Method and apparatus for lossless and minimal-loss color conversion
View Patent ↗A method and apparatus for perfectly lossless and minimal-loss interconversion of digital color data between spectral color spaces (RGB) and perceptually based luma-chroma color spaces (Y′C B C R ) is disclosed. In particular, the present invention provides a process for converting digital pixels from R′G′B′ space to Y′C B C R space and back, or from Y′C B C R space to R′G′B′ space and back, with zero error, or, in constant-precision implementations, with guaranteed minimal error. This invention permits digital video editing and image editing systems to repeatedly interconvert between color spaces without accumulating errors. In image codecs, this invention can improve the quality of lossy image compressors independently of their core algorithms, and enables lossless image compressors to operate in a different color space than the source data without thereby becoming lossy. The present invention uses fixed-point arithmetic with signed and unsigned rounding normalization at key points in the process to maintain reversibility.
1. A method of converting digital image data from a first format wherein colors are represented in a first color space to a second format wherein colors are represented in a second color space distinct from the first color space, the method comprising:
converting a first representation of the digital image data from the first color space to a second representation of the digital image data in the second color space;
converting the second representation of the digital image data from the second color space to a third representation of the digital image data in the first color space;
producing intermediate image data values, wherein a range of the values is centered about zero;
rounding the intermediate image data values, wherein rounding depends on a sign of an intermediate image data value, such that negative intermediate image data values are rounded upward and positive intermediate image data values are rounded downward; and
normalizing the intermediate image data values.
2. The method of claim 1 , wherein the steps of converting comprise using fixed-point multiplication and addition, wherein a conversion error, being any difference between the first representation of the digital image data and the third representation of the digital image data, is minimized.
3. The method of claim 2 , wherein the steps of converting further comprise augmenting precision of the second representation of the digital image data in the second color space, whereby the augmenting of precision in the second representation reduces the conversion error to zero.
4. The method of claim 1 , wherein the steps of converting comprise factoring a transform of digital image data into sequential steps, whereby the factoring reduces precision of the intermediate image data values and reduces an amount of computation.
5. The method of claim 4 , wherein the steps of converting further comprise omitting any multiplications by one and zero, whereby the amount of computation is further reduced by the omissions.
6. The method of claim 4 , wherein the step of converting the first representation of the digital image data further comprises a plurality of rounding and normalizing steps in a plurality of chroma converters, whereby the precision of intermediate image data values is further reduced.