A method and device for detecting boundary points of target

The embodiment of the invention discloses a method and a device for detecting boundary points of a target, and relates to the technical field of automatic driving, in particular to the technical fieldof autonomous parking. The method comprises the steps of obtaining an original panorama formed by splicing a plurality of fisheye images collected at the same moment; inputting the original panoramainto a trained target boundary point detection neural network model to obtain a panorama of boundary points of a marked target output by the trained target boundary point detection neural network model; wherein the target is a target surrounding the central point of the panoramic image; wherein the expected output adopted by the trained target boundary point detection neural network model in the training stage is a panorama sample of a boundary point of a marked target determined by scanning a plurality of fisheye image samples of the marked target. According to the method, the operation timeof the boundary points of t