License plate detection method based on convolutional neural network

Abstract

The invention discloses a license plate detection method based on a convolutional neural network. The method specifically includes the steps that an Adaboost license plate detector based on Haar characteristics detects license plate images to be detected, license plate roughing regions are acquired, a convolutional neural network complete license plate recognition model recognizes the license plate roughing regions, a final license plate candidate region is acquired, the final license plate candidate region is segmented through a multi-threshold segmentation algorithm, license plate Chinese characters, letters and numbers are acquired, a Chinese character, letter and number convolutional neural network recognition model recognizes the license plate Chinese characters, letters and numbers, and then a license plate recognition result is acquired. License plate images under different conditions can be accurately recognized through the Adaboost license plate detector based on the Haar characteristics and the convolutional neural network complete license plate recognition model, meanwhile, characters are segmented through the multi-threshold segmentation algorithm, character images can be more easily and conveniently segmented, and the good effect is achieved in engineering application.

Description

technical field [0001] The invention belongs to the technical field of computer vision recognition, and in particular relates to a license plate detection method based on a convolutional neural network. Background technique [0002] As a part of smart city and digital city, intelligent transportation system is mainly used for traffic flow monitoring, vehicle monitoring, highway toll station management, community intelligent management, parking lot management, traffic police law enforcement, etc., among which license plate detection is the most critical of intelligent transportation system It is an important link to realize the intelligentization of traffic management, and it is an important research topic of computer vision and pattern recognition technology in intelligent transportation system. License plate recognition technology is used in vehicle crossing and bridge crossing automatic non-stop toll collection, measurement of traffic flow control indicators, automatic veh...

AI Technical Summary

Problems solved by technology

Due to the differences in license plates in different regions at home and abroad, the portability of related products is poor, so the development of license plate detection and recognition technology suitable for mainland China is an urgent issue to be solved.

The license plate detection and recognition methods used by products on the existing market mainly face three major problems: 1) When the color of the vehicle body and the license plate color are similar, the license plate area cannot be accurately determined; 2) The current license plate character recognition based on character strokes and other methods are relatively Complicated and not very accurate

3) When the license plate image is unevenly illuminated, the license plate characters cannot be better segmented

But in fact this invention has the following three disadvantages: 1) simply rely on color and texture space features to locate the license plate with poor anti-interference; It is not easy to grasp the relevant color feature parameter values; 3) When screening the texture features of the license plate area, the number of jumps in each line is used to filter the license plate area, which does not have a good effect when the license plate area is unevenly illuminated

But in fact, this invention is relatively troublesome when super-resolution reconstruction is performed. On the other hand, the recognition method based on character strokes commonly used in the industry is adopted for character recognition. High, when the image is polluted, the recognition effect cannot be guaranteed, such as the recognition of confusing characters

Images