The purpose of this paper is to present an analysis of image process for computing the surface velocity of the granular flow to develop the video monitoring system of the debris flows, and the low cost commercial Video-base systems can be also used to measure the velocity of the debris flow. In order to study the suit image recognizing method for the debris flows, the experiments of the granular flow are simulated as the debris flows; moreover, image recognizing method, Cross-Correlation, MQD, and STDS are used to compute the surface velocity. According to the results, Cross-Correlation method is useful to compute the surface velocity in higher density of the granular particles, however, STDS method is good to measure the movement of the big particles. Lastly, the nature debris flow images from the monitor of Soil and Water Conservation Bureau are analyzed to obtain the surface velocity by the image process.