Deep learning-based traffic sign automatic identifying and marking method

Abstract

The invention provides a deep learning-based traffic sign automatic identifying and marking method, which is used in the technical field of environmental perception of intelligent vehicles. A semantic segmentation structure is adopted for locating and detecting traffic signs, and candidate regions with the traffic signs are obtained, wherein the semantic segmentation structure comprises two parts: an encoding network; and a decoding network and a pixel-based classification layer. Then, the traffic signs in the candidate regions are subjected to classified identification and locating through a convolutional neural network of a quick region. Based on the traffic sign automatic identifying and marking method, the invention furthermore correspondingly provides an effective map navigation information updating method. According to the method, the candidate regions with the traffic signs are located by using the semantic segmentation method, so that a new idea is provided, a training parameter quantity is reduced, the storage space is saved, and the computing time is shortened; and the method is high in identifying accuracy, and can perform relatively accurate traffic sign information updating on map navigation information, thereby better serving drivers conveniently.

Description

technical field [0001] The invention relates to the technical field of environment perception of smart cars, and is applicable to unmanned driving and assisted driving systems, in particular to an automatic recognition and labeling method of traffic signs based on deep learning. [0002] technical background [0003] With the development of society, cars have become an irreplaceable means of transportation for human daily life. However, it is followed by increasingly prominent security issues. Today, map navigation is a tool commonly used by people for daily travel. However, there are obvious shortcomings in map navigation. The road condition information it provides often changes due to reasons such as road maintenance, expansion, and renovation, and the information of traffic signs also changes from time to time. Therefore, it is very necessary to regularly update the information in the map navigation using the module that automatically recognizes traffic signs, which is r...

AI Technical Summary

Problems solved by technology

[0009] Aiming at the problem of poor recognition of traffic signs at present, the present invention provides a method for automatic recognition and labeling of traffic signs based on deep learning, which obtains image candidate areas containing traffic signs, and uses fast regional convolutional neural networks for the extracted traffic sign candidate areas Carry out classification and recognition, realize traffic sign recognition and labeling, the invention is used to update map navigation information, and at the same time provide new angles and ideas for the research of traffic sign recognition

Images