Fast traffic signboard recognition method based on convolution neural network

Abstract

The invention aims to solve the problems in the existing traffic signboard recognition method that the recognition target falls into a single group and the speed in doing so is slow. The invention, out of this concept, provides a fast traffic signboard recognition method based on convolution neural network, referred to as FTSR-CNN in abbreviation. This method comprises: using the convolution kernel sliding filter extracted characteristics; obtaining the loss of the network in the forward learning process, and ensuring the accuracy of the network model to the recognition of multiple categories of signboards; optimizing the network performances through the adjustment of the parameters, the activation of the function types and the reduction of dimensions for better accuracy and timeliness eventually; and at the same time, in order to make the samples more diverse, conducting data adding and expanding to the samples in the data set based on affine transformation. The recognition rates of the FTSR-CNN for two data set tests of the German traffic signboard data set GTSRB and the Tsinghua-Tencent 100K are recorded as 95.74% and 96.67% respectively. The results indicate that the recognition speed is increased on the same recognition accuracy level through the modification of a previous model network and the start up of different training strategies by the FTSR-CNN.

Description

technical field [0001] The invention relates to a fast recognition method for traffic signboards based on a convolutional neural network, which belongs to the technical field of image processing and can be applied to fast recognition of traffic signboards. Background technique [0002] As an important part of road traffic, traffic signs provide guidance, instructions, warnings and restrictions for drivers with words or symbols. Automatic identification of traffic signs is an integral part of the design of advanced driver assistance systems. Due to the wide variety of traffic signs, they are used in various scenes, and the contrast of sign images decreases with weather changes and lighting effects; physical damage and occlusion affect their inherent shape; motion blur caused by high-speed driving makes the automatic recognition of traffic signs very difficult. How to identify traffic signs quickly and accurately is a big challenge for designers. Several affected signage suc...

AI Technical Summary

Problems solved by technology

[0005] In view of this, for the problem of single classification, the convolutional neural network algorithm is used to achieve multi-class target classification. The convolutional neural network algorithm uses the original samples directly as the input of the network, and automatically obtains the most favorable features for classification through a large number of samples; On the basis of its data set, traffic signboards use affine transformation to increase the sample diversity, which improves the recognition rate to a certain extent; in order to solve the time-consuming problem caused by the huge amount of computation on the complex network in the recognition process, a variety of The small convolution kernel of the scale, and reduce the number of network outputs under the premise of ensuring the diversity of the underlying features, by adjusting the activation function and other factors in the network structure, an improved convolutional neural network FTSR-CNN is proposed. Signage recognition method Fast Traffic Sign Recognition-Convolution Neural Network, referred to as FTSR-CNN

Images