An automatic system for counting and capturing the pictures of moving vehicles in real-time

In recent years, there has been an increase in the interest placed on Intelligent Traffic Monitoring Systems (ITMS). This paper presents an inexpensive, portable, automatic vision based moving vehicle detection and capturing system in real time using a single video camera. Vehicle counting is done considering a smaller window from the main frame. Predefining the window gives the upper hand on excluding the irrelevant movements from a far distance and also less computation time. We define a specific method based on frame comparison with the background and pattern matching avoiding multiple counts with adaptive updating of the background with the elapsing periodic time. The system is implemented and the experimental results from the traffic scenes are provided which demonstrate the effectiveness of the approach.