License plate recognition method in haze environment

• Main Authors: CUI XIN, HU XIAOWAN, ZHANG MINFENG, WANG LIHE
• Format: Patent
• Language: Chinese; English
• Published: 03.06.2022
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; IMAGE DATA PROCESSING OR GENERATION, IN GENERAL; PHYSICS

The invention provides a license plate recognition method in a haze environment, and aims to reduce the influence of haze on the boundary and characteristics of a license plate in an image in the haze environment so as to correctly recognize the license plate. Firstly, a dark channel prior algorithm based on atmospheric light value local estimation is used to carry out preliminary defogging on a blurred image; and then the image is further defogged, and on the basis of the target detection convolutional neural network, the license plate area is detected through the joint defogging and area extraction model JFRM, so that the distortion of image restoration is greatly reduced, and the image with the defogged license plate is accurately extracted. And finally, the image super-resolution is realized through the convolution enhanced SRCNN, so that the license plate characters can be successfully recognized. The accuracy of license plate recognition in the haze environment is improved. 本发明提供了一种雾霾环境下的车牌识别方法，其目的在于在雾霾