A fuzzy license plate recognition method based on multi-camera

Abstract

The invention relates to a fuzzy license plate recognition method based on a multi-eye camera, comprising the following steps: 1) collecting multi-frame license plate images at one time through the multi-eye camera; 2) segmenting each frame image collected and classifying the image by using a support vector machine method to obtain a license plate image block; 3) carrying out Gaussian filtering and grayscale operation on that license plate image block to obtain the denoised license plate grayscale image block; 4) carrying out super-resolution reconstruction on that gray image block of the license plate to obtain a clear super-resolution license plate which is convenient for recognition; 5) a convolution neural network being used to recognize the character of the clear license plate after the super-division operation, and the final recognition result being obtained. Compared with the prior art, the invention improves the recognition rate of the resolution blurred license plate, effectively improves the recognition accuracy of the low-resolution blurred license plate photographed under the environment of long distance and long focal length, and meets the recognition demand of the license plate.

Description

technical field [0001] The invention relates to the technical field of intelligent traffic control, in particular to a multi-camera-based fuzzy license plate recognition method. Background technique [0002] License plate recognition is a technology that improves the automation and intelligence of traffic control systems based on computer, image processing, and pattern recognition technologies. Nowadays, license plate recognition technology is blooming everywhere and widely used in commercial and law enforcement departments, such as parking lots, residential areas, bridges and highway toll stations, highway flow monitoring stations and other actual traffic systems for authenticating license plates. [0003] Most of the existing license plate recognition systems on expressways use a camera to shoot road vehicles. Due to the long shooting distance and low resolution of the camera, the recognition rate is low, and the captured pictures cannot recognize clear license plate chara...

AI Technical Summary

Problems solved by technology

[0003] Most of the existing license plate recognition systems on expressways use a camera to shoot road vehicles. Due to the long shooting distance and low resolution of the camera, the recognition rate is low, and the captured pictures cannot recognize clear license plate characters. , which has a greater impact on capturing illegal vehicles

Images