Summary
International Technical Conference on Circuits/Systems, Computers and Communications
2008
Session Number:P2
Session:
Number:P2-60
Moving Object Segmentation using Weighted Coefficients Ratio Based on Discrete Wavelet Transform
Tae-Gyoun Lee, Ju-Young Kim, Tae-Yeon Jung, Jae-Sik Sohn, Duk-Gyoo Kim,
pp.-
Publication Date:2008/7/7
Online ISSN:2188-5079
Summary:
In this paper, it is proposed that edge detection using weighted coefficients ratio (WCR) video object segmentation algorithm based on discrete wavelet transform (DWT). The proposed method is applied DWT to two successive frames. Also, we utilize detection method with different thresholds in four wavelet sub-bands and make the map of boundary using Prewitt edge operator in wavelet domain. As mixing the difference and division operation, our algorithm improved the performance test that is able to find more accurate monving objects. on DWT. After the inverse discrete wavelet transform (IDWT), the robust edge map can be obtained. Through combination with the current frame's moving edges and previous can be detected and tracked. It is then used to extract video object planes (VOPs) by a simple filling technique. The proposed algorithm is robust to the entire motion object detection and can obtain fruther shape information, more accurate extraction of moving object. The experimental results are proved the effectiveness of our algorithm.