17 research outputs found
Moving shadow detection in video using cepstrum
No full textMoving shadows constitute problems in various applications such as image segmentation and object tracking. The main cause of these problems is the misclassification of the shadow pixels as target pixels. Therefore, the use of an accurate and reliable shadow detection method is essential to realize intelligent video processing applications. In this paper, a cepstrum-based method for moving shadow detection is presented. The proposed method is tested on outdoor and indoor video sequences using well-known benchmark test sets. To show the improvements over previous approaches, quantitative metrics are introduced and comparisons based on these metrics are made. © 2013 Cogun and Cetin; licensee InTech
Cepstrum based method for moving shadow detection in video
No full textDate of Conference: 22-24 September, 2010Conference name: Proceedings of the 25th International Symposium on Computer and Information Sciences, 2010Moving shadows constitute problems in various applications such as image segmentation and object tracking. Main cause of these problems is the misclassification of the shadow pixels as target pixels. Therefore, the use of an accurate and reliable shadow detection method is essential to realize intelligent video processing applications. In this paper, the cepstrum based method for moving shadow detection is presented. The proposed method is tested on outdoor and indoor video sequences using well-known benchmark test sets. To show the improvements over previous approaches, quantitative metrics are introduced and comparisons based on these metrics are made. © 2011 Springer Science+Business Media B.V
Integrated sensor detection/localization for multi-source data
No full textIn this paper, the detection/localization of a target based on radio-frequency (RF) and infra-red (IR) data sources problem is addressed. The target is assumed to radiate an RF signal to multiple widely distributed sensors in space and is imaged using multiple frames of an IR sensor. The goal is to integrate RF and IR data to reliably detect and localize the target. The generalized likelihood ratio test (GLRT) approach is employed to find the detector. In order to reduce the computation required by a straightforward GLRT, the random basis function (RBF) approach is used. © 2014 IEEE
Object tracking under illumination variations using 2D-cepstrum characteristics of the target
No full textDate of Conference: 4-6 Oct. 2010Most video processing applications require object tracking as it is the base operation for real-time implementations such as surveillance, monitoring and video compression. Therefore, accurate tracking of an object under varying scene conditions is crucial for robustness. It is well known that illumination variations on the observed scene and target are an obstacle against robust object tracking causing the tracker lose the target. In this paper, a 2D-cepstrum based approach is proposed to overcome this problem. Cepstral domain features extracted from the target region are introduced into the covariance tracking algorithm and it is experimentally observed that 2D-cepstrum analysis of the target object provides robustness to varying illumination conditions. Another contribution of the paper is the development of the co-difference matrix based object tracking instead of the recently introduced covariance matrix based method. ©2010 IEEE
Alternative approaches to data compression for distributed detection
No full textIn this paper, the distributed detection problem of linear and nonlinear signals embedded in white Gaussian noise (WGN) is considered. First, the asymptotically optimal generalized likelihood ratio test (GLRT) detector is derived for both signal models. It is found that the GLRT detector requires the submission of all observed data to the central processor which is practically infeasible. Thus, several test statistics based on compressing the observed data at each sensor are proposed. Monte Carlo simulations are carried out to plot the receiver operating characteristic (ROC) curves in order to compare the performance of the proposed detectors for a nonlinear signal example
Multimodal target detection via integrated GLRT
No full textThe focus of this paper is the multimodal detection and localization of a ground moving target based on radio frequency (RF) and infrared (IR) data. The target radiates a low probability of intercept (LPI) RF signal received by multiple passive RF sensors at scene and is imaged by using a stationary IR camera concurrently. To obtain the multimodal detector proposed in this paper, first, the generalized likelihood ratio test (GLRT) is employed to derive the RF and IR detectors individually. Then, the RF and IR detectors are integrated optimally to get the integrated GLRT. In order to avoid the computational complexity associated with the integrated GLRT, a suboptimal multimodal detector is implemented by applying the random basis functions (RBF) approach to the IR image sequence to reduce down the search space for the RF detector. It is shown that the suboptimal multimodal detector has better localization performance compared to the RF detector when one of the RF sensors is partially occluded
Moving Shadow Detection in Video Using Cepstrum
No full textMoving shadows constitute problems in various
applications such as image segmentation and object tracking.
The main cause of these problems is the misclassification of
the shadow pixels as target pixels. Therefore, the use of an
accurate and reliable shadow detection method is essential
to realize intelligent video processing applications. In this paper, a cepstrum‐based method for moving shadow
detection is presented. The proposed method is tested on
outdoor and indoor video sequences using well‐known
benchmark test sets. To show the improvements over
previous approaches, quantitative metrics are introduced
and comparisons based on these metrics are made
