Vehicle distance detection based on monocular vision

摘要

Keeping vehicle safety distance through early warning for road vehicles is very important and it could greatly reduce the occurrence of road traffic accidents. This article introduce a vehicle distance detection method based on monocular vision. Nowadays, due to the progress of deep neural network, object detection technology has reached a higher level, and its detection accuracy (recognition rate, location and object size) has been greatly enhanced. In this paper, the advanced object detection technology is applied to highway vehicle distance detection. Combined with the traditional visual detection research results, we propose a monocular vision ranging method based on convolutional neural network. This method has high accuracy In a certain distance range, and it is significant for the vehicle safety warning.