Captcha image authentication method and system

• US 8,607,331 B2
• Filed: 09/13/2011
• Issued: 12/10/2013
• Est. Priority Date: 07/15/2011
• Status: Active Grant

• Current Assignee: Industrial Technology Research Institute
• Original Assignee: Industrial Technology Research Institute
• Inventors: Sun, Hung-Min, Chen, Yao-Hsin, Yeh, Chun-Hao
• Primary Examiner(s): REVAK, CHRISTOPHER A
• Application Number:
  US13/230,924
• Publication Number:
  US 20130019278A1
• Time in Patent Office: 819 Days
• Field of Search: None
• US Class Current: 726/19
• CPC Class Codes: G06F 21/36 by graphic or iconic representation

Abstract

The present disclosure relates to a Completely Automated Public Turing Test to tell Computers and Humans Apart (CAPTCHA) image authentication method and system. The CAPTCHA image authentication method comprises the steps of: collecting a plurality of first objects; defining a plurality of variables so as to be used as basis for classifying and dividing the plural first objects into M groups accordingly while allowing each group in the M groups to correspond to at least one variable selected from the plural variables; selecting at least one group from the M groups while further grading and dividing the first objects in the selected group into subgroups of N grades based upon a standard unit of the variable corresponding to the selected group; sorting and storing the subgroups of N grades; and selecting a plurality of authentication objects from the subgroups of N grades to be used in an authentication process.

Claims

1. A Completely Automated Public Turing Test to tell Computers and Humans Apart (CAPTCHA) image authentication method, comprising the steps of:

collecting a plurality of first objects;

defining a plurality of variables so as to be used as basis for classifying and dividing the plurality of first objects into M groups accordingly while allowing each group in the M groups to correspond to at least one variable selected from the plurality of variables;

selecting one or more groups from the M groups while further grading and dividing the first objects in the selected one or more groups into subgroups of N grades based upon a standard unit of the variable corresponding to the selected one or more groups;

sorting and storing the subgroups of N grades for each of the one of more groups;

selecting a plurality of authentication objects from the subgroups of N grades for each of the one of more groups to be used in an authentication process;

defining a selection criteria for each of the one of more groups based on the plurality of variables corresponding to each group;

displaying the selection criteria and the plurality of authentication objects for each of the one of more groups on a display unit;

requesting a user to sort the plurality of authentication objects for each of the one of more groups according to the selection criteria; and

determining whether the plurality of authentication objects are sorted correctly for each of the one of more groups, and if so, the authentication process is determined to have passed, else selecting a new set of authentication objects for a future authentication process and the authentication process is determined to have failed.

2. The CAPTCHA image authentication method of claim 1, wherein each of the first objects is an object selected from the group consisting of:

a static picture, a dynamic image, or the combination of the two.

3. The CAPTCHA image authentication method of claim 1, wherein each of the plural variables is a measureable variable, and the plural variables may include size, speed, weight, height, volume, age, length, or area.

4. The CAPTCHA image authentication method of claim 1, further comprising the steps of:

defining a margin interval for each group to be used for defining a region evenly to the left and to the right from the border of any two neighboring subgroups of corresponding N grades for each of the one or more groups after being sorted, and accordingly removing a set of second objects, in those related neighboring subgroups that are located within corresponding margin intervals; and

selecting the plural authentication objects for each of the one of more groups from the subgroups of N grades for each of the one or more groups after having their corresponding second objects removed.

5. The CAPTCHA image authentication method of claim 1, further comprising the steps of:

randomly selecting P grades from the N grades of each of the one or more groups so as to select the plural authentication objects for each of the one or more groups from the subgroups relating to the selected P grades.

6. The CAPTCHA image authentication method of claim 1, further comprising the steps of:

after sorting the selected one or more groups into subgroups of N grades for each of the one or more groups, defining a margin interval to be used for defining a region evenly to the left and to the right from the border of any two neighboring subgroups of N grades for each of the one or more groups, and accordingly removing a set of second objects, in those related neighboring subgroups that are located within their corresponding margin intervals; and

randomly selecting P grades for each of the one or more groups from the N grades of each of the one or more groups so as to select the plural authentication objects for each of the one of more groups from the subgroups relating to the selected P grades for each of the one or more groups after having their corresponding second objects removed.

7. The CAPTCHA image authentication method of claim 4, wherein the margin interval is defined based upon a mathematics equation.

8. The CAPTCHA image authentication method of claim 7, wherein the margin interval is defined to be a present difference or a preset multiple between any two neighboring subgroups of N grades for each of the one of more groups.

9. The CAPTCHA image authentication method of claim 8, wherein the preset multiple is increasing or decreasing from a predesignated starting set for each of the one of more groups including two neighboring subgroups to the left or to the right to those other sets of two neighboring subgroups of each of the one of more groups.

10. The CAPTCHA image authentication method of claim 1, wherein the values of M and N are integral values.

11. The CAPTCHA image authentication method of claim 5, wherein the value of P is an integral value.

12. The CAPTCHA image authentication method of claim 1, wherein the selection criteria is any combination of texts, pictures and images.

13. A Completely Automated Public Turing Test to tell Computers and Humans Apart CAPTCHA image authentication system, comprising:

an acquisition unit, for collecting a plurality of first objects;

a processing unit, coupling to the acquisition unit to be used for defining a plurality of variables while allowing the plural variables to be used as basis for performing the following procedures;

classifying and dividing the plurality of first objects into M groups accordingly while allowing each group in the M groups to correspond to at least one variable selected from the plurality of variables;

selecting one or more groups from the M groups while further grading dividing the first objects in the selected one or more groups into subgroups of N grades based upon a standard unit of the variable corresponding to the selected one or more groups;

sorting and storing the subgroups of N grades for each of the one or more groups;

selecting a plurality of authentication objects from the subgroups of N grades for each of the one or more groups to be used in an authentication process;

defining a selection criteria for each of the one or more groups based on the plurality of variables corresponding to each group;

displaying the selection criteria and the plurality of authentication objects for each of the one or more groups on a display unit;

requesting a user to sort the plurality of authentication objects for each of the one or more groups according to the selection criteria; and

determining whether the plurality of authentication objects are sorted correctly for each of the one or more groups, and if so, the authentication process is determined to have passed, else selecting a new set of authentication objects for a future authentication process and the authentication process is determined to have failed; and

a storage unit, coupling to the processing unit to be used for storing the sorted subgroups of N grades for each of the one or more groups.

14. The CAPTCHA image authentication system of claim 13, wherein each of the first objects is an object selected from the group consisting of:

a static picture, a dynamic image, or the combination of the two.

15. The CAPTCHA image authentication system of claim 13, wherein each of the plural variables is a measureable variable, and the plural variables may include size, speed, weight, height, volume, age, length, or area.

16. The CAPTCHA image authentication system of claim 13, wherein after sorting the selected group into subgroups of N grades for each of the one or more groups, the processing unit is configured to perform the following procedure:

after sorting the selected one or more groups into subgroups of N grades for each of the one or more groups, defining a margin interval to be used for defining a region evenly to the left and to the right from the border of any two neighboring subgroups of N grades for each of the one or more groups, and accordingly removing a set of second objects, in those related neighboring subgroups that are located within their corresponding margin intervals; and

selecting the plural authentication objects for each of the one of more groups from the subgroups of N grades of each of the one or more groups after having their corresponding second objects removed.

17. The CAPTCHA image authentication system of claim 13, wherein after sorting the selected group into subgroups of N grades for each of the one or more groups, the processing unit is configured to perform the following procedure:

randomly selecting P grades from the N grades of each of the one or more groups so as to select the plural authentication objects for each of the one or more groups from the subgroups relating to the selected P grades.

18. The CAPTCHA image authentication system of claim 13, wherein after sorting the selected one or more groups into subgroups of N grades for each of the one or more groups, the processing unit is configured to perform the following procedure:

defining a margin interval to be used for defining a region evenly to the left and to the right from the border of any two neighboring subgroups of N grades for each of the one or more groups after being sorted, and accordingly removing each and every objects, referring as second objects, in those related neighboring subgroups that are located within their corresponding margin intervals; and

randomly selecting P grades from the N grades of each of the one or more groups so as to select the plural authentication objects from the subgroups relating to the selected P grades after having their corresponding second objects removed.

19. The CAPTCHA image authentication system of claim 16, wherein the margin interval is defined based upon a mathematics equation.

20. The CAPTCHA image authentication system of claim 19, wherein the margin interval is defined to be a present difference or a preset multiple between any two neighboring subgroups of N grades for each of the one or more groups.

21. The CAPTCHA image authentication system of claim 20, wherein the preset multiple is increasing or decreasing from a predesignated starting set of each of the one of more groups two neighboring subgroups to the left or to the right to those other sets of two neighboring subgroups for each of the one of more groups.

22. The CAPTCHA image authentication system of claim 13, wherein the values of M and N are integral values.

23. The CAPTCHA image authentication system of claim 17, wherein the value of P is an integral value.