License Plate Location Algorithm Research In Complex Condition

As an important application form of target automatic recognition, vehicle license plate recognition( LPR) technology is used in some fields, such as electron charging, pass controlling and automobile stream supervising etc.The paper introduces license plate location system based on the principle of Run Length and Tamura texture feature. The system is developed based on video streams, mainly includes vehicle motion region detection, license plate image processing, vehicle license plate location. The detection of vehicle motion region uses the background estimation methods based on the average of multi-frame; license plate image processing includes gray-scale transformation, median filtering, edge detection and binarization; the license plate location includes license plate horizontal location, license plate vertical position, color judgments and accurate location.The system is achieved in the C++ Builder environment, and it uses 1400 images that are of different plate sizes and under different lighting conditions to test, its location rates is up to 92% or more; but also the algorithm is packaged into a DLL library, it is tested in the Radar tester in video stream under various lighting conditions. The Test results prove that this algorithm is simple, fast, high accuracy, it meets real-time performance, versatility, and accuracy requirements, and it has a certain value.