Vehicle re-recognition method based on space-time constraint model optimization

Abstract

The invention discloses a vehicle re-recognition method based on space-time constraint model optimization. The method comprises the following steps: 1) obtaining a to-be-queried vehicle image; 2) fora given vehicle query image and a plurality of candidate pictures, extracting vehicle attitude features through a vehicle attitude classifier and outputting a vehicle attitude category; 3) fusing thevehicle attitude feature and the fine-grained identity feature of the vehicle to obtain a fusion feature of the vehicle based on visual information, and obtaining a visual matching probability; 4) estimating the relative driving direction of the vehicle, and establishing a vehicle space-time transfer model; 5) obtaining a vehicle space-time matching probability; 6) based on the Bayesian probability model, combining the visual matching probability and the space-time matching probability of the vehicle to obtain a final vehicle matching joint probability; and 7) arranging the joint probabilitiesof the queried vehicle and all candidate vehicles in a descending order to obtain a vehicle re-recognition sorting table. The method provided by the invention greatly reduces the false recognition rate of vehicles and improves the accuracy of a final recognition result.

Description

technical field [0001] The invention relates to vehicle re-identification technology, in particular to a vehicle re-identification method based on time-space constraint model optimization. Background technique [0002] In the urban transportation system, managers are faced with difficulties such as the increasing traffic volume and the huge three-dimensional transportation network. The analysis of vehicles has become the core of smart urban transportation. Technologies such as vehicle retrieval, tracking, and re-identification are of vital significance in the field of security. Vehicle re-identification is a cross-camera surveillance video automatic retrieval technology for specific vehicle objects, that is, matching the same vehicle object that appears at different times under multiple cameras with no overlap in the illuminated area. Vehicle re-identification technology aims to identify the same amount of vehicles in a complex environment, and has a wide range of applicati...

AI Technical Summary

Problems solved by technology

When encountering external factors such as vehicle attitude changes, low resolution, camera settings, etc., the above vehicle re-identification techniques cannot guarantee high accuracy

Especially for re-identification of vehicles with similar models and colors, visual information alone cannot solve the problem.

Images