TMC2272A datasheet(13/20 Pages) FAIRCHILD

13 / 20 page

TMC2272A

PRODUCT SPECIFICATION

REV. 1.1.3 10/25/00

Since Y = 256, U = V = 0, and G = B = R = 255 for saturated

white output, every Y coefﬁcient will be 255 / 255=1.0. The

full matrix for YUV to GBR conversion is:

G = 1.0 (Y)

-0.3443 (U)

-0.7142 (V)

B = 1.0 (Y)

+1.7727 (U)

R = 1.0 (Y)

+1.3965 (V)

Since the largest element is over 1.0 and the largest permissi-

ble coefﬁcient is 511, multiply all elements of the matrix by

256 to obtain the values to load into the TMC2272A:

G =

256 (Y')

- 88 (U')

- 83 (V')

dec.

100

3A8

349

hex

B =

256 (Y')

+ 454 (U')

dec.

100

1C6

hex

R =

256 (Y')

+ 359 (V')

dec.

100

167

hex

For the YUV to RGB conversion, the TMC2272A input data

alignment for 10-bit Y'U'V' is:

Y9Y8Y7Y6Y5Y4Y3Y2Y1Y00

U9 U9 U8 U7 U6 U4 U4 U3 U2 U1 U0 0

V9 V9 V8 V7 V6 V5 V4 V3 V2 V1 V0 0

where the doubled U9 and V9 sign bits denote two’s comple-

ment sign extensions. The TMC2272A output data align-

ment for 8-bit GBR is then:

0000G7 G6 G5 G4 G3 G2 G1 G0

0000B7 B6 B5 B4 B3 B2 B1 B0

0000R7 R6 R5 R4 R3 R2 R1 R0

Note that the inputs have to be doubled because the coefﬁ-

cient gain is 256, whereas the internal gain is 1 / 512, for a

net gain of 1/2.

Table 5. Summary of Colorspace Conversion Coefﬁcients

Table 6. Conversion Port Assignments and Alignments

Where XY-0 denotes right-justified, (e) denotes sign extension, and XY-1 denotes shifted one bit leftward from a right-justified

position.

Conversion

KAX

KAY

KAZ

KBX

KBY

KBZ

KCX

KCY

KCZ

RGB to YUV

099

3AA

100

12D

356

32A

03A

100

3D6

RGB to YCBCR

083

3B4

0E1

102

36B

344

032

0E1

3DB

YUV to RGB

100

000

3A8

1C6

167

349

000

YCBCR to RGB

149

000

3CE

102

0CC

398

000

Port

AIN

BIN

CIN

XOUT

YOUT

ZOUT

RGB to YUV

R7-0

G7-0

B7-0

Y7-0

U7-0(e)

V7-0(e)

RGB to YCBCR

R7-0

G7-0

B7-0

Y9-0

U9-0(e)

V9-0(e)

YUV to RGB

Y8-1(e)

U8-1(e)

V8-1(e)

R7-0

G7-0

B7-0

YCBCR to RGB

Y9-0

CB9-0(e)

CR9-0(e)

R7-0

G7-0

B7-0

HSV (HSI) Format Conversions

HSV (or HSI) refers to Hue (color), Saturation (vividness),

and Value (intensity or brightness), quantities which are

directly related to the human perception of light and color.

The V (or I) levels are simply the Y (or luminance) levels.

Hue and Saturation are derived from the R-Y and B-Y color

difference values of a signal.

HSV Calculations:

Value (V) = Intensity (I) = Y

Hue (H) = Arctan (B-Y/R-Y)

Saturation (S) =

R-Y = S*cos(H)

B-Y = S*sin(H)

One may use two 64Kx8 ROM look-up-tables to calculate

Hue and Saturation from R-Y and B-Y in an 8-bit system.

However, the ﬁnite size of this LUT may limit performance,

especially if the TMC2272A’s full precision is used. The

TMC2330A, developed to translate between rectangular and

polar coordinates, can perform the trigonometric transforma-

tions to 16 bit precision at 50MHz. These calculations are

the same as required in HSV calculations. A 4 Gigabyte x 32

bit LUT can achieve the same accuracy and precision as the

TMC2330A, if it is programmed correctly.

–

()

–

()

TMC2272A Datasheet(PDF) 13 Page - Fairchild Semiconductor

TMC2272A Datasheet(HTML) 13 Page - Fairchild Semiconductor

Html Pages

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

Datasheet Download

Link URL