Image processing method for converting non-linear RGB image data to L*a*b* image data

1. An image processing method comprising the steps of:

linearizing a Color Filter Array image having non-linear RGB pixel data to generate a linearized image, the step of linearizing using at least one lookup table for each of the three RGB colors, the step of linearizing further performed using an address for current RGB pixel data being read to index the lookup tables and incrementing the addresses for the RGB pixel data;

generating a histogram for each of three RGB image color planes of the linearized image data, and storing at least a high and a low threshold for each histogram;

planarizing the linearized image using the high and the low thresholds to generate a planarized linear RGB image; and

transforming the planarized linear RGB image from planarized linear RGB form to L*a*b* form, wherein

the step of transforming the planarized linear RGB image from planarized linear RGB form to L*a*b* form performs, on the planarized linear RGB image, additional steps of white balancing, range expansion, resampling, color conversion, and sharpening, the additional steps reading and writing to a plurality of buffers such that the additional steps are performed on the planarized linear RGB image in parallel.

2. An image processing method according to claim 1 , wherein the step of transforming said image from planarized linear RGB form to L*a*b* form references each of the histograms and applies a thresholding algorithm thereto.

3. An image processing method according to claim 1 , wherein the step of transforming said image from planarized linear RGB form to L*a*b* form is a tri-linear interpolation utilizing three lookup tables respectively for RGB to L*, RGB to a*, and RGB to b*.

4. An image processing method according to claim 1 , wherein the four least significant bits of the address are used to interpolate between values retrieved from the lookup tables.