The classification of mice behavior is very important in the study fields of neuroscience, biological science, drug development and so on. In behavioral analysis,according to the short limbs of mice, it is difficult to extract relevant information. Mouse behavior is classified using characteristics of mouse contour. Firstly, mouse contour is extracted from each frame behavior video. Then, the distance of eight directions are calculated which are from contour center to edge and calculate two main components of contour. Using these values to form ten dimensional vectors as feature vector. Finally, calculate the log-covariance distance and classify. Experimental results show that the correct rate of classification is 87.6%.