A color texture based visual monitoring system for automated surveillance

This paper describes a visual monitoring system that performs scene segment ation based on color and texture information. Color information is combined with texture and corresponding segmentation algorithms are developed to de tect and measure changes (loss/gain) in a given scene or environment over a period of time. The xyY color space is used to represent the color informa tion. The two chromaticity coordinates (x, y) are combined into one, thus, providing the chrominance (spectral) part of the image, while Y describes t he luminance (intensity) information. The proposed color texture segmentati on system processes luminance and chrominance separately. Luminance is proc essed in three stages: filtering, smoothing, and boundary detection. Chromi nance is processed in two stages: histogram multi-thresholding, and region growing. Two or more images may be combined at the end in order to detect s cene changes, using logical pixel operators. Bs a case study, the methodolo gy is used to determine wetlands loss/gain. For comparison purposes, result s in both the xyY and HIS color spaces are presented.