A traffic sign recognition method and system based on a deep space network

Abstract

The invention belongs to the technical field of traffic information processing and discloses a traffic sign recognition method and system based on a deep space network. The method includes: extractingimage information with space invariance by using a space transformation network; by improving the VGG network, fusing the features extracted from each stage of the network layer, using a spatial pyramid method to process the features of the fused image, and obtaining more features of the traffic sign images. A spatial transformation network is added to make up for the deficiency that the CNN cannot effectively extract the image information with spatial invariance. The VGG network is improved, fusion of the features extracted from each stage of network layer is carried out, the spatial pyramid method is utilized to process the fused image features, and more information of traffic signs image features are obtained. The experimental results show that the accuracy of traffic sign recognitionis improved on the basis of enhancing the network features.

Description

technical field [0001] The invention belongs to the technical field of traffic information processing, and in particular relates to a traffic sign recognition method and system based on a deep space network. Background technique [0002] At present, the existing technologies commonly used in the industry are as follows: [0003] Sun Wei [] and others proposed to use CNN network to extract multi-layer features, and use pyramid pooling method to pool features of each layer at multiple scales, combine these features to obtain multi-attribute traffic sign feature vectors, and input them into ELM classifier Carry out rapid identification of traffic signs; [0004] Traffic Sign Recognition (TSR) [1] It is an important component module of intelligent transportation system (ITS). Using image and information processing technology to conduct more accurate and timely traffic management to ensure driving safety. The traffic sign recognition system mainly includes two parts: detectio...

AI Technical Summary

Problems solved by technology

[0007] (1) The existing traffic sign recognition method based on deep learning cannot meet the generalization ability of the model for spatial invariance such as rotation, and the problem of insufficient feature extraction of small target images, resulting in poor accuracy in recognizing traffic sign images

[0008] (2) The existing traffic sign recognition method based on deep learning adopts the method of data enhancement (including transformation operations such as translation, rotation and scaling at a certain angle) to increase the diversity of samples and greatly increase the training traffic signs of deep learning network. time, although the accuracy of traffic sign recognition can be guaranteed, it is difficult to achieve fast and effective recognition of traffic signs

[0009] (3) The network layer of the neural network selected by the existing traffic sign recognition method based on deep learning is very deep, and there are many training parameters, which cannot be effectively applied to traffic sign recognition

[0011] Traffic sign images are all small target images, it is difficult to obtain enough features using the classic Let-5 network, and it will affect the final recognition rate

Images