MEDOUAKH, S. University of Biskra - Energie Systems Modeling Laboratory, Algeria , BOUMEHRAZ, M. University of Biskra - Energie Systems Modeling Laboratory, Algeria

MEAN SHIFT BASED OBJECT TRACKING: THE EFFECT OF COLOR SPACE

21712

The mean shift algorithm is widely used in object tracking because of its speed, efficiency and simplicity. This algorithm is used to track the location of a non rigid object in image sequences using the object s color histogram. Mean shift tracker maximizes iteratively the appearance similarity by comparing the color histogram of the target model and the target candidate. In this paper we focus on studying the effect of color spaces (HSV, YCrCb, YIQ, YUV, I1I2I3 Lab) on object tracking by mean shift algorithm since this algorithm is based on color information (color histogram) and traditional color histogram uses only the RGB color space. The experimental results show that each color space can influence the tracking robustness for different color targets having the same background.

67

Mean Shift algorithm , Color Histogram , Color Spaces , Object Tracking

Courrier Du Savoir

74