Vehicle re-identification method, system, equipment and medium

Abstract

According to the vehicle re-identification method, system and device and the medium, the visible light image and the infrared image of the to-be-identified vehicle are obtained, the visible light image is compared for the first time, the candidate image and the first comparison result are obtained, then the infrared image is compared with the candidate image, and the second comparison result is obtained; the candidate ranking of each candidate image is determined by combining the first comparison result and the second comparison result so as to realize the re-identification of the to-be-identified vehicle, the characteristic information used in the vehicle re-identification technology is enriched by simultaneously utilizing various spectral information of the visible light image and the infrared image and combining the glass transmission characteristic of the infrared image, and the vehicle re-identification efficiency is improved. The vehicle re-identification precision is improved, the vehicle re-identification with higher precision can be completed under the condition that the license plate cannot be identified through the assistance of the internal information of the vehicle, and the environmental adaptability is relatively strong.

Description

technical field [0001] The present invention relates to the field of computer technology, and in particular, to a vehicle re-identification method, system, device and medium. Background technique [0002] Vehicle re-identification technology has been widely used in traffic monitoring, public safety and other fields. Such as vehicle trajectory tracking, vehicle positioning, etc. In the related art, the vehicle re-identification technology is mainly based on the visible light imaging technology. Through deep learning technology, the characteristics of the human body to be recognized and the characteristics of the target to be retrieved are obtained, and the similarity is matched to complete the re-identification; if the imaging conditions are good, the license plate can also be added. Recognition technology to complete accurate vehicle re-identification by comparing license plate numbers. [0003] However, based on the actual application scenarios, the current vehicle re-ide...

AI Technical Summary

Problems solved by technology

[0003] However, based on the actual application scenario, the current vehicle re-identification technology has the following problems and difficulties: 1. The traffic monitoring environment is complex and changeable, and there are a large number of scenes where the license plate cannot be recognized, so the license plate recognition is difficult to apply; 2. The appearance of the vehicle is relatively similar, especially There are many popular vehicles of the same brand, same model, and same color, with a high degree of similarity. Except for license plate comparison, it is difficult to distinguish them, which may cause mismatches; 3. The characteristics of the front and rear of the vehicles are quite different, making matching difficult

Based on the above problems, the current vehicle re-identification accuracy is low, and it is difficult to re-identify in difficult scenarios

Images