Man-machine collaboration method and system for identifying verification code

Abstract

The invention discloses a verification code identification method and system in a man-machine collaborative manner. The method includes: an automatic verification code classification module obtains verification code data to be identified; the verification code automatic classification module performs type identification on the verification code data; verification The code task distribution module obtains a list of currently waiting and idle coding clients; according to the type of verification code data and the client list; Identify; perform coding; the feedback module calculates the coding time to update the average recognition speed of the client for this type of verification code; performs similarity calculation on the coding results of different coding clients. The system includes a verification code automatic classification module, a verification code task distribution module, a coding client and a feedback module. The method and the system thereof in the present invention can effectively reduce labor cost consumption and can effectively ensure the accuracy of verification code identification.

Description

technical field [0001] The invention relates to the technical field of verification code identification, in particular to a verification code identification method and system in a man-machine collaborative manner. Background technique [0002] Captcha is an effective way to distinguish service users from humans and automated programs, and has made a great contribution to the system's defense against malicious behaviors such as spam and brute force. Although the verification code provides a great guarantee for the security of the system, it also sets a great obstacle for developers to explore automated applications. For example, the information collection of complex websites has to use a third-party verification code identification (coding) service To bypass the verification code authentication link, the verification code recognition service is gradually indispensable in today's automated development link. [0003] Existing third-party verification code identification servic...

AI Technical Summary

Problems solved by technology

In addition, although the existing verification code recognition platforms all have the concept of "task difficulty", that is, the coding client can actively choose the tasks they want to perform according to the task difficulty, but there is no specific difficulty quantification standard, so the system cannot Estimate the recognition effect of the client on the verification code of the target system, so there will be a "remedial" human-computer cooperation coding method

Images