Character verification code recognition model training method and system, character verification code recognition model recognition method and system, equipment and medium

Abstract

The invention provides a character verification code recognition model training method, a recognition method and system, equipment and a medium. The method comprises the steps: building a standard character class library which comprises a plurality of characters and class vectors corresponding to the characters; acquiring a plurality of character verification code sample images, and naming characters contained in the corresponding images; obtaining a target position offset of each character in each character verification code sample image through a character position prediction model; matchingcharacters in the names of the character verification code sample images with characters in a standard character category library to obtain target category vectors of the characters in the characterverification code sample images; and training the character verification code recognition model according to the target category vector and the target position offset of each character in each character verification code sample image to obtain a target character verification code recognition model. According to the invention, the accuracy and efficiency of character verification code recognition can be improved, and the generation efficiency of training samples is improved.

Description

technical field [0001] The present invention relates to the field of deep learning, in particular to a character verification code recognition model training method, recognition method, system, equipment and media. Background technique [0002] Captcha is a public fully automatic procedure to distinguish whether a user is a computer or a human. Character verification codes are widely used in Internet services as a tool to determine whether network requests come from legitimate users, so as to prevent a large number of automatic requests from machines and ensure the stable operation of website servers. Character verification codes are currently the most commonly used type of verification codes. This type of verification code usually requires the user to complete a character recognition task, and the user needs to correctly identify each character in the character image generated by computer image technology to pass the verification. In order to increase the difficulty of ma...

AI Technical Summary

Problems solved by technology

Due to the large number of strokes and complex line structure of Chinese characters, it is difficult for traditional image recognition methods to effectively separate the character foreground from the interference background, resulting in a low recognition success rate and a long time-consuming

In addition, traditional image recognition methods often rely on a large amount of labeled data for training. Each image is manually labeled and verified. The whole process is time-consuming, laborious, and expensive.

Images