lab2double (Image Processing Toolbox User's Guide)

Image Processing Toolbox User's Guide

Convert data to double

Syntax

```
labd = lab2double(lab)
```

Description

labd = lab2double(lab) converts an M-by-3 or M-by-N-by-3 array of color values to class double. The output array labd has the same size as lab.

The Image Processing Toolbox follows the convention that double-precision arrays contain 1976 CIE values. arrays that are uint8 or uint16 follow the convention in the ICC profile specification (ICC.1:2001-4, www.color.org) for representing values as unsigned 8-bit or 16-bit integers. The ICC encoding convention is illustrated by these tables.

Value (L*)
uint8 Value
uint16 Value

0.0
0
0

100.0
255
65280

100.0 + (25500/65280)
None
65535

Value (a* or b*)
uint8 Value
uint16 Value

-128.0
0
0

0.0
128
32768

127.0
255
65280

127.0 + (255/256)
None
65535

*Value (L)**	uint8 Value	uint16 Value
`0.0`	`0`	`0`
`100.0`	`255`	`65280`
`100.0 + (25500/65280)`	None	`65535`

*Value (a or b)*	uint8 Value	uint16 Value
`-128.0`	`0`	`0`
`0.0`	`128`	`32768`
`127.0`	`255`	`65280`
`127.0 + (255/256)`	None	`65535`

Class Support

lab is a uint8, uint16, or double array that must be real and nonsparse. labd is double.

Example

Convert full intensity neutral color (white) from uint8 to double.

lab2double(uint8([255 128 128]))
ans =

   100     0     0

For a full list of the toolbox color space conversion functions, see Color Space Conversions.

isrgb lab2uint16