THE EFFECT OF COLOR SPACE ON MEAN SHIFT OBJECT TRACKING: A COMPARATIVE STUDY | ||
| JES. Journal of Engineering Sciences | ||
| Article 9, Volume 42, No 3, May and June 2014, Pages 756-768 PDF (377.03 K) | ||
| Document Type: Research Paper | ||
| DOI: 10.21608/jesaun.2014.115019 | ||
| Authors | ||
| Ahmed Nabil Mohamed1; Mohamed Moness Ali2 | ||
| 1Assistant Lecturer, Department of Computer and Information Systems, Sadat Academy | ||
| 2Professor, Department of Computers and Systems, Engineering Faculty, Minia University | ||
| Abstract | ||
| The mean shift algorithm is a widely used tool for robust and real-time object tracking that can use any feature space such as the color space. Therefore, in this article, we have studied and analyzed the various effects of using different color space configurations with the mean shift algorithm on the tracking process in surveillance videos. We have done so by investigating the potency of the gray space and four other color spaces (RGB, HSV, YIQ, YCbCr [each color space was studied under all the available seven configurations]). In the end, we have reached a conclusion that deciding which color space (or which configuration of the color space channels) to be used not only can be a very critical decision which may affect greatly the results of the tracking process, but also there is no color space configuration that can succeed all the time in all situations. Therefore, it is better to select the appropriate color space configuration according to the situation at hand. | ||
| Keywords | ||
| Tracking; Color Spaces; Mean Shift | ||
|
Statistics Article View: 94 PDF Download: 325 |
||
