DECONVOLUTION OF DIGITAL IMAGES

• US 20120087551A1
• Filed: 10/12/2010
• Published: 04/12/2012
• Est. Priority Date: 10/12/2010
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Bhagwan, Varun, Liu, Timothy, Ormont, Justin, Underwood, Heather
• Granted Patent: US 8,792,748 B2
• Time in Patent Office: Days
• US Class Current: 382/118
• CPC Class Codes: G06K 7/146 the method including quality enhancement steps
• CPC Class Codes: G06T 2207/10004 Still image; Photographic image
• CPC Class Codes: G06T 2207/10024 Color image
• CPC Class Codes: G06T 3/10 Selection of transformation methods according to the characteristics of the input images
• CPC Class Codes: G06T 3/14 Transformations for image registration, e.g. adjusting or mapping for alignment of images
• CPC Class Codes: G06T 5/73 Deblurring; Sharpening
• CPC Class Codes: G06V 10/20 Image preprocessing
• CPC Class Codes: G06V 30/10 Character recognition
• CPC Class Codes: G06V 30/1607 Correcting image deformation, e.g. trapezoidal deformation caused by perspective
• CPC Class Codes: G06V 40/166 using acquisition arrangements

Abstract

A method for deconvolution of digital images includes obtaining a degraded image from a digital sensor, a processor accepting output from the digital sensor and recognizing a distorted element within the image. The distorted element is compared with a true shape of the element to produce a degrading function. The degrading function is deconvolved from at least a portion of the image to improve image quality of the image. A method of indirectly decoding a barcode includes obtaining an image of a barcode using an optical sensor in a mobile computing device, the image comprising barcode marks and a textual character. The textual character is optically recognized and an image degrading characteristic is identified from the textual character. Compensating for the image degrading characteristic renders previously undecodable barcode marks decodable. A system for deconvolution of digital images is also included.

Claims

1. A method for deconvolution of digital images comprising:

obtaining a degraded image from a digital sensor, a processor accepting output of the digital sensor;

recognizing a distorted element within the degraded image using the processor;

comparing the distorted element with a true shape of the element to produce a degrading function; and

deconvolving the degrading function from at least a portion of the degraded image to improve image quality of the degraded image.

2. The method of claim 1, in which recognizing a distorted element comprises optical character recognition of text within the degraded image.

3. The method of claim 1, in which comparing the distorted element with a true shape of the element comprises comparing an optically recognized character with a portion of the degraded image from which the optically recognized character was extracted.

4. The method of claim 1, in which degraded image comprises a bar code and associated text.

5. The method of claim 4, in which:

recognizing a distorted element comprises optical character recognition of text within the degraded image; and

deconvolving the degrading function comprises deconvolving a portion of the degraded image containing the barcode.

6. The method of claim 5, further comprising:

decoding the barcode; and

comparing the optically recognized text and decoded barcode information.

7. The method of claim 6, further comprising verifying decoded information from the barcode using optically recognized text.

8. The method of claim 1, in which the degrading function quantifies an image blur path and distance.

9. The method of claim 1, in which recognizing a distorted element comprises facial recognition of a face within the degraded image.

10. The method of claim 1, in which a first degrading function is calculated at a first location in the degraded image and a second degrading function is calculated at a second location within the degraded image.

11. The method of claim 10, in which deconvolution of portions of the image between the first location and second location are made by interpolating the first degrading function and the second degrading function.

12. The method of claim 10, further comprising extrapolating from the first and second degrading functions to cover areas of the degraded image which are beyond both the first location and second location.

13. The method of claim 1, in which the degraded image is a frame in a video sequence, the deconvolution of the degrading function being applied to subsequent frames in the video sequence.

14. The method of claim 13, in which a first degrading function is calculated for a first frame in the video sequence and a second degrading function is calculated for a later frame in the video sequence, the deconvolution of the frames between the first frame and the later frame being based on an interpolation between the first degrading function and the second degrading function.

15. A method of indirectly decoding a barcode comprises:

obtaining an image of a barcode using an optical sensor in a mobile computing device, the image comprising barcode marks and textual characters;

optically recognizing a textual character;

identifying an image degrading characteristic from the textual character; and

compensating for the image degrading characteristic to render previously undecodable barcode marks decodable.

16. The method of claim 15, further comprising:

overlaying the optically recognized textual character on portions of the image from which the optically recognized textual character was extracted; and

comparing the optically recognized textual character with the portions of the image from which the optically recognized textual character was extracted to produce a degrading function which represents the image degrading characteristic.

17. The method of claim 16, in which comparing the optically recognized textual character with the portions of the image from which the optically recognized textual character was extracted comprises performing a fast Fourier transform.

18. The method of claim 15, in which identifying an image degrading characteristic comprises quantifying the image degrading function as a point spread function.

19. The method of claim 15, in which the textual character is associated with the barcode marks and contains the same information that is encoded in the barcode marks.

20. The method of claim 18, in which information decoded from the barcode marks is verified against the optically recognized textual character.

21. A system for deconvolution of digital images comprising:

an optical sensor for acquiring an image of a scene;

an element identification module for identification of distorted elements within the image;

a degrading function calculation module for calculating a degrading function from the difference between portions of the image which contain distorted elements and undistorted elements obtained from a known element library; and

a deconvolution module for deconvolving the degrading function from at least a portion of the image and outputting an enhanced image.

22. The system of claim 21, in which the element identification module comprises an optical character recognition and the known element library comprises alpha numeric characters in a plurality of fonts.

23. The system of claim 21, in which the element identification module comprises a facial recognition function and the known element library comprises known faces.

24. A computer program product for image enhancement by deconvolution, the computer program product comprising:

a computer readable storage medium having computer readable program code embodied therewith, the computer readable program code comprising;

computer readable program code configured to recognize a distorted element within an image;

computer readable program code configured to compare the distorted element with a corresponding undistorted element to produce a degrading function; and

computer readable program code configured to deconvolve the degrading function from at least a portion of the image to improve the image quality.