Unlock instant, AI-driven research and patent intelligence for your innovation.

A method for removing complex background noise interference of two-dimensional character graphic verification code

A technology of character graphics and complex backgrounds, applied in cross-fields, can solve problems such as difficult segmentation

Active Publication Date: 2019-03-22
BEIJING UNIV OF TECH
View PDF4 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, correct segmentation is an extremely difficult task due to interferences such as noise and adhesion during the design of captcha
Due to the variety of types of verification codes, no algorithm can be applied to all the divisions of graphic verification codes

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • A method for removing complex background noise interference of two-dimensional character graphic verification code
  • A method for removing complex background noise interference of two-dimensional character graphic verification code
  • A method for removing complex background noise interference of two-dimensional character graphic verification code

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0014] figure 1 It shows the effect of smoothing with EPSF filter and bilateral filter respectively. It can be seen that the edges of characters after EPSF filtering are still blurred, but the edges of bilateral filters are not smoothed.

[0015] The process of implementing this method is as follows:

[0016] S1 preprocessing

[0017] S1.1 Read in the verification code image and filter it with a bilateral filter; the basic idea of ​​the bilateral filter is to simultaneously consider the difference between the definition domain, that is, the pixel space neighborhood, and the value domain, that is, the pixel brightness space neighborhood. The value of the output pixel depends on the weighted sum of the neighboring pixel values, whose formula is as follows:

[0018]

[0019] Among them, g(i, j) represents the pixel brightness value of the i-th row and j-column of the output image; f(k, l) represents the pixel brightness value of the k-th row and l-column of the input image, ...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

A method for removing complex background noise interference of a two-dimensional character graphic verification code. After reading a verification code picture, the method performs a quantization operation on the background color, and by reducing the background color, the purpose of eliminating the cluttered background and highlighting the characters is achieved. . In the first smoothing step, an EPSF algorithm is designed to keep the edges from being blurred while smoothing; this algorithm uses a Gaussian kernel bilateral filter to replace the EPSF algorithm. Compared with the original EPSF algorithm, the Gaussian kernel The bilateral filter is better than the EPSF algorithm in smoothing the image while keeping the edge from being blurred, so that it can ensure that the character structure is not destroyed and lay the foundation for subsequent character recognition. When merging color classes, this algorithm adopts hierarchical clustering algorithm to merge colors adaptively. Practice shows that this algorithm can effectively remove the background noise of some verification codes in complex backgrounds.

Description

technical field [0001] The method of the invention belongs to the intersecting fields of computer vision and pattern recognition, and relates to a method for removing background noise of two-dimensional numbers or Chinese character verification code pictures based on color information. Background technique [0002] Verification codes (CAPTCHAs: Completely Automated Public Turing Test to TellComputers and Humans Apart, a fully automated public Turing test for distinguishing between humans and machines) are currently widely used by various websites to verify the login person and prevent hackers from automatically logging into the website A security defense method. There are various types of verification codes, which can be visual-based picture verification codes, or auditory verification codes, etc. Common visual verification codes include: verification codes based on numbers or characters, and verification codes based on natural scene graphs, such as: 12306 website's newly u...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Patents(China)
IPC IPC(8): G06K9/34G06K9/44
CPCG06V10/273G06V10/34G06V10/267G06V30/287
Inventor 李蓉
Owner BEIJING UNIV OF TECH