Multi-vehicle target identification method based on improved YOLOv2 model

Abstract

The invention discloses a multi-vehicle target identification method based on an improved YOLOv2 model, and the method comprises the steps: firstly collecting sample data in an actual traffic environment, and dividing the sample data into sample images of a training set and a test set according to a proportion of 7: 3; then performing data enhancement on the training set sample images, including random scaling of the sample images and adjustment of exposure and saturation so that the processed images are enabled to act as the input of the training model; carrying out target area feature vectorextraction on the processed training set through an improved Darknet-19 network; inputting the training set into a Darknet-19 network model for training to obtain a detection and recognition model; and finally, inputting the test set into the model for testing to obtain a multi-target vehicle identification result. According to the method, the problems of low detection rate, poor robustness and non-ideal classification effect of a road vehicle multi-target detection and vehicle type classification method in the prior art are solved.

Description

technical field [0001] The invention belongs to the technical field of image detection and classification, and in particular relates to an improved YOLOv2 model multi-vehicle target recognition method. Background technique [0002] Image detection and image classification technology is an important part of image processing technology, which is widely used in many fields, such as remote sensing image recognition, military criminal investigation, modern biomedicine, intelligent transportation, etc. However, traditional target detection and recognition methods, such as the Cascade classifier based on Haar features, are mainly for the detection of specific targets, which are limited to multi-category targets, and the process of target area selection is complicated, and the efficiency of detection and recognition is low. When selecting objects, its feature extraction has strong subjectivity, poor robustness, weak generalization ability and other shortcomings, and it is difficult ...

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to provide a multi-vehicle target recognition method that improves the YOLOv2 model, which solves the problems of low detection rate, poor robustness and unsatisfactory classification effect for road vehicle multi-target detection and vehicle classification methods in the prior art

Images