Gamma Correction - Analog Devices ADV8003 Hardware Manual

8.4.17. Gamma Correction

Generally, gamma correction is applied to compensate for the nonlinear relationship between the signal input and the output brightness level
(as perceived on a CRT). It can also be applied wherever nonlinear processing is used.
Gamma correction uses the function:
Signal = (Signal )
γ
OUT IN
where γ is the gamma correction factor.
Gamma correction is available for SD and ED/HD video. For both variations, there are twenty 8-bit registers, used to program the Gamma
Correction Curve A and Gamma Correction Curve B.
Gamma correction is performed on the luma data only. The user can choose one of two correction curves, Curve A or Curve B. Only one
of these curves can be used at a time.
The shape of the gamma correction curve is controlled by defining the curve response at 10 different locations along the curve. By altering
the response at these locations, the shape of the gamma correction curve can be modified. Between these points, linear interpolation is
used to generate intermediate values. Considering that the curve has a total length of 256 points, the 10 programmable locations are at the
following points: 24, 32, 48, 64, 80, 96, 128, 160, 192, and 224. The following locations are fixed and cannot be changed: 0, 16, 240, and
255.
From the curve locations, 16 to 240, the values at the programmable locations and, therefore, the response of the gamma correction curve,
should be calculated to produce the following result:
x = (x )
γ
DESIRED INPUT
where:
x is the desired gamma corrected output.
DESIRED
x is the linear input signal.
INPUT
γ is the gamma correction factor.
To program the gamma correction registers, the 10 programmable curve values are calculated using
where:
γ is the value to be written into the gamma correction register for point n on the gamma correction curve.
n
n = 24, 32, 48, 64, 80, 96, 128, 160, 192, or 224.
γ is the gamma correction factor.
For example, setting γ = 0.5 for all programmable curve data points results in the following y
y = [(8/224) × 224] + 16 = 58
0.5
24
y = [(16/224) × 224] + 16 = 76
0.5
32
y = [(32/224) × 224] + 16 = 101
0.5
48
y
= [(48/224) 0.5 × 224] + 16 = 120
64
y
= [(64/224) 0.5 × 224] + 16 = 136
80
y
= [(80/224) 0.5 × 224] + 16 = 150
96
y
= [(112/224) 0.5 × 224] + 16 = 174
128
y
= [(144/224) 0.5 × 224] + 16 = 195
160
y = [(176/224) × 224] + 16 = 214
0.5
192
y = [(208/224) × 224] + 16 = 232
0.5
224
Rev. B, August 2013
γ

n −
16
 

γ = ×
 

n
−
 240 16 

Equation 26: Gamma Correction Calculation
329
Equation


( 240 − 16 ) + 16


values:
n
ADV8003 Hardware Manual
26.