Fuzzy image comparator

• US 7,477,785 B2
• Filed: 11/19/2004
• Issued: 01/13/2009
• Est. Priority Date: 11/19/2004
• Status: Expired due to Fees

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Cauchy, Marc, Reissman, Pierre-Jean
• Primary Examiner(s): Desire; Gregory M
• Application Number:
  US10/993,538
• Publication Number:
  US 20060110047A1
• Time in Patent Office: 1,516 Days
• Field of Search: 348/246, 348/250, 382/168, 382/191, 382/218, 382/221, 382/260, 382/275, 382/278, 382/280, 382/286, 382/192
• US Class Current: 382/221
• CPC Class Codes: G06T 7/001 using an image reference approach
• CPC Class Codes: G06V 10/751 Comparing pixel values or logical combinations thereof, or feature values having positional relevance, e.g. template matching

Abstract

A system and method for improved image comparison is provided. An adjustable tolerance level may be set for comparing the images, and filtering may be applied to the images, for example, to suppress high frequencies in the images or to enhance low frequencies in the images. Furthermore, a geometrical transform, such as a 2D affine transform, may be applied to one of the images. The pixels of one of the images may be compared with corresponding pixels of another image to detect differences between the pixels compared. Then the outcome of comparing the pixels of the images may be determined by comparing the differences detected with the tolerance level set. The tolerance may be set by selecting one of a set of predefined tolerance levels, each of which may represent a different degree of tolerance to be used in comparison of the images.

Claims

1. A method for comparing images in a computer system, comprising:

preprocessing images to be compared, including setting an adjustable tolerance level for determining the results of comparing the images;

using the computing system to perform the following;

comparing pixels of one of the images with corresponding pixels of another image to detect differences between the pixels compared;

recording an error histogram representing the detected differences of the pixel of one of the images with the corresponding pixels of the another image, the error histogram comprising a 1 to n dimensional array representing a count of a particular percentage of error resulting from comparing pixels and wherein the difference is recorded by incrementing a column corresponding to that absolute difference;

determining the outcome of comparing the pixels of the images by comparing the differences detected with the tolerance level set; and

outputting results of the determination.

2. The method of claim 1 wherein preprocessing the images to be compared further comprises applying a filter to each of the images.

3. The method of claim 2 wherein applying a filter to each of the images comprises suppressing high frequencies in the images.

4. The method of claim 2 wherein applying a filter to each of the images comprises enhancing low frequencies in the images.

5. The method of claim 1 wherein preprocessing the images to be compared comprises performing a geometrical transform on one of the images.

6. The method of claim 5 wherein performing a geometrical transform comprises performing a 2D affine transform.

7. The method of claim 1 wherein setting an adjustable tolerance level for determining the results of comparing images comprises selecting one of a set of predefined tolerance levels.

8. The method of claim 7 wherein selecting one of a set of predefined tolerance levels comprises selecting one of a set of predefined tolerance curves.

9. The method of claim 7 wherein selecting one of a set of predefined tolerance levels comprises initializing the adjustable tolerance level to a default tolerance level.

10. The method of claim 1 wherein setting an adjustable tolerance level for determining the results of comparing images comprises automatically selecting one of a set of predefined tolerance levels based upon an image type of one of the images.

11. The method of claim 1 wherein comparing pixels of one of the images with corresponding pixels of another image to detect differences between the pixels compared comprises measuring the distance between the colors of the pixels of the images as the sum of the absolute difference between the R, G, B and alpha channels.

12. The method of claim 1 wherein comparing pixels of one of the images with corresponding pixels of another image to detect differences between the pixels compared comprises recording the result of the differences between pixels.

13. The method of claim 12 wherein recording the result of the differences between pixels comprises recording the result of the differences between pixels in a histogram.

14. A computer-readable storage medium having computer-executable instructions for performing the method of claim 1.