Adaptive piece-wise approximation method for gamma correction

• US 6,292,165 B1
• Filed: 08/13/1999
• Issued: 09/18/2001
• Est. Priority Date: 08/13/1999
• Status: Expired due to Fees

• Current Assignee: Transpacific IP I Limited (Transpacific IP Group Limited)
• Original Assignee: Industrial Technology Research Institute
• Inventors: Lin, Tsu-Ping, Kung, Chen-Pan, Cheng, Han-Min
• Primary Examiner(s): Liang, Regina
• Application Number:
  US09/373,565
• Time in Patent Office: 767 Days
• Field of Search: 345/147-149, 345/89, 345/63, 345/12, 345/20, 348/671, 348/673, 348/674, 348/687, 358/519, 358/521
• US Class Current: 345/690
• CPC Class Codes: G09G 2320/0276 for the purpose of adaptation to the characteristics of a display device, i.e. gamma correction
• CPC Class Codes: G09G 5/10 Intensity circuits
• CPC Class Codes: H04N 5/202 Gamma control circuits for controlling camera response irrespective of the scene brightness H04N23/82

Abstract

An adaptive piece-wise approximation method for gamma correction is disclosed. The method includes an adaptive way for segmenting the brightness intensity input range and after the segmentation, a set of selected sampling points is taken. A simplified gamma correction circuit is then designed to omit the subtraction and the division calculation in the gamma correcting function by choosing the segmentation points appropriately. The improved gamma correction circuit comprises a multiplexer, a subtractor, an offset block, a multiplier, a division block, and an adder. An intensity signal at the set of selected input points is sampled and the sampled values are stored in providing a top value and a bottom value for the video intensity value to be gamma corrected. The video intensity value is then sent into the gamma correction circuit to get a gamma corrected data word representing pixel intensity for its respective input video intensity value.

Claims

1. An improved piece-wise method for providing gamma correction of an input to display means, said method comprising the steps of:

providing a gamma correction curve having an end that is almost linear and an opposite end that is almost curving;

taking a zero level input and a maximum level input as two end points of an input range;

center parting said input range to form two equivalent segments;

center parting said two equivalent segments to form four equivalent segments;

further segmenting said gamma correction curve by using an adaptive segmentation method; and

repeat the previous stage until a desired width for the smallest segment is achieved.

2. The method in accordance with claim 1, wherein said adaptive segmentation method comprises a step of center parting two smallest segments of said input range.

3. The method in accordance with claim 1, wherein said adaptive segmentation method comprises a step of center parting more than two smallest segments of said input range.

4. The method in accordance with claim 2, wherein said adaptive segmentation method further comprises a combination of said step of center parting two smallest segments of said input range and said step of center parting more than two smallest segments of said input range.

5. The method in accordance with claim 1, wherein said center parting means dividing a segment into two smaller segments with equal length.

6. The method in accordance with claim 1, wherein said gamma correction curve can be in any style, as long as it is changing from almost curving to almost linear along said input range.

7. The method in accordance with claim 1, wherein said gamma correction curve can be in any style, as long as it is changing from almost linear to almost curving along said input range.

8. The method in accordance with claim 6, wherein said changing comprises changes in gradient from large to small.

9. The method in accordance with claim 7, wherein said changing comprises changes in gradient from small to large.

10. The method in accordance with claim 8, wherein said desired width can be as small as two but not less than, and it is heavily depending on the gradient value, the larger the value the smaller the width is required.

11. A circuit for providing gamma correction of an input IN in between two sampled points, A and B, to display means in accordance with a gamma correcting function G(IN), where G(IN)=(IN−

A)*(G(B)−

G(A))/(B−

A)+G(A), G(A) and G(B) are programmed values for their respective sampled points, A and B, said circuit comprising;

selector means responsive to upper bits of said input for choosing a top value G(B) and a bottom value G(A) for said input IN;

subtracting means responsive to said top value and said bottom value for subtracting said bottom value from said top value;

truncating means responsive to said input for removing a few upper bits of said input;

multiplier means responsive to an output of said subtracting means and an output of said truncating means for generating their product;

shifter means responsive to upper bits of said input and an output of said multiplier means for carrying out a division function; and

adder means responsive to said bottom value and an output of said shifter means for generating a gamma corrected data word representing pixel intensity.

12. The circuit in accordance with claim 11, wherein said selector means comprises a multiplexer having stored sampling values as its inputs.

13. The circuit in accordance with claim 11, wherein said subtracting means comprises an adder means and an inverting input port.

14. The circuit in accordance with claim 11, wherein said truncating means performs the subtracting function of IN−

A in said gamma correcting function.

15. The circuit in accordance with claim 11, wherein said shifter means performs the division function of (G(B)−

G(A))/(B−

A) in said gamma correcting function by right shifting said output of said multiplier means.

16. A method for gamma correcting an input video intensity value IN in between two sampled points, A and B, to display means in accordance with a gamma correcting function G(IN), where G(IN)=(IN−

A)*(G(B)−

G(A))/(B−

A)+G(A), G(A) and G(B) are programmed values for their respective sampled points, A and B, said method comprising the steps of;

providing a gamma correction circuit in accordance with said gamma correcting function;

applying an improved piece-wise method to select a set of input points to be sampled;

sampling an intensity signal at said selected set of input points;

storing sampled values to provide said programmed values G(A) and G(B) for said gamma correction circuit;

sending said input video intensity value IN into said gamma correction circuit;

selecting a top value G(B) and a bottom value G(A) for said input video intensity value IN;

subtracting said bottom value from said top value and outputs a difference value;

removing a few upper bits of said input video intensity value and outputs to a truncated value;

multiplying said difference value and said truncated value, the outcome is sent to a product value;

right shifting said product value to a length that is equivalent to the length of said input intensity video value, and outputs the result to a shifted value; and

adding said bottom value and said shifted value to produce a gamma corrected data word G(IN) representing pixel intensity for its respective input video intensity value IN.

17. The method in accordance with claim 16, wherein said gamma correction circuit comprises a selector means, a subtracting means, a truncating means, a multiplier means, a shifter means, and an adder means.

18. The method in accordance with claim 17, wherein said selector means responsive to upper bits of said input video intensity value for choosing said top value G(B) and said bottom value G(A) for said input video intensity value IN, furthermore, said selector means comprises a multiplexer having stored sampling values as its inputs.

19. The method in accordance with claim 17, wherein said subtracting means comprises an adder means and an inverting input port.

20. The method in accordance with claim 17, wherein said truncating means responsive to said input video intensity value for performing the subtracting function of IN−

A in said gamma correcting function.

21. The method in accordance with claim 17, wherein said shifter means responsive to upper bits of said input video intensity value and said product value for carrying out the division function of (G(B)−

G(A))/(B−

A) in said gamma correcting function.

22. The method in accordance with claim 16, wherein said improved piece-wise method comprising the steps of:

providing a gamma correction curve having an end that is almost linear and an opposite end that is almost curving;

taking a zero level input and a maximum level input as two end points of an input range;

center parting said input range to form two equivalent segments;

center parting said two equivalent segments to form four equivalent segments;

further segmenting said gamma correction curve by using an adaptive segmentation method; and

repeating the previous stage until a desired width for the smallest segment is achieved.

23. The method in accordance with claim 22, wherein said adaptive segmentation method comprises a step of center parting two smallest segments of said input range.

24. The method in accordance with claim 22, wherein said adaptive segmentation method comprises a step of center parting more than two smallest segments of said input range.

25. The method in accordance with claim 23, wherein said adaptive segmentation method further comprises a combination of said step of center parting two smallest segments of said input range and said step of center parting more than two smallest segments of said input range.

26. The method in accordance with claim 22, wherein said center parting means dividing a segment into two smaller segments with equal length.

27. The method in accordance with claim 22, wherein said gamma correction curve can be in any style, as long as it is changing from almost curving to almost linear along said input range.

28. The method in accordance with claim 22, wherein said gamma correction curve can be in any style, as long as it is changing from almost linear to almost curving along said input range.

29. The method in accordance with claim 27, wherein said changing comprises changes in gradient from large to small.

30. The method in accordance with claim 28, wherein said changing comprises changes in gradient from small to large.

31. The method in accordance with claim 29, wherein said desired width can be as small as two but not less than, and it is heavily depending on the gradient value, the larger the value the smaller the width is required.