A method to detect the fluid motion in microfluidic chip with imagery processing technology is introduced,which solves the problem that the fluid motion can’t be detected with common method. Firstly,with the fact that in flow process the brightness change is the biggest in the edge of tiny tube,the motion trail can be collected with frame difference method. Then this destination area is cropped to further reduce the amount of calculation in later steps. Secondly,repeat several times of open operation and close operation to obtain the images,which has smooth boundary and barely background noise. Thirdly,find all edge pixels of this motion trail image through boundary tracking algorithm and then obtain the motion trail lines with the Hough transform detection approach. At last,with the lines’ length and the motion time,the flow rate can be calculated easily. Furthermore,some characteristics of the flow can also be obtained,such as the shear stress and the viscosity of the fluid. This paper explicitly states this method process combining with a chip that use for viscosity measurement. And the results have high accuracy and good repeatability,which is demonstrated by the data that standard deviation is less than 0.7 and the precision is less than 1/100.