Identifying image transformations for improving optical character recognition quality

1. A method comprising:

producing, by a computer system, an ordered list of transformations to be applied to an image representing a character string, each transformation corresponding to a hypothesis asserted with respect to one or more characteristics of the image;

applying, to the image, a leading transformation on the list to produce a transformed image;

evaluating a quality of the transformed image to produce a quality estimate; and

updating the list in view of the quality estimate,

wherein evaluating the quality of the transformed image comprises performing an optical character recognition of the text comprises by the image.

2. The method of claim 1 , wherein each transformation of the list of transformations comprises one or more operations to be performed on the image.

3. The method of claim 1 , further comprising identifying the image as a fragment of a larger image comprising a textual block.

4. The method of claim 1 , wherein producing the ordered list of transformations comprises evaluating, in view of values of one or more parameters of the image, one or more criteria associated with a plurality of transformations.

5. The method of claim 1 , wherein updating the list comprises:

updating, in view of the quality estimate, integrated scores assigned to one or more transformations; and

selecting a new leading transformation in view of the updated integrated scores.

6. The method of claim 5 , wherein selecting the new leading transformation comprises applying one or more differential rules to compare at least two transformations on the list.

7. The method of claim 1 , wherein updating the list comprises at least one of: ordering the list, adding a transformation to the list, or removing a transformation from the list.

8. The method of claim 1 , further comprising: responsive to determining that the quality estimate fails short of a certain threshold value, repeating the operations of applying, evaluating, and updating.

9. The method of claim 1 , further comprising:

responsive to determining that the quality estimate exceeds a certain threshold value, associating, with the image, a text produced by the optical character recognition.

10. A computing device, comprising:

a memory;

a processor, coupled to the memory, the processor configured to: produce an ordered list of transformations to be applied to an image representing a character string, each transformation corresponding to a hypothesis asserted with respect to one or more characteristics of the image;

apply, to the image, a leading transformation on the list to produce a transformed image;

evaluate a quality of the transformed image to produce a quality estimate; and

update the list in view of the quality estimate,

wherein evaluating the quality of the transformed image comprises performing an optical character recognition of the text comprises by the image.

11. The computing device of claim 10 , wherein producing the ordered list of transformations comprises evaluating, in view of values of one or more parameters of the image, one or more criteria associated with a plurality of transformations.

12. The computing device of claim 10 wherein updating the list comprises:

updating, in view of the quality estimate, integrated scores assigned to one or more transformations; and

selecting a new leading transformation in view of the updated integrated scores.

13. The computing device of claim 12 , wherein selecting the new leading transformation comprises applying one or more differential rules to compare at least two transformations on the list.

14. A computer-readable non-transitory storage medium comprising executable instructions that, when executed by a computing device, cause the computing device to perform operations comprising:

producing, an ordered list of transformations to be applied to an image representing a character string, each transformation corresponding to a hypothesis asserted with respect to one or more characteristics of the image;

applying, to the image, a leading transformation on the list to produce a transformed image:

evaluating a quality of the transformed image to produce a quality estimate; and

updating the list in view of the quality estimate,

wherein evaluating the quality of the transformed image comprises performing an optical character recognition of the text comprises by the image.

15. The computer-readable non-transitory storage medium of claim 14 , wherein updating the list comprises:

updating, in view of the quality estimate, integrated scores assigned to one or more transformations: and

selecting a new leading transformation in view of the updated integrated scores.

16. The computer-readable non transitory storage medium of claim 14 , wherein updating the list comprises at least one of:

ordering the list, adding a transformation to the list, or removing a transformation from the list.

17. The computer-readable non-transitory storage medium of claim 14 , further comprising executable instructions causing the computing device to perform operations comprising:

responsive to determining that the quality estimate falls short of a certain threshold value, repeating the operations of applying, evaluating, and updating.

18. The computer-readable non-transitory storage medium of claim 14 , further comprising executable instructions causing the computing device to perform operations comprising:

responsive to determining that the quality estimate exceeds a certain threshold value, associating, with the image, a text produced by the optical character recognition.