Target detection and fusion method based on long-focus and short-focus multi-camera vehicle environment

Abstract

The invention discloses a target detection and fusion method based on a long-focus and short-focus multi-camera vehicle environment. The method comprises the following steps of: 1, performing target detection on images acquired by a long-focus and short-focus binocular camera by adopting a convolutional neural network to obtain target frame positions in the images acquired by the cameras with different focal lengths at the same moment; 2, according to a camera imaging principle and internal and external parameters K, R and T obtained by camera calibration, obtaining a mapping relation f of thespace target point P under a long-focus camera pixel coordinate system and a short-focus camera pixel coordinate system; and 3, obtaining the position of the corresponding target frame in the short-focus camera image through the mapping relationship f according to the position of the target frame in the long-focus camera image, and fusing the position of the corresponding target frame with the target in the original short-focus camera image, thereby realizing target detection tasks under different distance conditions. According to the invention, the limitation that a single focal length camera cannot adapt to target detection tasks in different distances is overcome, and the accuracy of target detection in a vehicle environment is improved. Meanwhile, the method is simple, easy to use, low in cost and high in real-time performance.

Description

technical field [0001] The invention belongs to the technical field of intelligent vehicle environment perception, and in particular relates to a target detection and fusion method in a long and short focal length multi-camera vehicle environment. Background technique [0002] In recent years, with the rapid development of artificial intelligence, machine vision and other fields, autonomous driving has become an important field of research in academia and industry. Environmental perception technology is one of the key technologies in the automatic driving system, and it is also the most basic module. It tells the surrounding environment of the vehicle like the eyes of the car. Target detection, localization, and motion state estimation are the most basic functions in the environment perception module. [0003] With the wide application of deep learning and the substantial increase in the computing power of computing devices, the environment perception technology based on de...

AI Technical Summary

Problems solved by technology

The field of view of the telephoto camera is narrow, and the target in the distance is large, which is easy to be detected; but the near target may not be collected due to the camera field of view.

Images