Systems and methods for local tone mapping of high dynamic range images

1. A method comprising:

calculating an illumination component of an image input signal of high dynamic range by processing the input signal with a shape adaptive filter;

processing the calculated illumination component to generate a processed illumination component in which a contrast of the input signal has been compressed;

processing the calculated illumination component to generate a processed reflectance component in which at least one feature of the input signal has been sharpened;

generating a processed input signal from the processed illumination component and the processed reflectance component, wherein the processed input signal comprises a contrast-compressed and image-sharpened version of the input signal; and

mapping the processed input signal into a dynamic range of a display device,

wherein calculating the illumination component further comprises:

calculating the input signal's associated luminance signal by transforming a RGB input signal into a YCgCo color space; and

processing the luminance signal with the shape adaptive filter to calculate the illumination component,

and wherein the shape adaptive filter calculates the illumination component through the equation:

L

⁡

(

x

,

y

)

=

∑

x

i

,

y

i

∈

W

⁢

ω

⁡

(

x

i

,

y

i

)

·

Y

⁡

(

x

i

,

y

i

)

∑

x

i

,

y

i

∈

W

⁢

ω

⁡

(

x

i

,

y

i

)

.

where (x,y) are the co-ordinates of a current pixel being processed, ω(x i , y i ) are filter weights of the shape adaptive filter, Y is a luminance signal, and W is a window of pixels centered at pixel (x,y).

2. The method of claim 1 , wherein ω comprises one or more filter weights of the shape adaptive filter defined by:

ω

⁡

(

x

i

,

y

i

)

=

{

3

,

if

⁢

⁢



Y

⁡

(

x

i

,

y

i

)

-

Y

⁡

(

x

,

y

)



<

c

1

⁢

σ

⁢

2

,

if

⁢

⁢

c

1

⁢

σ

≤



Y

⁡

(

x

i

,

y

i

)

-

Y

⁡

(

x

,

y

)



<

c

2

⁢

σ

1

,

if

⁢

⁢

c

2

⁢

σ

≤



Y

⁡

(

x

i

,

y

i

)

-

Y

(

x

,

y

⁢

)



<

c

3

⁢

σ

0

,

otherwise

⁢

where σ is a filter strength, and c1, c2 and c3 are filter parameters, of the shape adaptive filter.

3. The method of claim 2 , wherein c1=1, c2=2, c3=3, and wherein σ comprises a filter strength of the shape adaptive filter defined by:

σ

=

M

i

⁢

⁢

n

+

1

8

,

where M in is the dynamic range of the input signal.

4. The method of claim 1 , wherein generating the processed input signal comprises:

multiplying the processed illumination component by the processed reflectance component to generate a processed luminance signal; and

transforming the processed luminance signal into a processed red, green, blue signal, wherein the processed red, green, blue signal comprises the processed input signal.

5. A system comprising:

a decomposer operable to receive a high dynamic range (“HDR”) image signal and decompose the image signal into at least an illumination component and a reflectance component;

a contrast compressor operable to receive the illumination component and compress the contrast of the image signal by processing the illumination component;

an image sharpener operable to receive the reflectance component and sharpen an image of the image signal by processing the reflectance component;

a multiplier operable to multiply the processed illumination component by the processed reflectance component to calculate a processed luminance signal;

a RGB processor operable to transform the processed luminance signal into its associated red, green, and blue signals; and

a range mapper operable to map a dynamic range of the associated red, green, and blue signals into a dynamic range of a display device,

wherein the decomposer:

calculates the illumination component through the equation:

L

⁡

(

x

,

y

)

=

∑

x

i

,

y

i

∈

W

⁢

ω

⁡

(

x

i

,

y

i

)

·

Y

⁡

(

x

i

,

y

i

)

∑

x

i

,

y

i

∈

W

⁢

ω

⁡

(

x

i

,

y

i

)

,

where (x,y) are the co-ordinates of a current pixel being processed, ω(x i , y i ) are filter weights of the shape adaptive filter, Y is a luminance signal, and W is a window of pixels centered at pixel (x,y),

calculates the input signal's associated luminance signal by transforming a RGB input signal into a YCgCo color space; and

processes the luminance signal to calculate the illumination component.

6. The system of claim 5 , wherein the decomposer comprises a shape adaptive filter.

7. The system of claim 5 , further comprising:

a RGB to Y converter operable to:

transform the HDR image signal from its associated red, green, and blue signals into at least a luminance signal; and

transmit the luminance signal to the decomposer for decomposing into the illumination component and the reflectance component.

8. The system of claim 7 , wherein the decomposer is further operable to:

pass the luminance signal through a shape adaptive filter to calculate the luminance component; and

calculate the reflectance component by dividing the luminance signal by the illumination component.

9. The system of claim 5 , wherein the display device comprises at least one of a computer monitor, a television, and a display of a laptop computer.

10. The system of claim 5 , wherein the dynamic range of the display device is less than a dynamic range of the HDR image signal.

11. A system for performing local tone mapping of an input image, comprising:

a contrast compressor operable to receive and process an illumination component signal of the input image by:

executing a power function on the illumination component signal to generate an Lp signal; and

mapping a dynamic range of the Lp Signal into a dynamic range of the input image to generate a contrast-compressed signal;

an image sharpener operable to receive and process a reflectance component signal of the input image by:

executing a log function on the reflectance component signal to generate a Log(F) signal;

executing a noise coring function on the Log(F) signal to reduce magnification of noise, thereby generating a noise-cored signal; and

processing the received reflectance component signal and the noise-cored signal to generate a image-sharpened signal; and

a range mapper operable to:

generate a processed image signal from the contrast-compressed signal and the image-sharpened signal; and

map the dynamic range of the processed image signal into a dynamic range of a display device,

wherein the contrast compressor is further operable to:

execute a power function on the illumination component defined by:

P

⁡

(

L

⁡

(

x

,

y

)

)

=

M

i

⁢

⁢

n

·

(

L

⁡

(

x

,

y

)

M

i

⁢

⁢

n

)

[

(

1

-

L

⁡

(

x

,

y

)

M

i

⁢

⁢

n

)

·

γ

Low

+

(

L

⁡

(

x

,

y

)

M

i

⁢

⁢

n

)

·

γ

High

]

,

where L(x,y) is the illumination component at a pixel of interest in the input image, and M in is the dynamic range of the input image.

12. The system of claim 11 , wherein the contrast control parameters of the power function comprise the ranges: 0<γ Low ≦1 and 0<γ High <1.5.

13. The system of claim 11 , wherein the image sharpener is further operable to:

execute a noise coring function defined by:

F

s

⁡

(

x

,

y

)

=

{

0

,

⁢

if

⁢

⁢



log

⁢

⁢

F

⁡

(

x

,

y

)



≤

η

log

⁢

⁢

F

⁡

(

x

,

y

)

-

η

,

if

⁢

⁢

log

⁢

⁢

F

⁡

(

x

,

y

)

>

η

⁢

log

⁢

⁢

F

⁡

(

x

,

y

)

+

η

,

if

⁢

⁢

log

⁢

⁢

F

⁡

(

x

,

y

)

<

-

η

where (x,y) is a pixel of interest, and q is a variable of the log function.

14. The system of claim 13 , wherein the variable η of the log function comprise the range 0<η<0.5.

15. The system of claim 11 , wherein the range mapper is further operable to:

execute a standard gamma correction function of the display device on the processed image signal to map the dynamic range of the processed image signal into the dynamic range of the display device.

16. The system of claim 11 , wherein the input image comprises a high dynamic range image.