Character confidence degree-based secondary license plate identification method and apparatus

Abstract

The invention relates to the field of double dynamic license plate identification, and provides a character confidence degree-based secondary license plate identification method and apparatus for the problems existent in the prior art. The method comprises the steps of performing character identification through template matching; giving out confidence degrees of identification results; for the results with relatively low confidence degrees, performing video super-resolution processing to obtain a frame of high-quality image; and based on the image, performing secondary license plate identification through a neural network classifier. According to the method and the apparatus, a license plate confidence degree threshold Th is preset; a picture is captured from a front end and license plate locating and segmenting are performed; characters of a license plate are identified; character identification confidence degrees and license plate identification confidence degree are calculated; when the character confidence degrees are all higher than the threshold Th, a license plate identification result is directly given, otherwise, the video super-resolution processing is performed; a frame of high-quality image is obtained by utilizing time domain information; and based on the image, the to-be-identified characters are input to classifiers for performing identification according to a position relationship of the to-be-identified characters, so that a final license plate identification result is obtained.

Description

technical field [0001] The invention relates to the field of license plate recognition, in particular to a method and device for secondary license plate recognition of character confidence. Background technique [0002] As time goes by, the tentacles of "mobile electronic eyes" will penetrate into every corner of the city and play an increasingly important role in protecting our "safe city". A safe city is also an important measure of a city's modern management level, and an important measure to realize the security and stability of a city and even the entire country. Only by establishing a reasonable and effective urban video surveillance management system can the government management department discover problems at the first time and propose countermeasures and emergency plans. As an important information resource for smart cities, license plates have attracted much attention. [0003] When the license plate is captured by the social vehicle camera and each bayonet came...

AI Technical Summary

Problems solved by technology

The character recognition of the neural network has strong fault tolerance. As long as the sample selection is reasonable, large and diverse, the recognition rate of the trained classifier will be relatively high, but it depends too much on the sample, and the training time is relatively long, so it is not suitable for Occasions with strong real-time requirements

Moreover, it is difficult to determine the number of middle layers of the three-layer neural network. Generally, it depends on empirical values. Increasing the number can improve the recognition rate, but the time is too long.

Another method of recognition, template matching, is simple in calculation and fast in recognition speed, but it is difficult to meet the requirements of recognizing low-quality pictures (such as blurred pictures, missing strokes, etc.). For similar characters (such as 8, 9 and B, etc.) the recognition rate is very low

Images