Vision-Based Distance Estimation for Multiple Vehicles Using Single Optical Camera

This paper presents a novel algorithm which can be applied in the telematic systems to detect the multiple vehicles´ distances. The characteristic of the proposed algorithm is that it only makes use of a single optical camera to capture images of vehicles and without using other auxiliary equipment. The working principle of the proposed algorithm is to use width of the license plate as the known information, and then calculate the pixel size of the license plate appeared in the CCD image. The extraction of the license plate region is performed by the image processing software. This paper derives a distance-pixel relation equation from the evaluation CCD images with known vehicles´ distances. After a simple mathematical operation, one can immediately obtain the distance information of multiple vehicles within the CCD image using the distance-pixel relation equation. This distance measure method can display the individual distance of multiple vehicles located at difference positions. Moreover, in comparison with other traditional methods, the overall structure of the proposed algorithm is quite simple. It just requires only a camera with a data processor and is very applicable in telematic systems.