Image enhancement methods and apparatus therefor

• US 7,302,110 B2
• Filed: 04/10/2002
• Issued: 11/27/2007
• Est. Priority Date: 05/02/2001
• Status: Expired due to Term

• Current Assignee: Arm Limited (SoftBank Group Corp.)
• Original Assignee: Apical Limited (SoftBank Group Corp.)
• Inventors: Chesnokov, Vyacheslav
• Primary Examiner(s): Wu; Jingge
• Assistant Examiner(s): Krasnic; Bernard
• Application Number:
  US10/476,587
• Publication Number:
  US 20040213478A1
• Time in Patent Office: 2,057 Days
• Field of Search: 382/232, 382/233, 382/261, 382/308, 382/276, 382/260, 382/264, 382/274, 382/210, 345/555, 345/20, 345/60, 345/77, 348/384.1, 348/440.1, 358/426.1, 358/426.16, 375/122, 375/240.1
• US Class Current: 382/264
• CPC Class Codes: G06T 2207/20208 High dynamic range [HDR] image processing
• CPC Class Codes: G06T 5/40 using histogram techniques
• CPC Class Codes: G06T 5/92 based on global image properties
• CPC Class Codes: H04N 19/46 Embedding additional information in the video signal during the compression process H04N19/517, H04N19/68, H04N19/70 take precedence
• CPC Class Codes: H04N 19/80 Details of filtering operations specially adapted for video compression, e.g. for pixel interpolation H04N19/635, H04N19/86 take precedence
• CPC Class Codes: H04N 19/85 using pre-processing or post-processing specially adapted for video compression

Abstract

The invention relates to methods, devices and computer programs for processing images. The dynamic range of an image is adjusted to allow more detail in the original image to be reproducible, whilst preserving the natural look of the image. Improvements to lossy data compression/decompression algorithms are proposed which include elements of spatial non-uniformity, for example modifying relatively dark portions of an image to a greater extent than relatively bright portions of the image. The algorithms may be made reversible by storing parameters describing the compression process with a compressed image, which can then be used to reverse the process when decompressing the image.

Claims

1. An image processing method comprising a transformation step of processing an input image signal to generate an adjusted output image signal, wherein the intensity values I(x,y) for different positions (x,y) of an image are adjusted to generate an adjusted intensity value I_out(x,y) in accordance with:

I_out=Σ

_i=0^Nα

_i(I)LPF<sub>Ω

i</sub>[P_i(F(I))]·

Q_i(F(I))+β

(I),

where the P_i(γ

) form an orthogonal basis of functions of γ

defined in a range 0<

γ

<

1, the Q_i(..) are antiderivatives of the P_i(..);

the Q_i(F(I))=∫

₀^F(i)=P_i(η

)dη

or approximations thereto, the LPF_Ω

[..] is an operator of low-pass spatial filtering;

the Ω

_i are cut-off frequencies of the low-pass spatial filter, the F(..) is a weighting function, the α

_i(l) are i^th members of a transformation step strength function, the β

(I) is a strength function, the i and N are positive constants and the η

is a variable.

2. The image processing method according to claim 1, wherein the β

(I) is not equal to zero.

3. The image processing method according to claim 1, wherein the Ω

_i is different for different i such that the transformation step has a different degree of spatial nonuniformity in different brightness domains.

4. The image processing method according to claim 1, wherein

I_out=α

·

Σ

_i=0^NLPF_Ω

[P_i(F(I))]·

Q_i(F(I))+(1+α

)I,

where the α

is a transformation step strength value such that 0<

α

<

1.

5. The image processing method according to claim 1, wherein said orthogonal basis of functions are polynomial functions.

6. The image processing method according to claim 5, wherein said orthogonal basis of functions are Legendre polynomial functions.

7. The image processing method according to claim 4, wherein said orthogonal basis of functions are piecewise linear mapping functions.

8. The image processing method according to claim 4, wherein said weighting function F(I) is selected to provide a greater strength of transformation step in dark regions than in light regions.

9. The image processing method according to claim 8, wherein said weighting function varies substantially in accordance with:

10. The method according to claim 1, wherein the transformation step is representable by a transfer function which varies in shape nonlinearly in a spatial domain.

11. The method of claim 1, further comprising a method of compressing an image signal, comprising conducting reversible dynamic range compression according to the method of claim 1, and lossy data compression during an image data compression procedure to generate compressed image data.

12. The method according to claim 11, comprising conducting a reversible dynamic range compression algorithm followed by a lossy image data compression algorithm.

13. The method according to claim 11, comprising generating parameters relating to the dynamic range compression algorithm, and storing the generated parameters in combination with the compressed image data.

14. The method of claim 1, further comprising a method of decompressing compressed image data, comprising conducting decompression, corresponding to a lossy compression, and a reverse of the dynamic range compression according to the method of claim 1 in combination to provide a decompressed image signal.

15. The decompressing method according to claim 14, comprising reading parameters relating to said reversible dynamic range compression algorithm from data accompanying said compressed image data, and using said parameters in said reverse dynamic range compression algorithm.

16. The decompressing method according to claim 14, comprising conducting said reverse dynamic range compression using one or more parameters selected in dependence on an image display method to be utilized.

17. A method according to claim 11, wherein said dynamic range compression algorithm comprises processing an input image signal to generate an adjusted output image signal, wherein the intensity values I(x,y) for different positions (x,y) of an image are adjusted to generate an adjusted intensity value I_out(x,y) in accordance with:

I_out=Σ

_i=0^Nα

_i(I)LPF<sub>Ω

i</sub>[P_i(F(I))]·

Q_i(F(I))+β

(I),

where the P_i(γ

) form an orthogonal basis of functions of γ

defined in a range 0<

γ

<

1, the Q_i(..) are antiderivatives of the P_i(..);

the Q_i(F(I))=∫

₀^F(I)P_i(η

)dη

or approximations thereto, the LPF_Ω

[..] is an operator of low-pass spatial filtering;

the Ω

_i are cut-off frequencies of the low-pass spatial filter, the F(..) is a weighting function, the α

_i(I) are i^th members of a transformation step strength function, the β

(I) is a strength function, the i and N are positive constants and the η

is a variable.

18. A computer program adapted to implement the method of claim 11.

23. An image processing device comprising a processor adapted to carry out the method of claim 1.

19. An image processing method comprising the step of processing an input image signal to generate an adjusted output image signal, wherein the intensity values (I) for different positions of an image are adjusted to generate an adjusted intensity value (I′

) in accordance with;

20. The image processing method according to claim 19, wherein the input image signal is a video signal represented by I(t) and the arithmetic mean is:

21. The image processing method according to claim 19, wherein said orthogonal functions comprise a basis of square functions, or approximations thereto.

22. The image processing method according to claim 19, wherein said orthogonal functions comprise Legendre polynomials, or approximations thereto.