Research on the Detection and Recognition Algorithm of Click Chinese Character Verification Code

Network security has always been a hot topic of people’s attention. In recent years, click Chinese character verification codes have been used by major Chinese websites to prevent malicious attacks and malicious batch operations. How effective this verification code is and whether it will be cracked at a low cost is a worthy of discussion and research question. In response to this problem, this paper proposes a two-step cracking method of detection and classification. Using the more commonly used detection and classification networks, combined with some targeted changes, achieved an 83.33% cracking success rate on the test set at a relatively low cost, and made a partial answer to this question.