Visual identification-based illegal parking detection system

Abstract

The invention relates to the technical field of intelligent traffic, in particular to an illegal parking detection system based on visual identification, comprising the following steps: S100, acquiring acquisition information uploaded by a vehicle-mounted terminal, the acquisition information comprising image information and position information; s200, identifying a road sign, a vehicle position, vehicle information and a parking state in the image information; s300, whether illegal parking behaviors exist or not is judged according to the vehicle position, the parking state and the road sign; and S400, when it is judged that the illegal parking behavior exists, the image information, the position information, the vehicle information and a judgment result are sent to a supervision platform. According to the illegal parking detection system based on visual identification, detection of illegal parking in the whole road section can be realized, and the processing efficiency of illegal parking behaviors is improved.

Description

technical field [0001] The invention relates to the technical field of intelligent transportation, in particular to an illegal parking detection system based on visual recognition. Background technique [0002] The number of cars in the city has increased sharply, and the corresponding parking space growth is relatively lacking, which cannot meet the daily parking needs, especially in prosperous areas and road sections, where land resources are tight, parking spaces are few, and the vehicle density is high. In this case Many car owners parked their cars on both sides of the road, resulting in narrow roads, affecting the driving of vehicles, especially during peak hours, causing serious traffic congestion, and even causing vehicle scratches and collisions. [0003] The traditional illegal parking detection and law enforcement scheme mainly includes manual law enforcement and snapshot law enforcement. Manual law enforcement requires traffic law enforcement personnel to go to t...

AI Technical Summary

Problems solved by technology

[0002]The number of cars in the city has increased sharply, and the corresponding increase in parking spaces is relatively scarce, which cannot meet the daily parking needs, especially in prosperous areas and road sections, where land resources are tight. There are very few parking spaces, and the vehicle density is high. In this case, many car owners park their cars on both sides of the road, resulting in narrow roads and affecting the driving of vehicles. Especially during peak hours, it causes serious traffic congestion, and even causes traffic jams. Scratches and collisions

[0003]Traditional illegal parking detection and law enforcement solutions mainly include manual law enforcement and snapshot law enforcement. Manual law enforcement requires traffic law enforcement personnel to go to the scene to inspect and deal with it, which has low efficiency, high cost, and poor real-time performance And other issues

Capture law enforcement mainly uses cameras set up at intersections or no-parking areas to detect and identify illegal parking behaviors. Although it solves the problem of low efficiency of manual law enforcement to a certain extent, some vehicles will park at locations that cannot be photographed by monitoring. This solution has coverage The detection rate is low, and the detection of the whole road section cannot be realized. At the same time, there may be problems such as license plate recognition errors and false alarms due to unclear shooting.

Images