Image processing device to determine image quality and method thereof

• US 20070041657A1
• Filed: 08/17/2006
• Published: 02/22/2007
• Est. Priority Date: 08/19/2005
• Status: Abandoned Application

• Current Assignee: Samsung Electronics Co. Ltd.
• Original Assignee: Samsung Electronics Co. Ltd.
• Inventors: Rychagov, Michael, Efimov, Sergei
• Application Number:
  US11/505,440
• Publication Number:
  US 20070041657A1
• Time in Patent Office: Days
• US Class Current: 382/274
• CPC Class Codes: G06T 2207/20021 Dividing image into blocks, subimages or windows
• CPC Class Codes: G06T 2207/30144 Printing quality
• CPC Class Codes: G06T 5/40 using histogram techniques
• CPC Class Codes: G06T 5/92 based on global image properties
• CPC Class Codes: G06T 7/0004 Industrial image inspection
• CPC Class Codes: G06V 10/993 Evaluation of the quality of the acquired pattern

Abstract

An image processing device that determines the image quality includes a division unit dividing image data into a plurality of divided images, a calculation unit calculating luminance histograms of the respective divided images, and a judgment unit judging qualities of the respective divided images using the luminance histograms of the divided images and judging the quality of the image data in consideration of rates of the judged qualities. Accordingly, the quality of the image data can be determined more accurately and more precisely.

Claims

1. An image processing device comprising:

a division unit dividing image data into a plurality of divided images;

a calculation unit calculating a plurality of luminance histograms of the respective divided images; and

a judgment unit judging qualities of the respective divided images using the plurality of luminance histograms of the divided images and judging the quality of the image data in consideration of rates of the judged qualities.

2. The image processing device of claim 1, further comprising a contrast adjustment unit adjusting a contrast of the image data to be divided and providing the adjusted contrast to the division unit.

3. The image processing device of claim 2, wherein the contrast adjustment unit adjusts the whole contrast of the image data using the following equations,

$\mathrm{low}=\{\begin{array}{c}\min \end{array}$

{ i / H ⁡

( i ) >

H 0 , i <

T ⁢

 

⁢

or ⁢

 

⁢

i = T } T , i >

T ⁢

⁢

high = max ⁢

 

⁢

( i ) ⁢

;

⁢

H ⁡

[ i ] >

T ⁢

⁢

R ⁡

( r , c ) = ( 2 n - 1 ) * { ( R ⁡

( r , c ) - low } / ( high - low ) ⁢

⁢

G ⁡

( r , c ) = ( 2 n - 1 ) * { ( G ⁡

( r , c ) - low } / ( high - low ) ⁢

⁢

B ⁡

( r , c ) = ( 2 n - 1 ) * { ( B ⁡

( r , c ) - low } / ( high - low )

where;

low and high denote lower and upper limit values for contrast adjustment,

i is luminance,

H(i) is luminance frequency,

r and c are horizontal and vertical coordinates of a pixel,

T is a threshold value for preventing an image from being excessively dark,

H₀ is a threshold value of histogram levels,

n is the number of image bits, and

R(r, c), G(r, c), and B(r, c) are R, G, and B component values, respectively.

4. The image processing device of claim 1, for each divided image, wherein the judgment unit:

judges the divided image as an underexposed image when an average value of the calculated luminance histograms for the divided image is below a preset first threshold value;

judges the divided image as an overexposed image when the average value of the calculated luminance histograms for the divided image is over a preset second threshold value;

judges the divided image as a low-contrast image when the average value of the calculated luminance histograms for the divided image is at or between the preset first threshold value and the preset second threshold value, and a standard deviation is below a preset threshold value of the standard deviation; and

judges the divided image as a normal image when the average value of the calculated luminance histograms for the divided image is at or between the preset first threshold value and the preset second threshold value, and when the standard deviation is over the preset threshold value of the standard deviation.

5. The image processing device of claim 4, wherein the judgment unit judges the image data as a low-quality data when the rate of the divided image that is judged as the normal image among the plurality of divided images is below a preset threshold rate.

6. The image processing device of claim 1, wherein the division unit divides the image data into the plurality of divided images after converting the image data into a color gamut in the form in which luminance information and color information are separated from the image data.

7. The image processing device of claim 1, further comprising a storage unit storing the image data;

wherein the division unit reads the image data stored in the storage unit, and divides the read image data into the plurality of divided images.

8. The image processing device of claim 1, further comprising an input unit receiving the image data from an external source;

wherein the division unit divides the image data received through the input unit into the plurality of divided images.

9. The image processing device of claim 1, further comprising a control unit discarding the image data when the quality of the image data is judged as low-quality.

10. The image processing device of claim 1, further comprising an image processor performing an image process for improving the quality of the image data when the quality of the image data is judged as low-quality.

11. An image determining method comprising:

dividing image data into a plurality of divided images;

calculating a plurality of luminance histograms for the respective divided images;

judging qualities of the respective divided images using the plurality of luminance histograms for the corresponding divided images; and

judging the quality of the image data in consideration of rates of the judged qualities.

12. The image determining method of claim 11, further comprising, before dividing the image data, adjusting the whole contrast of the image data.

13. The image determining method of claim 12, wherein the adjusting the contrast uses following equations,

$\mathrm{low}=\{\begin{array}{c}\min \end{array}$

{ i / H ⁡

( i ) >

H 0 , i <

T ⁢

 

⁢

or ⁢

 

⁢

i = T } T , i >

T ⁢

⁢

high = max ⁢

 

⁢

( i ) ⁢

;

⁢

H ⁡

[ i ] >

T ⁢

⁢

R ⁡

( r , c ) = ( 2 n - 1 ) * { ( R ⁡

( r , c ) - low } / ( high - low ) ⁢

⁢

G ⁡

( r , c ) = ( 2 n - 1 ) * { ( G ⁡

( r , c ) - low } / ( high - low ) ⁢

⁢

B ⁡

( r , c ) = ( 2 n - 1 ) * { ( B ⁡

( r , c ) - low } / ( high - low )

where

low and high denote lower and upper limit values for contrast adjustment,

i is luminance,

H(i) is luminance frequency,

r and c are horizontal and vertical coordinates of a pixel,

T is a threshold value for preventing an image from being excessively dark,

H₀ is a threshold value of histogram levels,

n is the number of image bits, and

R(r, c), G(r, c), and B(r, c) are R, G, and B component values, respectively.

14. The image determining method of claim 11, wherein, for each divided image, the judging the qualities of the respective divided images includes:

judging the divided image as an underexposed image when an average value of the calculated luminance histograms for the divided image is below a preset first threshold value;

judging the divided image as an overexposed image when the average value of the calculated luminance histograms for the divided image is over a preset second threshold value;

judging the divided image as a low-contrast image when the average value of the calculated luminance histograms for the divided image is at or between the preset first threshold value and the preset second threshold value, and a standard deviation is below a preset threshold value of the standard deviation; and

judging the divided image as a normal image when the average value of the calculated luminance histograms for the divided image is at or between the preset first threshold value and the preset second threshold value, and the standard deviation is over the preset threshold value of the standard deviation.

15. The image determining method of claim 14, wherein the judging the quality of the image data includes judging the quality of the image data as a low-quality when the rate of the divided image that is judged as the normal image among the plurality of divided images is below a preset threshold rate.

16. The image determining method of claim 11, wherein the dividing the image data includes dividing the image data into the plurality of divided images after converting the image data into a color gamut in the form in which luminance information and color information are separated from the image data.

17. A computer-readable medium encoded with a program that implements an image-quality judgment process as recited in claim 11, performed by at least one computer.

18. The computer-readable medium of claim 17, wherein the image-quality judgment process further comprises adjusting the whole contrast of the image data before dividing the image data.

19. The computer-readable medium of claim 17, wherein, for each divided image, the judging the qualities of the respective divided images comprises:

judging the divided image as an underexposed image when an average value of the calculated luminance histograms for the divided image is below a preset first threshold value;

judging the divided image as an overexposed image when the average value of the calculated luminance histograms for the divided image is over a preset second threshold value;

judging the divided image as a low-contrast image when the average value of the calculated luminance histograms for the divided image is over the preset first threshold value and is below the preset second threshold value, and when a standard deviation is below a preset threshold value of the standard deviation; and

judging the divided image as a normal image when the average value of the calculated luminance histograms for the divided image is over the preset first threshold value and is below the preset second threshold value, and the standard deviation is over the preset threshold value of the standard deviation.

20. The computer-readable medium of claim 19, wherein the judging the quality of the image data includes judging the quality of the image data as low-quality when the rate of the divided image that is judged as the normal image among the plurality of divided images is below a preset threshold rate.

21. An image processing device comprising:

a calculation unit that calculates, for each of a plurality of divided portions of an image, a corresponding luminance histograms to produce a plurality of luminance histograms; and

a judgment unit that uses each of the plurality of luminance histograms to judge qualities of the each of the divided portions, and judges an overall quality of the image according to the judged qualities for each of the divided portions.

22. The image processing device of claim 21, further comprising a contrast adjustment unit that adjusts a contrast of the image prior to the image being received at the calculation unit.

23. The image processing device of claim 22, wherein the contrast adjustment unit adjusts the contrast of the image using the following equations:

$\mathrm{low}=\{\begin{array}{cc}\min <br><br>\{i/H<br><br>\left(i\right)><br><br>H_0,& i<<br><br>T<br><br>\text{\hspace{1em}<br><br>}<br><br>\mathrm{or}<br><br>\text{\hspace{1em}<br><br>}<br><br>i=T\}\\ T,& i><br><br>T\end{array}<br><br><br><br>\mathrm{high}=\max <br><br>\text{\hspace{1em}<br><br>}<br><br>\left(i\right)<br><br>\text{;<br><br>}<br><br>H<br><br>\left[i\right]><br><br>T<br><br><br><br>R<br><br>\left(r,c\right)=\left(2^n-1\right)*\{\left(R<br><br>\left(r,c\right)-\mathrm{low}\right\}/\left(\mathrm{high}-\mathrm{low}\right)<br><br><br><br>G<br><br>\left(r,c\right)=\left(2^n-1\right)*\{\left(G<br><br>\left(r,c\right)-\mathrm{low}\right\}/\left(\mathrm{high}-\mathrm{low}\right)<br><br><br><br>B<br><br>\left(r,c\right)=\left(2^n-1\right)*\{\left(B<br><br>\left(r,c\right)-\mathrm{low}\right\}/\left(\mathrm{high}-\mathrm{low}\right)$

where;

low and high denote lower and upper limit values for contrast adjustment,

i is luminance,

H(i) is luminance frequency,

r and c are horizontal and vertical coordinates of a pixel,

T is a threshold value for preventing an image from being excessively dark,

H₀ is a threshold value of histogram levels,

n is the number of image bits, and

R(r, c), G(r, c), and B(r, c) are R, G, and B component values, respectively.

24. The image processing device of claim 21, wherein the judgment unit:

calculates a standard deviation for the divided portion;

judges the divided portion as a low-contrast image when the calculated standard deviation is at or below a threshold value of the standard deviation; and

judges the divided portion as a normal image when the calculated standard deviation is over the threshold value of the standard deviation.

25. The image processing device of claim 21, wherein the judgment unit:

judges the divided portion as an underexposed image when an average value of the calculated luminance histograms is below a first value; and

judges the divided portion as an overexposed image when the average value of the plurality of the luminance histograms is over a second value.

26. The image processing device of claim 25, wherein, when the average value of the calculated luminance histograms for the divided image is at or between the first value and the second value, the judgment unit:

calculates a standard deviation for the divided portion;

judges the divided image as a low-contrast image when the calculated standard deviation is below a threshold value; and

judges the divided image as a normal image when the calculated standard deviation is at or over the threshold value.

27. The image processing device of claim 26, wherein the average value of the calculated luminance histograms is calculated according to the following equation:

$H_{\mathrm{mean}}=\frac{1}{N}<br><br>\underset{i=1}{\stackrel{N}{Q}}<br><br>H<br><br>\left(i\right)$

where

H_mean is the average value of luminance histograms,

i is a luminance value,

H(i) is a luminance frequency, and

N is the number of luminance histograms.

28. The image processing device of claim 26, wherein the standard deviation is calculated by the following equation:

$\mathrm{\sigma <br><br>}=\sqrt{\frac{1}{N-1}<br><br>{\underset{i=1}{\stackrel{N}{Q}}<br><br>\left(H<br><br>\left(i\right)-H_{\mathrm{mean}}\right)}^2}$

where

σ

is the standard deviation,

H_mean is the average value of luminance histograms,

i is a luminance value,

H(i) is a luminance frequency, and

N is the number of luminance histograms.

29. The image processing device of claim 24, wherein the judgment unit judges the image as low-quality when a number of the divided portions are judged as the normal image as compared to the number of divided portions being judged is below a threshold.

30. The image processing device of claim 21, further comprising a division unit that converts the image into a color gamut to separate luminance information from color information, and divides the image into the plurality of divided portions.