Cross-regional similar vehicle retrieval method and device

Abstract

The invention provides a cross-regional similar vehicle retrieval method and a cross-regional similar vehicle retrieval device. The method comprises the following steps: according to an image to be inquired comprising a vehicle and a first region to which the image to be inquired belongs, determining vehicle model information of the image to be inquired in the first region; determining a second region to which all images in a target database belong, and if the second region is different from the first region, determining a vehicle model template library of vehicle model information of the second region; selecting a plurality of sample images which meet time information of the target database and illumination conditions at different angles and in different scenes from the vehicle model template library to form an inquiry image set of the second region; acquiring a retrieval result of each sample image in the inquiry image set and all images in the target database; determining vehicles similar to the vehicle in the image to be inquired in the target database according to the retrieval result, wherein the vehicle model template library comprises a plurality of vehicle sample images under different illumination conditions, at different shooting angles and in different scenes. According to the method, the problem of drop of the vehicle retrieval performance caused by the difference of monitoring point vision models of different regions in a true scene can be solved.

Description

technical field [0001] The invention relates to the technical field of intelligent transportation, in particular to a cross-regional similar vehicle retrieval method and device. Background technique [0002] With the rapid development of my country's economy, the continuous expansion of urban scale, and the substantial increase in the number of vehicles, my country's transportation system is gradually becoming intelligent. Traffic surveillance video is an important data basis for public security services, and plays a vital role in maintaining social security and stability and combating crimes. Among them, retrieving the target vehicle from a large number of surveillance videos is a basic requirement. However, when it is necessary to retrieve a certain target vehicle in different regions, such as different cities, the existing technology generally directly uses the query image to search in databases in various places. Vehicle vision models vary greatly, resulting in a sharp...

AI Technical Summary

Problems solved by technology

[0004] Aiming at the defects in the prior art, the present invention provides a cross-regional similar vehicle retrieval method and device, which can solve the problem of vehicle retrieval performance degradation caused by differences in visual models of monitoring points in different regions in real scenes

Images