Mei, B., Guo, X., Tan, G., Xu, Y. et al., "Vehicle Velocity Measurement Based on Image Registration," SAE Technical Paper 2017-01-0035, 2017, doi:10.4271/2017-01-0035.
Vehicle speed is an important factor to driving safety, which is directly related to the stability and braking performance of the vehicle. Besides, the precise measurement of the vehicle speed is the basis of some vehicle active safety systems. Even in the future intelligent transportation, high quality speed information will also play an important role.The commonly used vehicle speed measurement techniques are based on the wheel speed sensors, which are not accurate, especially when the wheels’ slip rate is not equal to zero. Focusing on these issues, image matching technology has been used to measure the vehicle speed in this paper. The image information of the road in the front of the vehicle is collected, and the pixel displacement of the vehicle is calculated by the matching system, thus accurately vehicle speed can be obtained. Compared with conventional speed measure technology, it has the advantages of wide measuring range, and high accuracy.Scale Invariant Feature Transform (SIFT) algorithm has been used in this paper to match the images. Firstly, the two-dimensional image Gauss scale-space has been established to extract the features in different scale space by the convolution integrated of input image and the Gauss kernel. To detect the feature points, Difference of Gaussian (DOG) scale-space has been built through the subtraction of adjacent scale-space images. Then, excluding the low contrast points and the edge points from local extreme points in DOG scale space, we can find the key points. To determine the displacement and the trajectory of the vehicle on the image, key points have been matched. So the global geometry transformation relation can be determined and the real-time speed of the vehicle can be calculated combined with the position of the camera.The research shows that the speed measurement system based on image registration is feasible on dry asphalt road and has good adaptability to the illumination. It can accurately detect the driving speed of the vehicle under conditions of acceleration, deceleration and turning.