CONTEXTUAL ABNORMALITY CAPTCHAS

• US 20110113378A1
• Filed: 11/09/2009
• Published: 05/12/2011
• Est. Priority Date: 11/09/2009
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Boden, Edward B., Greenblatt, Michael L.
• Granted Patent: US 8,495,518 B2
• Time in Patent Office: Days
• US Class Current: 715/837
• CPC Class Codes: G06F 21/36 by graphic or iconic representation

Abstract

A method, computing device, and computer program product are provided for automatically generating contextual image tests implemented on a computer. An image is received by a computer. Sections of the image are altered to appear out of context or wrong to human users. The user must then identify the errors to pass the CAPTCHA.

Claims

1. A method for automatically generating contextual image tests implemented on a computer, comprising:

receiving, by a computer, an image;

dividing, by the computer, the image into objects such that the objects can be selected;

selecting, by the computer, for replacement an original object from the objects; and

replacing, by the computer, the original object with a similarly shaped replacement object, the original object being swapped for the replacement object in the image, wherein the replacement object is an abnormality in the image.

2. The method of claim 1, further comprising:

recognizing a selection of the replacement object as a correct answer.

3. The method of claim 1, wherein the replacement object is the abnormality because the replacement object is configured not to adhere to a context of the image.

4. The method of claim 1, wherein replacing the original object with the replacement object causes the image to visually appear abnormal to a human within a context of the image.

5. The method of claim 1, wherein the image is configured to allow a user to select the replacement object in the image by clicking on the replacement object.

6. The method of claim 1, wherein the selecting and replacing steps comprise:

selecting a remove object from the objects, the remove object being in an area on the image;

cloning an area around the remove object to create a clone replacement from the area; and

replacing the remove object with the clone replacement in the image.

7. The method of claim 6, wherein the clone replacement is created to visually blend in with the area.

8. The method of claim 6, wherein the clone replacement is selectable as a correct answer.

9. The method of claim 6, wherein the clone replacement is configured as an abnormality in the image.

10. The method of claim 6, further comprising:

removing the remove object from the area such that the absence of the remove object visually appears as an abnormality within a context of the image.

11. The method of claim 6, wherein replacing the remove object with the clone replacement in the image causes the image to visually appear abnormal to a human within a context of the image.

12. The method of claim 6, wherein the image is configured to allow a user to select the clone replacement in the image by clicking on the clone replacement.

13. A computing device configured to automatically generate contextual image tests, comprising:

memory for storing a program; and

a processor, functionally coupled to the memory, the processor being responsive to computer-executable instructions contained in the program and operative for;

receiving an image;

dividing the image into objects such that the objects can be selected; and

selecting for replacement an original object from the objects; and

replacing the original object with a similarly shaped replacement object, the original object being swapped for the replacement object in the image, wherein the replacement object is an abnormality in the image.

14. The device of claim 13, wherein the processor is further operative for:

recognizing a selection of the replacement object as a correct answer.

15. The device of claim 13, wherein the replacement object is the abnormality because the replacement object is configured not to adhere to a context of the image.

16. The device of claim 13, wherein replacing the original object with the replacement object causes the image to visually appear abnormal to a human within a context of the image.

17. The device of claim 13, wherein the image is configured to allow a user to select the replacement object in the image by clicking on the replacement object.

18. The device of claim 13, wherein the selecting and replacing comprise:

selecting a remove object from the objects, the remove object being in an area on the image;

cloning the area around the remove object to create a clone replacement of the area; and

replacing the remove object with the clone replacement in the image.

19. The device of claim 18, wherein the clone replacement is created to visually blend in with the area.

20. The device of claim 18, wherein the clone replacement is selectable as a correct answer.

21. The device of claim 18, wherein the clone replacement is configured as an abnormality in the image;

or

wherein replacing the remove object with the clone replacement in the image causes the image to visually appear abnormal to a human within a context of the image.

22. The device of claim 18, wherein the processor is further operative for:

removing the remove object from the area such that the absence of the remove object visually appears as an abnormality within a context of the image.

23. A method for automatically generating contextual image tests implemented on a computer, comprising:

receiving, by the computer, an image;

selecting, by the computer, one or more image parts of the image for change based on image analysis;

determining, by the computer, modifications for the selected one or more image parts and a modification source for each selected image part; and

applying, by the computer, the modifications to the one or more image parts to create a modified image.

24. The method of claim 23, wherein the modifications for the selected image part comprise:

replacing the selected image part;

changing a color of the selected image part;

changing a color gradient of the selected image part;

changing a size of the selected image part;

duplicating the selected image part;

changing an orientation with respect to image axes for the selected image part;

changing a texture of the selected image part;

or

adding an object at a location of the selected image part.

25. The method of claim 24, wherein adding the object at the location of the selected image part comprises inserting the object in the image;

and

wherein the object is duplicated from the image as the modification source or the object is copied from a different image as the modification source.