Automated removal of narrow, elongated distortions from a digital image

• US 6,266,054 B1
• Filed: 11/05/1997
• Issued: 07/24/2001
• Est. Priority Date: 11/05/1997
• Status: Expired due to Term

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Qadir, Mahmood G., Lawton, Daryl T.
• Primary Examiner(s): Huynh, Ba
• Application Number:
  US08/964,629
• Time in Patent Office: 1,357 Days
• Field of Search: 382/22, 382/115-118, 382/164-167, 382/190-196, 382/199, 382/202, 382/205, 382/203, 382/254-275, 382/307-311, 382/323, 345/326
• US Class Current: 345/581
• CPC Class Codes: G06T 5/80 Geometric correction

Abstract

An apparatus including a method of digital image processing which allows a user to remove narrow, elongated areas of image distortion from a digitized image. According to the method, the user selects a portion of an image to be edited and, once the portion has been identified, the image distortion within the portion is automatically removed. The procedure for removing the distortion includes the steps of delineating the distortion within the portion of the image selected and replacing the delineated area with a correction image which is created as a function of the attributes of the neighboring portions of the image.

Claims

1. In a computer system, a method of removing from a digital image a narrow, elongated region of image distortion having a primary orientation, the method comprising the steps of:

identifying the primary orientation;

identifying and storing in a memory of the computer system a set of attributes of a portion of the digital image, the attributes comprising a plurality of substantially one-dimensional pixel slices oriented by reference to the primary orientation;

delineating the image distortion within the portion of the digital image as a function of the attributes stored in the memory; and

removing the delineated image distortion utilizing a pixel replacement process.

2. The method as recited in claim 1, further comprising the step of allowing a user of the computer system to select the portion of the digital image.

3. The method as recited in claim 1, further comprising the steps of displaying the digital image on a display device of the computer system and allowing the user to select the portion of the digital image by interacting with the displayed digital image.

4. The method as recited in claim 3, wherein a movable selection rectangle under the control of the user is displayed on the display device which the user employs to interact with the displayed digital image.

5. The method as recited in claim 4, wherein the selection rectangle has a width and a length which are adjustable by the user.

6. The method as recited in claim 3, wherein a sequence of movable selection rectangles under the control of the user are displayed on the display device which the user employs to interact with the displayed digital image.

7. The method as recited in claim 6, further comprising the step of causing the sequence of movable selection rectangles to automatically overlap.

8. The method as recited in claim 1, wherein the step of delineating the distortion comprises the step of evaluating each slice to determine if a scratch-pixel-sequence exists therein.

9. The method as recited in claim 8, wherein the step of delineating the distortion within each slice further comprises the step of redefining the scratch-pixel-sequence for each slice as a function of the scratch-pixel-sequence of neighboring slices.

10. The method as recited in claim 1, wherein the pixel replacement process comprises the steps of locating within the delineated distortion a pixel to be removed, establishing a local neighborhood of pixels around the pixel to be removed, determining from the values of the neighborhood of pixels values for a replacement pixel, and replacing the pixel to be removed with the replacement pixel.

11. The method as recited in claim 10, further comprising the steps of storing the replacement pixel within a digital correction image and thereafter compositing the digital correction image with the digital image.

12. The method as recited in claim 10, wherein only those pixels within the neighborhood of pixels which are located outside of the portion of the image selected and generally perpendicular to a major axis of the portion of the image selected, located within the portion of the image selected and not within the delineated distortion, and located within the portion of the image selected and within the delineated distortion but previously replaced with the replacement pixel are utilized in determining the values for the replacement pixel.

13. The method as recited in claim 10, wherein only those pixels within the neighborhood of pixels which lie outside a predetermined distance from a mean value of those pixels which comprise the delineated distortion are utilized in determining the values for the replacement pixel.

14. The method as recited in claim 10, wherein only those pixels within the neighborhood of pixels which meet a predefined criteria are utilized in determining the values for the replacement pixel.

15. A computer-readable medium having computer-executable instructions for performing a method of removing a narrow, elongated region of image distortion from a digital image, the method steps comprising:

identifying a primary orientation of the image distortion;

identifying and storing in a memory of a computer system a set of attributes of a portion of the digital image resident on the computer system, the attributes comprising a plurality of substantially one-dimensional pixel slices oriented by reference to the primary orientation;

delineating a narrow, elongated region of image distortion within the portion of the digital image as a function of the attributes stored in the memory; and

removing the delineated image distortion utilizing a pixel replacement process.

16. The computer-readable medium as recited in claim 15, further performing the steps of displaying the digital image on a display device associated with the computer system and identifying the portion of the digital image in response to a user interacting with the displayed digital image.

17. The computer-readable medium as recited in claim 15, wherein the step of delineating the image distortion comprises the step of evaluating each slice to determine if a scratch-pixel-sequence exists therein.

18. The computer-readable medium as recited in claim 15, wherein the pixel replacement process comprises the steps of locating within the delineated distortion a pixel to be removed, establishing a local neighborhood of pixels around the pixel to be removed, determining from the values of the neighborhood of pixels values for a replacement pixel, and replacing the pixel to be removed with the replacement pixel.

19. A computer system comprising:

a memory on which is stored a digital image comprising a plurality of pixels and having a narrow-elongated region of visual distortion;

a user interface for allowing a user to identify a pixel area of the image which includes the visual distortion; and

a tool for automatically removing the visual distortion from within the pixel area, the tool comprising

     1) a row/column slicer for identifying a primary orientation of the visual distortion and for creating a plurality of substantially one-dimensional pixel slices oriented by reference to the primary orientation;

     2) an extraction filter for automatically identifying and collecting attributes of the pixel slices and identifying a subset of the pixel area which defines the visual distortion as a function of the collected attributes; and

     3) a replacement pixel value calculator for determining values for use in forming a corrected subset of the pixel area based, at least in part, on the collected attributes, wherein the subset of the pixel area is replaced with the corrected subset of the pixel area resulting in the effective removal of the visual distortion.

20. The computer system as recited in claim 19, further comprising a display device for use in displaying the digital image and wherein the user interface further comprises an input device for allowing the user to select the pixel area by interacting with the digital image when it is displayed on the display device.

21. The computer system as recited in claim 19, wherein the extraction filter identifies the subset of the pixel area by evaluating the attributes of each slice to determine if a scratch-pixel sequence exists therein.

22. The computer system as recited in claim 19, further comprising an iterator and a compositor, wherein the iterator locates a pixel within the subset of the pixel area which is to be removed, the replacement pixel value calculator establishes a neighborhood of pixels around the pixel to be removed and determines an image value for a replacement pixel based upon image values of the neighborhood of pixels, and the compositor composites the replacement pixel with the pixel to be removed.

23. A method for automatically removing from a digital image an unwanted region, the method comprising the steps of:

identifying a portion of the digital image containing at least a portion of the unwanted region;

identifying the primary orientation of the portion of the unwanted region contained within the identified image portion;

storing in a memory of a computer system a set of attributes of the image portion, the attributes comprising pixel value attributes of a plurality of substantially one-dimensional pixel slices oriented by reference to the primary orientation of the portion of the unwanted region;

identifying the unwanted region portion within the image portion using the set of attributes stored in the memory;

extracting the portion of the unwanted region;

repairing the extracted unwanted region portion using information from a set of attributes describing visual characteristics of an area of a portion of the digital image surrounding the unwanted region; and

, compositing the repaired region and the digital image so that the repaired region replaces the unwanted region.