Vehicle re-identification method based on double sub-networks

Abstract

The invention discloses a vehicle re-identification method based on double-subnetworks. The method comprises the following five modules: obtaining a vehicle image data set, preprocessing vehicle image data, constructing a vehicle image feature extraction neural network, training a vehicle re-identification model, extracting a to-be-identified vehicle image feature vector, and calculating the feature similarity between the to-be-identified vehicle image feature vector and a test set image. The vehicle re-identification method based on the double sub-networks aims to construct a neural network model with a strong discrimination capability, further solves the influence of factors such as illumination and a visual angle on a vehicle re-identification result, and improves the accuracy of vehicle re-identification. According to the invention, a double-subnetwork is designed on the basis of fusion of CNN and Transform structures, and effective global feature information can be obtained while multi-granularity local features of a vehicle image region are learned cooperatively, so that the discriminative characterization capability of a vehicle re-identification model is improved.

Description

technical field [0001] The invention relates to the technical field of vehicle re-identification, in particular to a dual-subnetwork-based vehicle re-identification method. Background technique [0002] As one of the key technologies of smart city and smart transportation, vehicle re-identification technology has received extensive attention and in-depth research from academia and industry in recent years. Vehicle re-identification, also known as Vehicle Re-identification, is essentially the same as pedestrian re-identification and is a type of re-identification task. This problem can be regarded as an image retrieval problem, which finds the same target vehicle image under the multi-view task of cross-camera scene based on the given vehicle image. [0003] With the rapid development of the field of deep learning, vehicle re-identification technology based on deep learning has also made great progress. The method based on deep learning usually first needs to build a neural...

AI Technical Summary

Problems solved by technology

This type of method ignores that discriminative features may appear anywhere in the vehicle, so it is weak for samples with local subtle differences;

[0007] Based on the method of generative confrontation network, some works use GAN to generate difficult samples to assist the network to deal with the problem of viewpoint change, but due to the limitation of GAN generation ability and the lack of confrontational samples, the authenticity of the generated content is poor

Images