Image quality improving apparatus and method

• US 20050249430A1
• Filed: 05/06/2005
• Published: 11/10/2005
• Est. Priority Date: 05/07/2004
• Status: Active Grant

• Current Assignee: S-Printing Solution Co., Ltd. (Samsung Electronics Co. Ltd.)
• Original Assignee: Samsung Electronics Co. Ltd.
• Inventors: Lim, Sung-Hyun
• Granted Patent: US 7,746,505 B2
• Time in Patent Office: Days
• US Class Current: 382/266
• CPC Class Codes: G06T 2207/20012 Locally adaptive
• CPC Class Codes: G06T 2207/20192 Edge enhancement; Edge preservation
• CPC Class Codes: G06T 2207/30176 Document
• CPC Class Codes: G06T 5/20 using local operators
• CPC Class Codes: G06T 5/73 Deblurring; Sharpening
• CPC Class Codes: G06T 7/13 Edge detection
• CPC Class Codes: G06T 7/155 involving morphological operators
• CPC Class Codes: H04N 1/4092 Edge or detail enhancement

Abstract

An image quality improving apparatus and method that distinguish an edge of an actual outline imagely recognized in an image inputted via an image input unit from those printed according to dither patterns, to emphasize the actual outline, and thereby acquire the image without distortion. The image quality improving apparatus comprises: a first unit estimating edge pixels of a text area in a predetermined detection area to be edge-detected from an input picture data containing a text and an image; and a second unit judging actual edge pixels based upon a number of edge-estimated pixels in a multiple window area and saturation values of individual edge-estimated pixels, and emphasizing individual actual edge pixels with different edge emphasis coefficients according to saturation values of the actual edge pixels.

Claims

1. An image quality improving apparatus comprising:

a first unit estimating edge pixels of a text area in a predetermined detection area to be edge-detected from an input picture data containing a text and an image; and

a second unit judging actual edge pixels based upon a number of edge-estimated pixels in a multiple window area and saturation values of individual edge-estimated pixels, and emphasizing individual actual edge pixels with different edge emphasis coefficients according to saturation values of the respective actual edge pixels.

2. The image quality improving apparatus according to claim 1, wherein the first unit comprises:

a lightness variation measuring section measuring lightness variation from individual pixels in the predetermined detection area to be edge-detected; and

an edge estimating section calculating binary data of the individual pixels according to multiple windows, the binary data being composed by the lightness variation measuring section according to different sampling procedures from an area having lightness variation, the edge estimating section also judging connectedness and similarity of the calculated binary data according to the individual windows, to estimate whether a given pixel is an edge.

3. The image quality improving apparatus according to claim 2, wherein the edge estimating section estimates a given pixel as a non-edge if the lightness variation is greater than or equal to a predetermined lightness variation reference value, connected component values of the binary data do not have connectedness, or windows do not have similarity.

4. The image quality improving apparatus according to claim 2, wherein the windows do not have similarity if connected component values of individual windows are not the same.

5. The image quality improving apparatus according to claim 1, wherein the second unit comprises:

an edge detecting section performing a first comparison, comprising comparing the saturation values of the individual edge-estimated pixels of multiple windows and the number of the edge-estimated pixels with a predetermined reference saturation value and a reference estimation number, respectively, to judge whether the edge-estimated pixels are actual edge pixels or not, based upon a result of the first comparison; and

an edge emphasizing section performing a second comparison, comprising comparing the saturation values of the actual edge pixels with the predetermined reference saturation value to execute edge emphasis on the individual pixels using different edge emphasis coefficients according to a result of the second comparison.

6. The image quality improving apparatus according to claim 5, wherein the edge detecting section judges the edge-estimated pixels as actual edge pixels if the saturation values of the individual edge-estimate pixels are larger than the predetermined reference saturation value and the number of the edge-estimated pixels is larger than a predetermined first reference edge number, and judges the edge-estimate pixels as not being actual edge pixels if the saturation values of the individual edge-estimate pixels are smaller than or equal to the predetermined reference saturation value and the number of the edge-estimated pixels is smaller than or equal to the predetermined first reference edge number.

7. The image quality improving apparatus according to claim 5, wherein:

a first pixel edge emphasis coefficient is applied to a given actual edge pixel when the saturation value of the given actual edge pixel is larger than the predetermined reference saturation value;

a second pixel edge emphasis coefficient is applied to a given actual edge pixel when the saturation value of the given actual edge pixel is smaller than the predetermined reference saturation value; and

the second pixel edge emphasis coefficient is larger than the first pixel edge emphasis coefficient.

8. An image quality improving method, comprising the operations of:

estimating edge pixels of a text area in a predetermined detection area to be edge-detected from an input picture data containing a text and an image;

judging actual edge pixels based upon a number of edge-estimated pixels in a multiple window area and saturation values of individual edge-estimated pixels; and

emphasizing individual actual edge pixels with different edge emphasis coefficients according to saturation values of the actual edge pixels.

9. The image quality improving method according to claim 8, wherein the edge pixel estimating operation comprises:

measuring lightness variation from individual pixels in the predetermined detection area to be edge-detected; and

calculating binary data of the individual pixels according to multiple windows, the binary data being composed according to different sampling procedures from an area having lightness variation, and judging connectedness and similarity of the calculated binary data according to the individual windows, to estimate whether a given pixel is an edge.

10. The image quality improving method according to claim 9, wherein the edge pixel estimating operation comprises:

estimating a given pixel as a non-edge if the lightness variation is greater than or equal to a predetermined reference value, connected component values of the binary data do not have connectedness, or windows do not have similarity.

11. The image quality improving method according to claim 10, wherein according to a result of the lightness variation measuring operation, windows do not have similarity if connected component values of individual windows are not the same.

12. The image quality improving method according to claim 9, wherein the different sampling procedures are at least one selected from a group consisting of 5×

5, 5×

7, and 5×

9 window mask sampling procedures.

13. The image quality improving method according to claim 8, wherein the edge emphasizing operation comprises:

in a first comparison, comparing the saturation values of the individual edge-estimated pixels of multiple windows and the number of the edge-estimated pixels with a predetermined reference saturation value and a reference estimation number, respectively, to judge whether the edge-estimated pixels are the actual edge pixels or not, based upon a result of the first comparison; and

in a second comparison, comparing the saturation values of the actual edge pixels with the predetermined reference saturation value to emphasize the individual actual edge pixels using different edge emphasis coefficients according to a result of the second comparison.

14. The image quality improving method according to claim 13, wherein the actual pixel judging operation comprises:

judging the edge-estimated pixels as actual edge pixels if the saturation values of the individual edge-estimate pixels are larger than the predetermined reference saturation value and the number of the edge-estimated pixels is larger than a predetermined first reference edge number; and

judging the edge-estimate pixels as not being actual edge pixels if the saturation values of the individual edge-estimate pixels are smaller or equal to than the predetermined reference saturation value and the number of the edge-estimated pixels is smaller than or equal to the predetermined first reference edge number.

15. The image quality improving method according to claim 13, wherein the edge emphasizing operation further comprises:

emphasizing a given actual edge pixel with a first edge emphasis coefficient if the saturation value of the given actual edge pixel is larger than the predetermined reference saturation value; and

emphasizing a given actual edge pixel with a second edge emphasis coefficient if the saturation value of the given actual edge pixel is smaller than or equal to the predetermined reference saturation value.

16. The image quality improving method according to claim 15, wherein the first edge emphasis coefficient is smaller than the second edge emphasis coefficient.

17. An image quality improving method, comprising:

estimating whether pixels within a detection area of an input picture are edges using lightness variations of the pixels;

judging whether edge-estimated pixels are edges using pixel saturation values and a number of edge-estimated pixels; and

applying a plurality of edge emphasis coefficients to edge-judged pixels in accordance with respective saturation values of the edge-judged pixels.

18. The method according to claim 17, wherein the estimating whether the pixels within the detection area of the input picture are edges using lightness variations of the pixels comprises:

for each pixel having a lightness value variation exceeding a predetermined reference lightness variation value, comparing respective lightness differences of surrounding pixels to calculate binary data representing a given pixel with varied lightness and the surrounding pixels;

using the binary data, calculating a binary data connected component for each of multiple windows representing surrounding pixel sample sizes, the multiple windows being centered on the given pixel with varied lightness;

determining whether the multiple windows have a connectedness using the respective binary data connected components;

for windows determined to have connectedness, determining whether the windows have similarity; and

for windows determined to have similarity, estimating the given pixel with varied lightness to be an edge.

19. The method according to claim 18, wherein the determining whether the windows have similarity comprises:

determining whether the binary data of the multiple windows satisfies the following equation

$\underset{i=1}{\overset{3}{\sum <br><br>}}<br><br>\underset{j=1}{\overset{3}{\sum <br><br>}}<br><br>a_{\mathrm{ij}}<br><br>b_{\mathrm{ij}}<br><br>c_{\mathrm{ij}}\ge <br><br>{\mathrm{TH}}_a$

wherein a_ij, b_ij, and c_ij are the binary data values of multiple windows, respectively, and TH_a is a predetermined reference value.

20. The method according to claim 17, wherein the judging whether the edge-estimated pixels are edges using the pixel saturation values and the number of edge-estimated pixels comprises:

judging a given edge-estimated pixel to be an edge if the saturation value of the given edge-estimated pixel exceeds a predetermined saturation reference value and the number of edge-estimated pixels exceeds a first reference number; and

judging the given pixel to be an edge if the saturation value of the given edge-estimated pixel does not exceed the predetermined saturation reference value and the number of edge-estimated pixels exceeds a second reference number.

21. The method according to claim 17, wherein the applying the plurality of edge emphasis coefficients to the edge-judged pixels in accordance with the respective saturation values of the edge-judged pixels comprises:

applying edge emphasis to a given edge judged pixel using a first predetermined edge emphasis coefficient if the saturation value of the given edge-judged pixel does not exceed a predetermined saturation reference value; and

applying edge emphasis to the given edge judged pixel using a second predetermined edge emphasis coefficient if the saturation value of the given edge-judged pixel exceeds the predetermined saturation reference value, the first predetermined edge emphasis coefficient being greater than the second predetermined edge emphasis coefficient.

22. The method according to claim 17, wherein the respective saturation values of the edge-judged pixels are determined using the following equations:

Cb=(−

0.167969*R)−

(0.331055*G)+(0.5*B);


Cr=(0.5*R)−

(0.417969*G)−

(0.0810547*B); and


S=|C_b|+|C_r|,

wherein R, G, and B are red, green, and blue values, respectively, of a given edge-judged pixel, and S is the saturation value of the given edge-judged pixel.

23. An image quality improving apparatus, comprising:

an edge sorting section estimating whether pixels within a detection area of an input picture are edges using lightness variations of the pixels;

an edge detecting section judging whether edge-estimated pixels are edges using pixel saturation values and a number of edge-estimated pixels; and

an edge emphasizing section applying a plurality of edge emphasis coefficients to edge-judged pixels in accordance with respective saturation values of the edge-judged pixels.

24. The image quality improving apparatus according to claim 23, wherein the an edge sorting section comprises:

a lightness variation measuring section determining whether lightness value variations of pixels within the detection area exceed a predetermined reference lightness variation value; and

a halftone image detecting section

comparing respective lightness differences of surrounding pixels, for each pixel having a lightness value variation exceeding the predetermined reference lightness variation value, to calculate binary data representing a given pixel with varied lightness and the surrounding pixels,

using the binary data, calculating a binary data connected component for each of multiple windows representing surrounding pixel sample sizes, the multiple windows being centered on the given pixel with varied lightness,

determining whether the multiple windows have a connectedness using the respective binary data connected components,

for windows determined to have connectedness, determining whether the windows have similarity, and

for windows determined to have similarity, estimating the given pixel with varied lightness to be an edge.

25. The image quality improving apparatus according to claim 24, wherein the determining whether the windows have similarity comprises:

determining whether the binary data of the multiple windows satisfies the following equation

$\underset{i=1}{\overset{3}{\sum <br><br>}}<br><br>\underset{j=1}{\overset{3}{\sum <br><br>}}<br><br>a_{\mathrm{ij}}<br><br>b_{\mathrm{ij}}<br><br>c_{\mathrm{ij}}\ge <br><br>{\mathrm{TH}}_a$

wherein a_ij, b_ij, and c_ij are the binary data values of multiple windows, respectively, and TH_a is a predetermined reference value.

26. The image quality improving apparatus according to claim 23, wherein:

the edge detecting section

judges a given edge-estimated pixel to be an edge if the saturation value of the given edge-estimated pixel exceeds a predetermined saturation reference value and the number of edge-estimated pixels exceeds a first reference number, and

judges the given pixel to be an edge if the saturation value of the given edge-estimated pixel does not exceed the predetermined saturation reference value and the number of edge-estimated pixels exceeds a second reference number.

27. The image quality improving apparatus according to claim 23, wherein:

the edge emphasizing section

applies edge emphasis to a given edge judged pixel using a first predetermined edge emphasis coefficient if the saturation value of the given edge-judged pixel does not exceed a predetermined saturation reference value, and

applies edge emphasis to the given edge judged pixel using a second predetermined edge emphasis coefficient if the saturation value of the given edge-judged pixel exceeds the predetermined saturation reference value, the first predetermined edge emphasis coefficient being greater than the second predetermined edge emphasis coefficient.

28. The image quality improving apparatus according to claim 23, wherein the edge emphasizing section determines the respective saturation values of the edge-judged pixels using the following equations:

Cb=(−

0.167969*R)−

(0.331055*G)+(0.5*B);


Cr=(0.5*R)−

(0.417969*G)−

(0.0810547*B); and


S=|C_b|+C_r|,

wherein R, G, and B are red, green, and blue values, respectively, of a given edge-judged pixel, and S is the saturation value of the given edge-judged pixel.

29. An image quality improving method, comprising:

determining whether differences between maximum and minimum lightness values of pixels within a predetermined detection area in which edge detection is to be made with respect to an input picture exceed a predetermined reference lightness value;

for a given pixel, determining that the lightness of the given pixel is varied if the difference between the maximum and minimum lightness values exceeds the predetermined reference lightness value;

for each pixel with varied lightness, comparing respective lightness differences of surrounding pixels to calculate binary data representing a given pixel with varied lightness and the surrounding pixels;

using the binary data, calculating a binary data connected component for each of multiple windows representing surrounding pixel sample sizes, the multiple windows being centered on the given pixel with varied lightness;

determining whether the multiple windows have a connectedness using the respective binary data connected components;

for windows determined to have connectedness, determining whether the windows have similarity;

for windows determined to have similarity, estimating the given pixel with varied lightness to be an edge;

for pixels estimated to be an edge, determining whether calculated saturation values of the respective pixels exceed a predetermined saturation reference value;

if the saturation value of a given pixel exceeds the predetermined saturation reference value and a number of edge-estimated pixels exceeds a first reference number, judging the given pixel to be an edge;

if the saturation value of a given pixel does not exceed the predetermined saturation reference value and the number of edge-estimated pixels exceeds a second reference number, judging the given pixel to be an edge;

for each pixel judged to be an edge, determining whether the saturation values of the respective pixels exceed the predetermined saturation reference value;

if the saturation value of a given pixel judged to be an edge does not exceed the predetermined saturation reference value, edge emphasis is applied to the given pixel using a first predetermined edge emphasis coefficient; and

if the saturation value of a given pixel judged to be an edge exceeds the predetermined saturation reference value, edge emphasis is applied to the given pixel using a second predetermined edge emphasis coefficient, the first predetermined edge emphasis coefficient being greater than the second predetermined edge emphasis coefficient.