Adaptive histogram-based video contrast enhancement
View Patent ↗The adaptive contrast enhancer uses an adaptive histogram equalization-based approach to improve contrast in a video signal. For each video frame, the histogram of the pixel luminance values is calculated. The calculated histogram is divided into three regions that are equalized independently of the other. The equalized values are averaged with the original pixel values with a weighting factor that depends on the shape of the histogram. The weighting factors can be also chosen differently for the three regions to enhance the darker regions more than the brighter ones. The statistics calculated from one frame are used to enhance the next frame such that frame buffers are not required. Many of the calculations are done in the inactive time between two frames.
1. A contrast enhancement system, comprising:
a histogram updater configured to:
provide a histogram based on an image;
determine arithmetic means for at least two regions of the histogram; and
calculate variances for each of the determined arithmetic means;
a weighting factor calculator configured to calculate weighting factors based on the calculated variances; and
an equalizer configured to transform pixel values of the image using the weighting factors.
2. The system of claim 1 , wherein the histogram updater is configured to determine the arithmetic means in accordance with the equation
m
i
=
∑
j
y
i
N
where y j represents luma of points in a particular region i and N is the total number of points in that region.
3. The system of claim 2 , wherein the histogram updater is configured to calculate the variances in accordance with the equation
σ
i
=
1
N
∑
j
n
(
y
j
)
(
y
j
-
m
i
)
where n(y j ) is a count of the number of pixels of luma y j and σ i is the variance of the ith region.
4. The system of claim 1 , wherein the equalizer is further configured to transform pixel values of the image based on a cumulative histogram of the image.
5. The system of claim 1 , wherein the equalizer is further configured to transform pixel values of the image based on a cumulative histogram of a previous image.
6. The system of claim 5 further comprising two memories configured to store the histogram of the image and the cumulative histogram of the previous image on an alternating basis.
7. The system of claim 6 , wherein the equalizer is further configured to transform pixel values of the image based on a cumulative density function obtained from the cumulative histogram of the previous image.
8. The system of claim 7 , wherein the cumulative density function is obtained from the cumulative histogram of the previous image using a reciprocal of a number of points in each region.
9. The system of claim 1 , wherein the weighting factor calculator is configured to calculate the weighting factors in accordance with a weighting factor curve that defines a weighting factor for each calculated variance.
10. The system of claim 9 , wherein each region has a respective weighting factor curve that depends on a desired level of enhancement.
11. A method for enhancing contrast of an image, comprising:
generating a histogram based on an image;
determining arithmetic means for at least two regions of the histogram;
calculating variances for each of the determined arithmetic means;
calculating weighting factors based on the calculated variances; and
transforming pixel values of the image using the weighting factors.
12. The method of claim 11 , wherein the arithmetic means are determined in accordance with the equation
m
i
=
∑
j
y
j
N
,
where y j represents luma of points in a particular region i and N is the total number of points in that region.
13. The method of claim 12 , wherein the variances are calculated in accordance with the equation
σ
i
=
1
N
∑
j
n
(
y
j
)
(
y
j
-
m
i
)
,
where n(y j ) is a count of the number of pixels of luma y j and σ i is the variance of the ith region.
14. The method of claim 11 , wherein the pixel values of the image are transformed based on a cumulative histogram of the image.
15. The method of claim 11 , wherein the pixel values of the image are transformed based on a cumulative histogram of a previous image.
16. The method of claim 15 , wherein the pixel values of the image are transformed based on a cumulative density function obtained from the cumulative histogram of the previous image, and wherein the cumulative density function is obtained from the cumulative histogram of the previous image using a reciprocal of a number of points in each region.
17. The method of claim 11 , wherein the weighting factors are calculated in accordance with a weighting factor curve that defines a weighting factor for each calculated variance, and wherein each region has a respective weighting factor curve that depends on a desired level of enhancement.
18. A non-transitory computer readable medium encoded with instructions for enhancing contrast of an image, the computer readable medium comprising instructions for:
generating a histogram based on an image;
determining arithmetic means for at least two regions of the histogram;
calculating variances for each of the determined arithmetic means;
calculating weighting factors based on the calculated variances; and
transforming pixel values of the image using the weighting factors.
19. The non-transitory computer readable medium of claim 18 further comprising instructions for determining the arithmetic means in accordance with the equation
m
i
=
∑
j
y
j
N
,
where y j represents luma of points in a particular region i and N is the total number of points in that region.
20. The non-transitory computer readable medium of claim 19 further comprising instructions for calculating the variances in accordance with the equation
σ
i
=
1
N
∑
j
n
(
y
j
)
(
y
j
-
m
i
)
,
where n(y j ) is a count of the number of pixels of luma y j and σ i is the variance of the ith region.