A license plate location method and system based on cascade morphological transformation

Abstract

The present invention relates to a license plate location method and system based on cascaded morphological transformation, comprising: performing down-sampling and grayscale processing on the original image to obtain a grayscale image; performing cascaded morphological transformation and binarization on the grayscale image, And preliminarily screen the China Unicom area to obtain several candidate areas; calculate the similarity between the color histogram information of each candidate area and the color histogram information of the standard license plate, and judge the license plate area according to the similarity; Each license plate area is mapped to the original image to obtain the corresponding original license plate area. Through grayscale processing, cascading morphological transformation and binarization processing on the image, the robustness of the license plate candidate area location and the license plate positioning speed are improved, while the interference of strong light is suppressed, the contrast of the license plate area text is enhanced, and the license plate area is improved. The screening accuracy is suitable for different lighting environments and complex scenes, and it can locate multiple license plates at the same time.

Description

technical field [0001] The invention relates to the technical field of image processing, in particular to a license plate location method and system based on cascade morphological transformation. Background technique [0002] Intelligent transportation is widely used in the management of vehicles in smart cities. License plate recognition is one of the core technologies, which is the application of computer image processing technology and pattern recognition technology in the field of intelligent transportation. License plate recognition technology is widely used in highway bayonet capture systems, intelligent parking systems and community vehicle management systems. The license plate recognition process is basically divided into three steps: license plate positioning, character cutting and character recognition. Technology is a crucial step in the license plate recognition technology. As the first step in the entire license plate recognition process, the successful positio...

AI Technical Summary

Problems solved by technology

The matching method of prior knowledge is more easily affected by light, pollution, and the environment, and the accuracy is lower

[0004] Most of the existing license plate location methods are only suitable for scenes with little background changes, such as toll booths and security cards. However, for various natural environment scenes in complex panoramas (for example, street view panoramas), the backgrounds are ever-changing. Especially for those panoramas containing a large number of trees and striped textured items, the existing license plate positioning methods are difficult to apply

Moreover, the existing license plate positioning methods have high requirements for the quality of the collected pictures, and are sensitive to light. If the license plate is too bright or too dark, the existing license plate positioning methods cannot handle it well, and its anti-interference Poor (that is, its robust--robustness is not high)

Images