Method and apparatus for automatic brightness adjustment of image signal processor
An automatic brightness adjusting method and apparatus for image signal processor (ISP) is provided. The image processing apparatus may include a histogram generating unit, a cumulative distribution function calculator, and a histogram equalization (HE) unit. The histogram generating unit may generate a histogram of brightness values of pixels in an input image. The cumulative distribution function calculator may generate a cumulative distribution function and an inverse cumulative distribution function, based on the generated histogram. The HE unit may generate a conversion function based on the cumulative distribution function and the inverse cumulative distribution function, and may apply HE to the input image based on the conversion function so as to generate an output image.
1. A method of automatically adjusting a brightness of an image, the method comprising:
generating a histogram of brightness values of pixels of an input image;
generating a conversion function based on a cumulative distribution function and an inverse cumulative distribution function of the histogram; and
generating an output image by applying, based on the conversion function, histogram equalization (HE) to the input image;
wherein the HE is based at least partly on the cumulative distribution function or the inverse cumulative distribution function, a total numbers of pixels used for generating the histogram, and a maximum brightness value of the brightness values of the pixels of the input image; and
wherein the inverse cumulative distribution function is generated according to Equation 2:
CDF inv ( n )=Pixel total −CDF( n )+Hist( n ), Equation 2
where CDF inv denotes the inverse cumulative distribution function, Pixel total denotes a number of pixels used for generating the histogram, CDF denotes the cumulative distribution function, Hist denotes the histogram, and n denotes an integer greater than or equal to 0 and less than or equal to a maximum value of the brightness values.
2. The method of claim 1 , wherein the generating of the histogram comprises:
sampling the pixels of the input image; and
generating the histogram based on the sampled pixels.
3. The method of claim 1 , wherein the generating of the conversion function comprises:
generating the cumulative distribution function and the inverse cumulative distribution function, based on the histogram; and
generating the conversion function based on the cumulative distribution function and the inverse cumulative distribution function.
4. The method of claim 1 , wherein the cumulative distribution function is generated according to Equation 1:
C
D
F
(
n
)
=
∑
k
=
0
n
Hist
(
k
)
,
[
Equation
1
]
where CDF denotes the cumulative distribution function, Hist denotes the histogram, k denotes a brightness value, and n denotes an integer greater than or equal to 0 and less than or equal to a maximum value of the brightness values.
5. The method of claim 1 , wherein the inverse cumulative distribution function is generated based on a number of pixels used for generating the histogram, the cumulative distribution function, and the histogram.
6. A method of automatically adjusting a brightness of an image, the method comprising:
generating a histogram of brightness values of pixels of an input image;
generating a conversion function based on a cumulative distribution function and an inverse cumulative distribution function of the histogram; and
generating an output image by applying, based on the conversion function, histogram equalization (HE) to the input image;
wherein the HE is based at least partly on the cumulative distribution function or the inverse cumulative distribution function, a total numbers of pixels used for generating the histogram, and a maximum brightness value of the brightness values of the pixels of the input image;
wherein the conversion function is generated based on a first HE function that is based on the cumulative distribution function and a second HE function that is based on the inverse cumulative distribution function; and
wherein the second HE function is generated according to Equation 4:
HE
inv
(
n
)
=
Pixel
Total
-
C
D
F
inv
(
n
)
Pixel
total
×
B
max
,
[
Equation
4
]
where HE inv denotes the second HE function, CDF inv denotes the inverse cumulative distribution function Pixel total denotes a number of pixels used for generatin the histogram, B max denotes a maximum value of the brightness values of the pixels of the input image, and n denotes an integer greater than or equal to 0 and less than or equal to B max .
7. The method of claim 6 , wherein the first HE function is generated according to Equation 3:
HE
(
n
)
=
C
D
F
(
n
)
Pixel
total
×
B
max
,
[
Equation
3
]
where HE denotes the first HE function, CDF denotes the cumulative distribution function, Pixel total denotes a number of pixels used for generating the histogram, B max denotes a maximum value of the brightness values of the pixels of the input image, and n denotes an integer greater than or equal to 0 and less than or equal to B max .
8. The method of claim 6 , wherein the conversion function is a weighted-sum of the first HE function and the second HE function.
9. The method of claim 6 , wherein the conversion function is generated according to Equation 5:
Proposed_HE
(
n
)
=
HE
(
n
)
2
+
HE
inv
(
n
)
2
,
[
Equation
5
]
where Proposed_HE denotes the conversion function, HE denotes the first HE function, HE inv denotes the second histogram, and n denotes an integer greater than or equal to 0 and less than or equal to a maximum value of the brightness values.
10. The method of claim 1 , wherein the conversion function is generated according to Equation 6:
Proposed_HE
(
n
)
=
Pixel
total
+
C
D
F
(
n
)
-
C
D
F
inv
(
n
)
Pixel
total
×
2
×
B
max
,
[
Equation
6
]
where Proposed_HE denotes the conversion function, Pixel total denotes a number of pixels used for generating the histogram, CDF denotes the cumulative distribution function, CDF inv denotes the inverse cumulative distribution function, B max denotes a maximum value of the brightness values of the pixels of the input image, and n denotes an integer greater than or equal to 0 and less than or equal to B max .
11. The method of claim 1 , wherein the conversion function is generated according to Equation 7:
Proposed_HE
(
n
)
=
C
D
F
(
n
)
-
Hist
(
n
)
2
Pixel
total
×
B
max
,
[
Equation
7
]
where Proposed_HE denotes the conversion function, Pixel total denotes a number of pixels used for generating the histogram, CDF denotes the cumulative distribution function, Hist denotes the histogram, B max denotes a maximum value of the brightness values of the pixels of the input image, and n denotes an integer greater than or equal to 0 and less than or equal to B max .
12. The method of claim 1 , wherein the conversion function is configured as a look-up table.
13. An image processing apparatus, the apparatus comprising:
a histogram generating unit to generate a histogram of brightness values of pixels of an input image;
a cumulative distribution function calculator to generate a cumulative distribution function and an inverse cumulative distribution function; and
a histogram equalization (HE) unit to generate a conversion function based on the cumulative distribution function and the inverse cumulative distribution function, and to generate an output image by applying, based on the conversion function, HE to the input image;
wherein the HE is based at least partly on the cumulative distribution function or the inverse cumulative distribution function, a total numbers of pixels used for generating the histogram, and a maximum brightness value of the brightness values of the pixels of the input image; and
wherein the inverse cumulative distribution function is generated according to Equation 2:
CDF inv ( n )=Pixel total −CDF( n )+Hist( n ), Equation 2
where CDF inv denotes the inverse cumulative distribution function, Pixel total denotes a number of pixels used for generating the histogram, CDF denotes the cumulative distribution function, Hist denotes the histogram, and n denotes an integer greater than or equal to 0 and less than or equal to a maximum value of the brightness values.
14. The apparatus of claim 13 , wherein:
the histogram generating unit comprises a dual-port static random access memory (SRAM), and
the HE unit comprises a single-port SRAM.
15. The apparatus of claim 14 , wherein a brightness value of a pixel of the input image is input as an address of the dual-port SRAM, and a data value of the address of the dual-port SRAM is increased by 1 due to the input.
16. The apparatus of claim 13 , wherein the HE unit applies, based on the conversion function, HE to the input image by utilizing a look-up table.
17. The apparatus of claim 13 , wherein the cumulative distribution function calculator applies a shift operation to a value indicating a number of pixels of the input image so as to calculate an approximate value of a number of pixels used for generating the histogram.
18. The apparatus of claim 13 , wherein the histogram generating unit performs sampling of the pixels of the input image, and generates the histogram based on the sample pixels.