The purposes of this research were to classify and count the types of vehicles moving on a road by used digital image processing. The object of this research was a video footage of the flow of vehicles. Then the horizontal line is used as the area classification and counted of moving vehicles. Furthermore, each object in the frame of the video processed using background subtraction algorithm with binary method to obtain the foreground and background images. Foreground image is filtered and performed morphological processes to remove noise in the image. Then the results by blob detection to get a part of the pixel which is the object moves. The next stage, each object obtained was classified by the value of object width,then it will be used as a parameter for vehicle classification. If the value of the parameter it will be recognized as type that is defined. The results of tested application can classify vehicles in three types of vehicles are motorcycle, car, and truck/bus with the high average total accuracy was 92.07%