A method for identifying the verification code of the nine-square grid pattern based on contour extraction

Abstract

The invention discloses a method for identifying a nine-square pattern verification code based on outline extraction. The present invention comprises the following steps: (1) performing binarization and Gaussian blur preprocessing on the input Jiugongge verification code containing arrows, and then performing noise reduction; (2) performing expansion processing on the picture, then performing canny contour detection, and calculating all The center of gravity of the closed contour; (3) Classify the extracted contours, calculate the number of Jiugong grid points and the position of the arrow according to the position and distribution of the center of gravity, and extract the minimum closed triangle from the contour of the non-overlapping arrow; The in-out degree of each Jiugongge mark point is obtained, and then the Hamiltonian path in the Jiugongge graphic verification code is calculated; (5) Any non-overlapping arrow is extracted, and the direction of the arrow is obtained by calculating the relationship between the triangle vertices and the center of gravity, as The Hamiltonian path indicates the direction, and finally outputs the verification result. The invention greatly improves recognition efficiency and user experience.

Description

technical field [0001] The invention belongs to the field of computer identification and detection, and in particular relates to a method for identifying a Jiugongge pattern verification code based on outline extraction. Background technique [0002] Jiugongge verification code is widely used in mobile communication, network media and Internet finance and other fields. Service providers help users confirm information by sending graphic verification codes to prevent accidental touches. However, the graphic verification code is time-consuming and laborious for the user to identify. It is necessary to switch the process to view the verification code on the mobile device, and many verification codes are difficult for users to recognize, and they often fail to pass the verification after entering multiple times, which greatly affects the user experience. , reducing the recognition efficiency. [0003] At present, the identification of graphic verification codes is carried out b...

AI Technical Summary

Problems solved by technology

However, the graphic verification code is time-consuming and laborious for the user to identify. It is necessary to switch the process to view the verification code on the mobile device, and many verification codes are difficult for users to recognize, and they often fail to pass the verification after entering multiple times, which greatly affects the user experience. , which reduces the recognition efficiency

[0003] At present, the identification of graphic verification codes is carried out by manual recognition, and there is no intelligent computer identification method, mainly because of the problems of complex picture background, many key elements and difficult control of recognition accuracy in graphic verification codes.

Using manual identification has obvious disadvantages, which greatly reduces the user experience for individual users; for corporate groups, manual identification of a large number of verification codes is extremely time-consuming, time-consuming and labor-intensive

Images