting effects as it changes the orientation of targets relative to the light sources which illuminate them.(36,37) For instance, if a detector is performed to recognize computer screens using a fixed camera, the change of the position, height, or angle of computer screens cause the variations of shape, size, brightness for the camera, which may result in more detection errors....
[...]
...Hence, varying illumination often causes issues since the observed color and the type of any object, which is being presented, may be perceived to be different.(35,36) On the other hand, the change of the position of targets is another factor influencing the lighting effects as it changes the orientation of targets relative to the light sources which illuminate them....
[...]

Proceedings Article•DOI•

Research on Vehicle Detection and Tracking Algorithm Based on the Methods of Frame Difference and Adaptive Background Subtraction Difference

[...]

Yiqin Cao¹, Xiao Yun, Tao Zhong¹, Xiaosheng Huang¹•Institutions (1)

East China Jiaotong University¹

20 Nov 2016

TL;DR: Experimental result shows that the improving algorithm can extract all moving objects, which was endowed with strong background adaptability and better real-time performance.

...read moreread less

Abstract: This paper proposed methods of vehicle detection and tracking algorithm in real-time traffic. In the detection of realtime moving vehicle, vehicle areas would be determined through road line detection. Then, the main color information of moving and non-moving area would be obtained through frame difference. Filling the main color information in vehicle moving area would lead to a similar background image. At last, moving vehicles would be determined through adaptive Background Subtraction difference. In the tracking of moving vehicles, firstly, all characteristic corners can be got by using Harris detection. Then, all characteristic corner set in the separate moving area would be collected through cluster analysis. For each characteristic individual corner set can generate a circle embracing all characteristics, some problems like vehicle barrier could be analyzed by using the radius of characteristic circle. At last, conduct feature matching tracking by using the center of feature circle. Experimental result shows that the improving algorithm can extract all moving objects, which was endowed with strong background adaptability and better real-time performance. Keywords-target detection; frame difference method; background subtraction difference method; harris corner detection; clustering analysis

...read moreread less

2 citations

Cites methods from "Performance evaluation of object de..."

...At present, the major approaches to detect moving vehicles are inter-frame difference method, Background Subtraction difference method, optical flow method and etc [3-4]....
[...]
...Background Subtraction difference method[3-4] requires first setting up the Gaussian background model as the background image, then calculating the difference of pixel brightness between current video sequence image and the known background image and taking the absolute value....
[...]

Proceedings Article•DOI•

Evaluation of the detection ability of a multi-sensor system under weak light

[...]

Zheng An Qiu, Yingce Wang, Changyu Shen, Xianglian Liu, Yangting Fu, Yingwei He, Haiyong Gan - Show less +3 more

27 Mar 2022

TL;DR: An evaluation method that uses the response characteristics of lidar, depth camera, and RGB camera to the surrounding environment, takes the surface illuminance and BRDF of the target as input variables, and uses the minimum distance classification method to realize the lateral evaluation and comparison of different sensors is proposed.

...read moreread less

Abstract: Target detection plays an important role in the field of autonomous driving. During the driving of the vehicle, the detection system is susceptible to the low illumination of the road at night, resulting in the inability to detect the target. Therefore, night detection becomes a difficult problem in target detection. With the continuous development of sensor types and related algorithms used in night detection systems, how to evaluate the detection capabilities of night detection systems has become an urgent problem to be solved. To solve this problem, an evaluation method for evaluating multi-sensor target detection systems is proposed. It is used to evaluate the target detection ability of the target detection system in the night road environment. This method uses the response characteristics of lidar, depth camera, and RGB camera to the surrounding environment, takes the surface illuminance and BRDF of the target as input variables, and uses the minimum distance classification method to realize the lateral evaluation and comparison of different sensors. And it is obtained that the BRDF limit of the detected samples is 0.057127 sr-1 under the 0.51lx illumination and the BRDF difference limit of the distinguished samples is 0.01953 sr-1 under the 0.98lx illumination of the RGB camera. Finally, the effectiveness of the evaluation method is proved by the test of the response signal of a single sensor.

...read moreread less

References

PDF

Open Access

More filters

Journal Article•DOI•

Object tracking: A survey

[...]

Alper Yilmaz¹, Omar Javed, Mubarak Shah²•Institutions (2)

Ohio State University¹, University of Central Florida²

25 Dec 2006-ACM Computing Surveys

TL;DR: The goal of this article is to review the state-of-the-art tracking methods, classify them into different categories, and identify new trends to discuss the important issues related to tracking including the use of appropriate image features, selection of motion models, and detection of objects.

...read moreread less

Abstract: The goal of this article is to review the state-of-the-art tracking methods, classify them into different categories, and identify new trends. Object tracking, in general, is a challenging problem. Difficulties in tracking objects can arise due to abrupt object motion, changing appearance patterns of both the object and the scene, nonrigid object structures, object-to-object and object-to-scene occlusions, and camera motion. Tracking is usually performed in the context of higher-level applications that require the location and/or shape of the object in every frame. Typically, assumptions are made to constrain the tracking problem in the context of a particular application. In this survey, we categorize the tracking methods on the basis of the object and motion representations used, provide detailed descriptions of representative methods in each category, and examine their pros and cons. Moreover, we discuss the important issues related to tracking including the use of appropriate image features, selection of motion models, and detection of objects.

...read moreread less

5,318 citations

"Performance evaluation of object de..." refers background in this paper

...Moving object tracking and detection is an important research area for wide spread application in diverse disciplines like the visual surveillance, human computer interaction, driving assistance system, image stabilization for digital cameras, security surveillance, autonomous navigation, traffic flow analysis and so on [1]....
[...]

Journal Article•DOI•

Mean shift, mode seeking, and clustering

[...]

Yizong Cheng¹•Institutions (1)

University of Cincinnati¹

01 Aug 1995-IEEE Transactions on Pattern Analysis and Machine Intelligence

TL;DR: Mean shift, a simple interactive procedure that shifts each data point to the average of data points in its neighborhood is generalized and analyzed and makes some k-means like clustering algorithms its special cases.

...read moreread less

Abstract: Mean shift, a simple interactive procedure that shifts each data point to the average of data points in its neighborhood is generalized and analyzed in the paper. This generalization makes some k-means like clustering algorithms its special cases. It is shown that mean shift is a mode-seeking process on the surface constructed with a "shadow" kernal. For Gaussian kernels, mean shift is a gradient mapping. Convergence is studied for mean shift iterations. Cluster analysis if treated as a deterministic problem of finding a fixed point of mean shift that characterizes the data. Applications in clustering and Hough transform are demonstrated. Mean shift is also considered as an evolutionary strategy that performs multistart global optimization. >

...read moreread less

3,924 citations

"Performance evaluation of object de..." refers methods in this paper

...It can be generated like clustering algorithms [9] and was originally developed for data clustering algorithm....
[...]

Journal Article•DOI•

Alignment by Maximization of Mutual Information

[...]

Paul A. Viola¹, William M. Wells¹•Institutions (1)

Massachusetts Institute of Technology¹

01 Sep 1997-International Journal of Computer Vision

TL;DR: A new information-theoretic approach is presented for finding the pose of an object in an image that works well in domains where edge or gradient-magnitude based methods have difficulty, yet it is more robust than traditional correlation.

...read moreread less

Abstract: A new information-theoretic approach is presented for finding the pose of an object in an image. The technique does not require information about the surface properties of the object, besides its shape, and is robust with respect to variations of illumination. In our derivation few assumptions are made about the nature of the imaging process. As a result the algorithms are quite general and may foreseeably be used in a wide variety of imaging situations. Experiments are presented that demonstrate the approach registering magnetic resonance (MR) images, aligning a complex 3D object model to real scenes including clutter and occlusion, tracking a human head in a video sequence and aligning a view-based 2D object model to real images. The method is based on a formulation of the mutual information between the model and the image. As applied here the technique is intensity-based, rather than feature-based. It works well in domains where edge or gradient-magnitude based methods have difficulty, yet it is more robust than traditional correlation. Additionally, it has an efficient implementation that is based on stochastic approximation.

...read moreread less

3,584 citations

"Performance evaluation of object de..." refers background or methods in this paper

...Also the kullback divergence [14] has been used to find the pose of an object in an image....
[...]
...First the definition of kernels is given and then the generalized mean shift algorithm [14] is discussed....
[...]

Proceedings Article•DOI•

Real-time tracking of non-rigid objects using mean shift

[...]

Dorin Comaniciu¹, Visvanathan Ramesh¹, Peter Meer•Institutions (1)

Princeton University¹

14 Feb 2000

TL;DR: The theoretical analysis of the approach shows that it relates to the Bayesian framework while providing a practical, fast and efficient solution for real time tracking of non-rigid objects seen from a moving camera.

...read moreread less

Abstract: A new method for real time tracking of non-rigid objects seen from a moving camera is proposed. The central computational module is based on the mean shift iterations and finds the most probable target position in the current frame. The dissimilarity between the target model (its color distribution) and the target candidates is expressed by a metric derived from the Bhattacharyya coefficient. The theoretical analysis of the approach shows that it relates to the Bayesian framework while providing a practical, fast and efficient solution. The capability of the tracker to handle in real time partial occlusions, significant clutter, and target scale variations, is demonstrated for several image sequences.

...read moreread less

3,368 citations

"Performance evaluation of object de..." refers methods in this paper

...The results obtained by our proposed methods are compared with the mean shift algorithm [12] which is presented in the Section-II....
[...]
...It is observed that proposed method increase the detection rate and performs slightly better than the mean shift algorithm [12] in most of the video sequences....
[...]
...Table 1: Results obtained by Mean-shift algorithm [12]...
[...]
...The proposed method is compared with the mean shift algorithm [12] and it gives the better tracking results....
[...]
...[12] is implemented on our video datasets....
[...]

Journal Article•DOI•

Incremental Learning for Robust Visual Tracking

[...]

David A. Ross¹, Jongwoo Lim², Ruei-Sung Lin³, Ming-Hsuan Yang²•Institutions (3)

University of Toronto¹, Honda², Motorola³

01 May 2008-International Journal of Computer Vision

TL;DR: A tracking method that incrementally learns a low-dimensional subspace representation, efficiently adapting online to changes in the appearance of the target, and includes a method for correctly updating the sample mean and a forgetting factor to ensure less modeling power is expended fitting older observations.

...read moreread less

Abstract: Visual tracking, in essence, deals with non-stationary image streams that change over time. While most existing algorithms are able to track objects well in controlled environments, they usually fail in the presence of significant variation of the object's appearance or surrounding illumination. One reason for such failures is that many algorithms employ fixed appearance models of the target. Such models are trained using only appearance data available before tracking begins, which in practice limits the range of appearances that are modeled, and ignores the large volume of information (such as shape changes or specific lighting conditions) that becomes available during tracking. In this paper, we present a tracking method that incrementally learns a low-dimensional subspace representation, efficiently adapting online to changes in the appearance of the target. The model update, based on incremental algorithms for principal component analysis, includes two important features: a method for correctly updating the sample mean, and a forgetting factor to ensure less modeling power is expended fitting older observations. Both of these features contribute measurably to improving overall tracking performance. Numerous experiments demonstrate the effectiveness of the proposed tracking algorithm in indoor and outdoor environments where the target objects undergo large changes in pose, scale, and illumination.

...read moreread less

3,151 citations

"Performance evaluation of object de..." refers methods in this paper

...The online multiple instance learning algorithms [6] successfully tracked an object in real time where lighting conditions change and object occlusion occurs....
[...]
...[6] proposed an adaptive tracking method that shows robustness to large changes in pose, scale, and illumination via incremental principal component analysis....
[...]