System and method for detecting vehicle
View Patent ↗The invention relates to a vehicle detection method which includes: photographing a front-view of a vehicle; detecting an edge from an image photographed by a camera and generating a horizontal binary image and a vertical binary image; detecting a bottom surface candidate group relating to a preceding vehicle from the horizontal binary image and detecting a side surface candidate group relating to the preceding vehicle from the vertical binary image; when one of a pair of side surface candidate groups and bottom surface candidate groups of the pair of side surface candidate groups is not detected, estimating the side surface candidate group and bottom surface candidate group and generating a vehicle candidate group; and filtering the vehicle candidate group and detecting the preceding vehicle. Therefore, even when all of vehicles are not perfectly photographed, a preceding vehicle can be more precisely detected.
1. A vehicle detection method comprising:
photographing a front-view of a vehicle;
detecting an edge from an image photographed by a camera and generating a horizontal binary image and a vertical binary image;
detecting a bottom surface candidate group relating to a preceding vehicle from the horizontal binary image and detecting a side surface candidate group relating to the preceding vehicle from the vertical binary image;
when one of a pair of side surface candidate groups and bottom surface candidate groups of the pair of side surface candidate groups is not detected, estimating the side surface candidate group and bottom surface candidate group and generating a vehicle candidate group; and
filtering the vehicle candidate group and detecting the preceding vehicle.
2. The method of claim 1 , wherein when the pair of side surface candidate groups only is detected, the generating of a vehicle candidate group is to generate the vehicle candidate group by estimating a bottom surface candidate group by the lower end of the pair of side surface candidate group.
3. The method of claim 1 , wherein when one side surface candidate group only is detected and there is a bottom surface candidate group adjacent to the one side surface candidate group, the generating of the vehicle candidate group is to generate the vehicle candidate group by estimating the other side surface candidate group facing the one side surface candidate group so as to be adjacent to the bottom surface candidate group.
4. The method of claim 3 , wherein the generating of the vehicle candidate group comprises generating the other side surface candidate group at the position spaced apart from the one side surface candidate group by the width of a vehicle.
5. The method of claim 1 , wherein when there are the pair of side surface candidate groups and a bottom surface candidate group between the pair of side surface candidate groups, the generating of a vehicle candidate group is to generate a vehicle candidate group by connecting the pair of side surface candidate groups and the bottom surface candidate group.
6. The method of claim 1 , further comprising setting a region-of-interest on the basis of the maximum distance at which a preceding vehicle can be detected from an image photographed by the camera,
wherein, in generating the horizontal binary image and the vertical binary image, an edge is detected from a region-of-interest among images photographed by the camera and the horizontal binary image and the vertical binary image are generated.
7. The method of claim 1 , wherein, in detecting the preceding vehicle, the preceding vehicle is detected from the vehicle candidate group by using a cascade-type filtering model generated by an adaptive boosting method.