Study On Obstacle Detection Technology Of Road Surface Of The Driver Assistance

The vehicle assistant driving technology has been always one of the hot research content in recent years, and has good application prospects in the field of military, civil and industrial.To solve the problem of roadway vehicle obstacle detection in the vehicle assistant driving technology, this paper studied an algorithm of the roadway obstacle detection based on the vehicle bottom shadow. Firstly, on account of the characteristics of more noise interference of the actual road image, the image was needed to preprocess to reduce the impact of the noise information of the obstacle detection. Secondly the preprocessed image edge was extracted, and the Hough transform method was used to extract the boundary information of the road according to the linearity assumption of the road so that the range of obstacle detection could be narrowed and the amount of the data was reduced. Afterwards, a dual adaptive threshold segmentation method was used to exclude the impact of the shadow of the road and the roadside trees to the shadow detection of the vehicle bottom shadow. And the vehicle bottom shadow feature was utilized to complete the roadway vehicle obstacle detection. Finally, a hardware system implementation on the assistant driving was introduced and a design of the DSP module was especially recommend. Simulation results shows, the algorithm of the roadway vehicle obstacle detection based on the vehicle bottom shadow can detect the preceding moving vehicle, and has good reliability and robustness.