A cross-domain pedestrian re-identification method, device and storage medium

Abstract

The present invention provides a cross-domain pedestrian re-identification method, device, and storage medium. The method includes: respectively obtaining multiple images corresponding to respective target areas from photographing devices preset in multiple target areas, and using multiple images of each target area A plurality of corresponding image sets are respectively obtained for each image; image set combination is performed according to the number of multiple image sets, and image style conversion processing is performed on the combined image set to obtain an image data set to be identified; The image data set is recognized and processed to obtain the recognition result. The present invention enhances the diversity of data set samples compared to before conversion, reduces the inter-domain differences of samples, makes the model have better generalization ability, can effectively improve the accuracy of cross-domain pedestrian re-identification, and can effectively Solve the problem of low accuracy of cross-domain pedestrian re-identification, so that it can better serve the social public security system.

Description

technical field [0001] The present invention mainly relates to the technical field of image data set processing, and in particular to a cross-domain pedestrian re-identification method, device and storage medium. Background technique [0002] Image dataset processing techniques transform an input image dataset into another image dataset with desired characteristics. For example, the output image data set can be processed to have a higher signal-to-noise ratio, or the details of the image data set can be highlighted through enhanced processing, so as to facilitate the operator's inspection. In computer vision research, image processing technology is often used for preprocessing and feature extraction. [0003] Image dataset processing is the technique of analyzing image datasets with a computer to achieve desired results. Also known as image processing. Image dataset processing generally refers to digital image dataset processing. A digital image data set refers to a larg...

AI Technical Summary

Problems solved by technology

It is designed to make up for the visual limitations of fixed cameras, and can be combined with pedestrian detection / pedestrian tracking technology. It can be widely used in intelligent video surveillance, intelligent security and other fields. Existing recognition methods have high requirements for samples, and due to target data It is difficult to obtain and lead to problems such as low accuracy of cross-domain pedestrian re-identification

Images