Segmentation of Motion Objects from Four Successive Video Frames Simultaneously Using Multiple Correlation

摘要

Identifying moving objects from a video sequence is a fundamental and critical task in many computer vision applications. We develop an efficient adaptive segmentation algorithm for color video surveillance sequences, which uses four frames simultaneously and works in non-stationary background in indoor environment. At runtime, segmentation is done by checking color intensity values at corresponding pixels P(x,y) in four frames simultaneously using temporal differencing. Background modeled with statistical multiple correlation coefficient using pixel-level based approach. The segmentation starts from a seed in the form of 3×3 image blocks to avoid the noise. Usually, temporal differencing generates holes in motion objects. After subtraction, holes are filled using image fusion, which uses spatial clustering as criteria to link motion objects. The emphasis of this approach is on the robust detection of moving objects even under noise or environmental changes (indoor).