Automatic security-inspection vehicle tracking and locating method and system of open scene

Abstract

The invention relates to an automatic security-inspection vehicle tracking and locating method and system of an open scene. The method includes: collecting video images of an inbound vehicle, and carrying out license-plate recognition according to the video images to obtain a license-plate recognition result; reading vehicle model information, which is in a database and corresponds to a license plate, and a corresponding time period of event occurrence according to the license-plate recognition result, wherein the vehicle model information includes color and outline features of a vehicle body;comparing the color of the vehicle body with background color of a target detection area to select a tracking algorithm; according to the selected tracking algorithm, judging, frame by frame, whethera lowest edge of the vehicle body appears in a monitoring area which is directly above a ground trench; starting feature-point key-point frame-by-frame matching tracking technology of the license plate to track license-plate location changes frame by frame if the lowest edge of the vehicle body appears in the monitoring area which is directly above the ground trench; and determining, according toa license-plate location change result, whether license-plate center coordinate changes are under a specific threshold value, and if the license-plate center coordinate changes are under the specificthreshold value, determining that the vehicle is stationary, and starting vehicle light locating and wiper motion recognition technology at the same time for automatic security inspection.

Description

technical field [0001] The invention relates to a fully automatic unattended vehicle safety routine inspection occasion, and cooperates with technologies such as computer vision-based car light positioning and wiper motion detection to form a complete set of fully automatic vehicle safety routine inspection and annual inspection system solutions. In particular, it relates to an automatic tracking and positioning method and system for security inspection vehicles in an open scene. Background technique [0002] With the development of information and artificial intelligence technologies such as big data, Internet +, and deep machine learning, in order to solve the contradictions of increasing traffic congestion and unbalanced allocation of load and capacity resources, the National Development and Reform Commission and the Ministry of Transport jointly proposed in 2016 The overall framework and development plan of the intelligent transportation (ITS) system, and emphasized the ...

AI Technical Summary

Problems solved by technology

The disadvantage is that large holes are caused by the uniformity of the gray distribution inside the object, which makes it impossible to find the connected domain of the moving target.

[0009] The shortcomings of the existing technology are as follows: 1) The existing motor vehicle safety routine inspection and annual inspection are mainly due to the fact that most of them rely on manual inspection, which is inefficient, poor in reliability, and the professional level of security personnel is different, and the security inspection standards are not uniform, so it cannot be realized. Automatic detection; 2) The detection results cannot be informatized, which is not conducive to information review management, responsibility tracking, and prevention in advance; 3) The current vehicle tracking and positioning technology based on computer vision is about the statistics of traffic flow and the analysis of traffic accidents , its algorithm focuses on the tracking of general vehicles, and is not suitable for tracking specific vehicles with rich prior information and identifying inspection items such as the location of each vehicle light and brightness change in the later stage; the current vehicle tracking algorithm cannot filter out the surrounding Irrelevant vehicles and pedestrians, unable to identify false moving targets caused by the driver, co-pilot and other personnel coming in and out and talking, unable to overcome the interruption of tracking targets caused by camera shake, and unable to overcome false vehicle targets caused by puddle reflections on the floor in rainy days Misjudgment formed

Images