Method and a system for generating character adversarial verification code based on frequency domain noise addition

Abstract

The invention discloses a method for generating a character adversarial verification code based on frequency domain noise addition. The method comprises the following steps of generating a string of character sequences by a random algorithm, and respectively converting characters in the character sequences into a character image containing the single corresponding character in sequence; convertingthe character image from a spatial domain to a frequency domain, adding adversarial noise to the character image of the frequency domain according to an adversarial learning method, and converting the character image added with the adversarial noise of the frequency domain from the frequency domain to the spatial domain; and splicing the character images added with the adversarial noise togetheraccording to the sequence of verification code answers in order to form the character adversarial verification code. The invention further discloses a system for generating the character adversarial verification code based on the frequency domain noise addition. On the basis of the adversarial learning method, the adversarial noise is added to the character images at the frequency domain, and thegenerated character adversarial verification code can effectively avoid or reduce the image preprocessing adopted by an attacker, so that the safety of the character verification code is improved.

Description

Technical field [0001] The invention relates to the field of machine learning model security and counter verification code security, in particular to a method and system for generating character counter verification codes based on frequency domain noise addition. Background technique [0002] The verification code technology is mainly to be able to distinguish whether the current user is a normal human or a machine. Character verification code, as a relatively common form, has good application value in practical applications such as website login and email verification. [0003] The usual character verification code design needs to follow two principles: 1) Human recognizability: Since the main function of the verification code is to distinguish whether the user is a human or a machine, a human user must be able to easily identify the content in the verification code. This requires the characters in the character verification code not to be too much interference, otherwise it will...

AI Technical Summary

Problems solved by technology

This requires that the characters in the character verification code cannot be disturbed too much, otherwise it will greatly reduce the experience of real users

2) Machine unrecognizability: the characters in the character verification code need to be difficult to be cracked by automatic programs

Therefore, it is unwise to directly abandon the research of character verification codes

Images