A method to segment moving human cast shadow based on wavelet multi-resolution features

In this paper, texture analysis method is proposed to segment and removal the human´s shadow, shadows always produced by the light. The method is composed of three step: firstly, using the Gaussian mixture model extract the foreground and background from the Human motion sequences, secondly, using wavelet decomposition to extract texture features and color features of the foreground and background and established the eigenvector that based on the pixels, finally, using SVM classified the eigenvector and if the foreground´s eigenvector and background´s eigenvector are the same, so the pixel is the shadow. The results showed that using the texture analysis method segment shadows clearly and it can do well at various environmental, for example the illumination changes or color deviation.