Traffic sign recognition method based on YOLO v4-tiny

Abstract

The invention discloses a traffic sign recognition method based on YOLO v4tiny. The method comprises the steps: collecting a traffic sign data set, carrying out the data enhancement of an initial sample traffic sign data set, and dividing the initial sample traffic sign data set into a training set, a verification set and a test set; for a real target frame in the training set, clustering six priori frame sizes with different sizes by taking an intersection-parallel ratio as an index, and embedding a channel attention mechanism and a space attention mechanism into a YOLO v4tiny framework to obtain a YOLO v4tinyCBAM network model; and training the network model through the training set, performing verification through the verification set, and finally testing the performance of the networkmodel through the test set. According to the method, a channel attention and spatial attention mechanism is introduced into the YOLO v4tiny lightweight network, so that the generalization ability is stronger, and the recognition precision is higher.

Description

technical field [0001] The invention relates to the field of image processing, in particular to an improved traffic sign recognition method based on the YOLO v4-tiny lightweight network. Background technique [0002] Today, with the development of AI intelligence, many breakthroughs have been made in the field of autonomous driving technology. For example, the intelligent vision system has achieved a qualitative leap in the recognition effect in just a few years. Technological breakthroughs have allowed more and more prototype smart cars to go out of the laboratory and be tested in real road environments, gradually moving towards practical applications. Traffic signs are used to transmit instruction information to vehicles and pedestrians, and are the key factors to ensure the smoothness and safety of traffic lines. Therefore, accurate and rapid recognition of traffic signs is the key link to realize the safety guarantee of automatic driving. For the task of traffic sign r...

AI Technical Summary

Problems solved by technology

Different from the R-CNN series algorithm, the network model parameters of the one-stage algorithm represented by SSD (Single Shot MultiBox Detector) and YOLO (You Only Look Once: Unified, Real-Time Object Detection) are relatively small, and in real time Performance is superior, but the accuracy is a bit unsatisfactory

With the update of the YOLO series, the fourth-generation algorithm YOLO v4 has been able to achieve high recognition accuracy on the basis of maintaining the recognition speed, especially the lightweight network YOLO v4-tiny series models have fewer parameters and are more suitable for deployment On edge devices such as vehicle systems, but for traffic scenes that contain many targets and small traffic sign targets, the recognition needs to be improved

Images