License plate character automatic detection, positioning and recognition method in complex background environment

Abstract

The invention discloses a license plate character automatic detection, positioning and recognition method in a complex background environment. The method comprises the following steps: 1, the front color image of a vehicle is acquired, the acquired color image is preprocessed, screening conditions are selected according to the national standard for a license plate, whether an image area meets thescreening conditions exists is judged, if yes, the coarse position of a license plate initial area image is acquired initially and the image is saved separately, or otherwise, a next frame of color image is acquired for preprocessing again; 2, precise positioning on a license plate area is completed, and accurate segmentation on a single character of the license plate area is carried out; and 3, the similarity between a feature vector of a target character and a feature vector of a built standard character feature vector library is calculated, license plate character recognition is completed by using a similarity discrimination method, and target license plate comparison and judgment are carried out. The method of the invention has the advantages of fast operation speed, high recognition rate and strong background environment interference-resisting ability.

Description

technical field [0001] The invention relates to the technical field of intelligent transportation and image processing, in particular to a method for automatic detection, positioning and recognition of license plate characters in a complex background environment. Background technique [0002] In recent years, Intelligent Transportation System (ITS) is developing in full swing. It is an emerging concept that efficiently integrates advanced information collection and transmission technology, various intelligent sensing technologies, and computer processing technology into the field of traffic control. As the unique information of a car, the license plate corresponds to the vehicle one by one. It is the most effective way to identify traffic violation vehicles and fake license plate vehicles. Therefore, intelligent transportation systems have an urgent need for intelligent identification of license plate numbers. The method of automatic detection, positioning and recognition of...

AI Technical Summary

Problems solved by technology

[0003] The background environment of the current license plate recognition method is relatively simple, mostly used for vehicle identification in the access control system, while the license plate recognition requirements of the intelligent transportation system and the automatic detection, positioning and recognition of license plate characters in complex background environments are relatively few.

In the complex background environment, the currently used license plate recognition methods still have disadvantages such as being susceptible to background interference, low ability to resist background environment interference, slow running speed, and low recognition rate.

Images