Image Processing Apparatus And Image Processing Method

• US 20070019243A1
• Filed: 07/18/2006
• Published: 01/25/2007
• Est. Priority Date: 07/20/2005
• Status: Active Grant

• Current Assignee: Seiko Epson Corporation (Seiko Group)
• Original Assignee: Seiko Epson Corporation (Seiko Group)
• Inventors: Katagiri, Masahiro, Sato, Yoshitake, Furuhata, Tadashi
• Granted Patent: US 7,760,958 B2
• Time in Patent Office: Days
• US Class Current: 358/3.270
• CPC Class Codes: H04N 1/38 Circuits or arrangements for blanking or otherwise eliminating unwanted parts of pictures H04N1/387 takes precedence
• CPC Class Codes: H04N 1/403 Discrimination between the two tones in the picture signal of a two-tone original
• CPC Class Codes: H04N 1/4092 Edge or detail enhancement

Abstract

Background images are removed and only the desired payment information is extracted from a grayscale image of a check or similar financial instrument. A density distribution improvement process applied to grayscale raw image data 40 acquired by scanning a check corrects the density distribution of the raw image data 40 to separate the density range of the desired payment information from the density range of the background image. Image sharpening or contrast enhancement can be used as the density distribution improvement method. A threshold level 44 for clearly separating the payment information from the background image is then calculated from the characteristics of the density distribution of the grayscale improved image data 42 acquired by the density distribution improvement process. Histograms of the density distributions are used to determine the characteristics of the density distribution. The improved image data 42 is then converted to binary image data 46 using the threshold level 44. Most of the background image is white and most of the desired payment information is black in the binary image data 46.

Claims

1. An image processing apparatus comprising:

a density distribution improvement means for receiving grayscale raw image data containing both background image information and foreground image information, modifying a density distribution of the raw image data to increase a difference between a density distribution range of the background image information and a density distribution range of the foreground image information, and generating grayscale improved image data;

a threshold level calculation means for receiving the improved image data and calculating a threshold level for two-level digital conversion based on a density distribution in the improved image data; and

a digitizing means for receiving the improved image data and the threshold level, digitizing the improved image data using the threshold level, and outputting two-level image data.

2. The image processing apparatus described in claim 1, wherein the density distribution improvement means applies an image sharpening process to the raw image data.

3. The image processing apparatus described in claim 2, further comprising a noise removal means for receiving the two-level image data and removing noise from the two-level image data.

4. The image processing apparatus described in claim 1, wherein the threshold level calculation means detects a junction between a density range of the background image of the improved image data and a density range of the foreground image of the improved image data, and sets the threshold level within the junction.

5. The image processing apparatus described in claim 4, wherein the threshold level calculation means calculates a histogram for the improved image data, and detects the junction from slopes in the histogram.

6. The image processing apparatus of claim 5, wherein a starting point of the junction adjoining the foreground image data is defined by determining where the density distribution of the foreground image changes from descending to ascending or level.

7. The image processing apparatus of claim 5, wherein an end point of the junction adjoining the background image data is defined by first identifying a provisional end point as the point where the density distribution of the background image changes from ascending to descending or level, and then determining the slope of the background image density from the provisional end point toward an apex of the background image density, and identifying said end point as a point where the slope is determined to be not less than a predefined positive value.

8. The image processing apparatus described in claim 4, wherein the threshold level calculation means calculates a histogram for the improved image data, finds transition points in the histogram for transitioning between background and foreground image information, and sets the threshold level based on the transition points.

9. An image processing apparatus, comprising:

a density distribution improvement module coupled to receive grayscale raw image data containing both background image information and foreground image information, said density distribution improvement module modifying a density distribution of the raw image data to increase a difference between a density distribution range of the background image information and a density distribution range of the foreground image information, and generating grayscale improved image data;

a threshold level calculation module coupled to receive the improved image data and calculate a threshold level for two-level digital conversion based on a density distribution in the improved image data; and

a digitizing module coupled to receive the improved image data and the threshold level, digitize the improved image data using the threshold level, and output two-level image data.

10. The image processing apparatus described in claim 9, wherein the density distribution improvement module further applies an image sharpening process to the raw image data.

11. The image processing apparatus described in claim 10, further comprising a noise removal module coupled to receive the two-level image data and remove noise from the two-level image data.

12. The image processing apparatus described in claim 9, wherein the threshold level calculation module detects a junction between a density range of the background image of the improved image data and a density range of the foreground image of the improved image data, and sets the threshold level within the junction.

13. The image processing apparatus described in claim 12, wherein the threshold level calculation module calculates a histogram for the improved image data, and detects the junction from slopes in the histogram.

14. The image processing apparatus of claim 13, wherein a starting point of the junction adjoining the foreground image data is defined by determining where the density distribution of the foreground image changes from descending to ascending or level.

15. The image processing apparatus of claim 13, wherein an end point of the junction adjoining the background image data is defined by first identifying a provisional end point as a point where the density distribution of the background image changes from ascending to descending or level, and then determining the slope of the background image density from the provisional end point toward an apex of the background image density, and identifying said end point as the point where the slope is determined to be greater than a predefined positive value.

16. The image processing apparatus described in claim 12, wherein the threshold level calculation means calculates a histogram for the improved image data, finds transition points in the histogram for transitioning between background and foreground image information, and sets the threshold level based on the transition points.

17. An image processing method comprising:

a density distribution improvement step for receiving grayscale raw image data containing both background image information and foreground image information;

modifying a density distribution of the raw image data to increase a difference between a density distribution range of the background image information and a density distribution range of the foreground image information, and generating grayscale improved image data;

a threshold level calculation step for receiving the improved image data and calculating a threshold level for two-level digital conversion based on a density distribution in the improved image data; and

a digitizing step for receiving the improved image data and the threshold level, digitizing the improved image data using the threshold level, and outputting two-level image data.

18. The image processing method described in claim 17, wherein the density distribution improvement step further applies an image sharpening process to the raw image data.

19. The image processing method described in claim 18, further comprising a noise removal step for receiving the two-level image data and removing noise from the two-level image data.

20. The image processing method described in claim 17, wherein the threshold level calculation step detects a junction between a density range of the background image of the improved image data and a density range of the foreground image of the improved image data, and sets the threshold level within the junction.

21. The image processing method described in claim 20, wherein the threshold level calculation step calculates a histogram for the improved image data, and detects the junction from slopes in the histogram.

22. The image processing method of claim 21, wherein a starting point of the junction adjoining the foreground image data is defined by determining where the density distribution of the foreground image changes from descending to ascending or level.

23. The image processing method of claim 21, wherein an end point of the junction adjoining the background image data is defined by first identifying a provisional end point as the point where the density distribution of the background image changes from ascending to descending or level, and then determining the slope of the background image density from the provisional end point toward an apex of the background image density, and identifying said end point as a point where the slope is determined to be not less than a predefined positive value.

24. The image processing method described in claim 20, wherein the threshold level calculation step calculates a histogram for the improved image data, finds transition points in the histogram for transitioning between background and foreground image information, and sets the threshold level based on the transition points.