Parking regulation enforcement system

Abstract

A video camera mounted on a parking enforcement patrol vehicle and connected to a computer near the operator. The system is driven along a patrol route where parked vehicles are governed by a posted time limit. The system enforces the local parking regulation by automatically determining whether or not each parked car has been parked longer than the posted time limit. Violations are detected by applying a License Plate Recognition algorithm to the images. Each license plate number is time-tagged, geo-referenced and entered into a local database. When the patrol vehicle re-traces the patrol route after the posted parking time limit has expired, the database is searched to flag vehicles that were observed at the same location during the previous circuit and therefore in violation of the parking regulations. When the system detects a parking violation, it prints a parking citation that the operator affixes to the offending parked vehicle.

Description

BACKGROUND OF THE INVENTION1. Field of the InventionThis invention relates to law enforcement and more particularly to an automated means for detecting vehicles that have been parked for longer than the legally prescribed period.2. BackgroundMunicipal governments enact regulations to govern the parking of cars along city streets. Typically, time limits are posted along each street and parking fines are levied on vehicle owners who park their cars for longer than the posted time. Two benefits result from the practice of making and enforcing on-street parking regulations:1) Traffic congestion is reduced by forcing motorists parked for long periods to find suitable off-street parking arrangements, thereby vacating their more convenient, on-street parking spaces for use by motorists wishing to stop for short periods.2) The parking fines levied on motorists who violate parking regulations create revenue the municipality.In order to reap these benefits, the fundamental technical problem f...

AI Technical Summary

Problems solved by technology

Spurious shadows in the vehicle's raster image will degrade the spat ial fidelity of vectors extracted from it.

However, there will typically be many vectorized rectangles in the vector-model that complicate isolating the plate-model (dealer logos, bumper stickers, parking permits, decorative trim etc.).

Full Recognition Mode LPR algorithms cannot recognize the plate-strings of all observed vehicles with 100 percent accuracy.

However, for some applications a certain number of plate-string errors is acceptable.

However, Pattern-matching LPR cannot extract useful information from a single image and cannot make the legal link to the vehicle's owner (only Full Recognition Mode LPR can provide that information).

Images