We present a method to learn and recognize object class models from unlabeled and unsegmented cluttered scenes in a scale invariant manner. Objects are modeled as flexible constellations of parts. A probabilistic representation is used for al

能量熵的计算，表示出两帧图像间各个像素点的相对情况，实现了图像的灰度化并进一步用于视频监视控。- Energy entropy calculation, Between two images showing the relative circumstances of each pixel, Achieve a grayscale image and further control for video surveillance.