Calculation of Speed in Car Accidents with the Image Processing Method

Calculating the speed of accidents is an important part of the collision scene. In the last two decades the design of the car brake system has changed and most cars are equipped with ABS anti-lock system. However, the braking function of the vehicles can be found in a significant way, and this causes the cars to crash when they collide with them. As in such cars, the effect of braking on the road surface is unstable and does not last long. The purpose of this article is to use the processing of images obtained from video footage to calculate the speed of vehicles involved in an accident.

In this research, to solve the problem of determining the speed of vehicles equipped with (ABS) in traffic accidents, the processing of images from municipal surveillance cameras or any project cameras in the city, according to their location and angle, and a homographic method has been used. This method directly determines the speed of the vehicle by using the video of a car accident and converting it into several frames, regardless of the dimensions of the vehicle.

The research findings show that the relative error between the calculated speed and the actual speed is less than 4%.

The proposed method generally overcomes the shortcomings of conventional methods of calculating speed. Also, this method has simpler calculations than other image processing methods, such as the nail-hole model (Pine-Hole) or image conversions, and requires less information and measurements on the ground.