Fusion-feature-based video vehicle re-identification method and system

Abstract

The invention discloses a fusion-feature-based video vehicle re-identification method and system. The method comprises: a target vehicle and a matching range are determined; according to the target vehicle and the matching range, a color feature similarity degree between a target vehicle image and a to-be-matched vehicle image is calculated by using a color-histogram-based similarity calculation method; on the basis of the target vehicle and the matching range, a similarity degree of coding direction gradient histogram features of the target vehicle image and the to-be-matched vehicle image iscalculated by using a direction gradient histogram feature similarity calculation method based on local linear constraint coding and weighted spatial pyramid; and weighting fusion is carried out on the calculated color feature similarity degree and the coding direction gradient histogram feature similarity degree. According to the invention, on the basis of the weighting fusion of the calculatedcolor feature similarity degree and the coding direction gradient histogram feature similarity degree, the similarity result of vehicle re-identification is obtained; and the accuracy, robustness, anduniversality are improved. The method and system are applied to the image processing field widely.

Description

technical field [0001] The invention relates to the field of image processing, in particular to a video vehicle re-identification method and system based on fusion features. Background technique [0002] Vehicle re-identification refers to the process of re-identifying the vehicle after the monitoring scene, lighting conditions, target angle and other conditions change in the video image (that is, the same vehicle is identified from surveillance videos or images collected at different times and locations). target vehicle), which is of great significance in the analysis of road traffic status and the fight against crimes. First of all, in real traffic flow, it often happens that vehicle information such as license plates cannot be collected due to occlusion or other environmental conditions. Through vehicle re-identification, license plate information can be obtained from another camera, and the vehicle’s driving path and Information such as vehicle travel time is used for s...

AI Technical Summary

Problems solved by technology

Another method is to train and learn the similarity measure criterion, so that the image similarity of the same vehicle is high, and the similarity between different vehicles is low.

Although the method of training and learning can effectively obtain features or measurement criteria with a high degree of discrimination, due to the large changes in vehicle appearance characteristics between different cameras, the generality of the target feature model or similarity measurement model obtained by it is not high. One of the more difficult problems of learning methods

Images