Preprocessing for information pattern analysis

• US 7,400,777 B2
• Filed: 05/25/2005
• Issued: 07/15/2008
• Est. Priority Date: 05/25/2005
• Status: Expired due to Fees

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Dang, Yingnong, Wang, Jian, Chen, Liyong
• Primary Examiner(s): Kassa; Yosef
• Application Number:
  US11/138,959
• Publication Number:
  US 20060269134A1
• Time in Patent Office: 1,147 Days
• Field of Search: 382/260, 382/274, 382/275, 382/300
• US Class Current: 382/254
• CPC Class Codes: G06V 30/10 Character recognition
• CPC Class Codes: G06V 30/142 using hand-held instruments; Constructional details of the instruments
• CPC Class Codes: G06V 30/162 Quantising the image signal

Abstract

Pre-processing techniques for processing an image to improve the distinctiveness of an information pattern captured in the image before the information pattern is analyzed in a decoding process. The brightness of an image first is normalized by dividing the image into blocks of areas, such as pixels. A brightness distribution value then is determined for each area of the image by fitting the brightness of its surrounding blocks using bilinear interpolation and extrapolation, and a normalized brightness value for each area can then be obtained by dividing the original brightness value by the brightness distribution value. Next, masks are created to distinguish the information pattern from content captured in the image. The masks may be generated based upon contrast differences between the brightness of pixels representing the information pattern, the brightness of pixels representing content, and the brightness of pixels representing the background of the writing medium.

Claims

1. A method of normalizing brightness in an image, comprising:

receiving an image made up of areas, each area having an original brightness value;

dividing the image up into blocks of areas;

determining an estimated brightness of each block;

determining a brightness distribution value for each area by fitting the estimated brightness determined for each block surrounding the area using bilinear interpolation and extrapolation;

normalizing a brightness value for each area by dividing an original brightness value for the area by the brightness distribution value determined for the area to produce a normalized brightness value; and

creating a new image using the normalized brightness value for each area.

2. The method recited in claim 1, wherein an area is a pixel.

3. The method recited in claim 1, further comprising

multiplying the normalized brightness value for each area by a uniform brightness level to produce an adjusted normalized brightness value for each area;

and

creating the new image using the adjusted normalized brightness value for each area.

4. The method recited in claim 1, further comprising determining the estimated brightness of each block to be the 90th percentile of brightness values of the areas making up the block.

5. The method recited in claim 1, wherein the image is a black-and-white image;

and the brightness values are gray level values.

6. The method recited in claim 1, wherein

the image is a color image, and

the brightness level values are color brightness values.