Adaptive threshold for motion detection in outdoor environment using computer vision

In this paper, an adaptive threshold method for motion detection in outdoor environment using computer vision is presented. This alternative solves the problem of several methods of thresholding when only one class exists in the captured images. A method of adaptive threshold is described and a modification to this method is proposed. The histogram of a distant matrix and the variance of a mean image are used to obtain an adaptive threshold that takes into account the varied illumination and weather conditions present in outdoor environment. To show the validity and robustness of this method, several banks of images were used to compare our proposal with another method described in this paper.