Abstract:

In complex weather conditions,the distortion of the license plate imaging caused by the fog or glare makes it difficult to accurately identify plate information by the traditional image-based license plate identification method.By the Dark Channel Prior proposed by He Kaiming and histogram equalization method,the interference caused by weather conditions can be effectively eliminated,followed by image restoration and correction by morphology method,and finally,the license plate information can be recognized accurately based on the BP neural network.Experimental results show that the restored license images are clear and the recognition effect is good under the bad weather conditions.

Key words: image enhancement;binary morphology;image segmentation;BP neural network;license plate recognition