SYSTEM FOR CORRECTING COLOR
IMAGES USING TETRAHEDRAL
INTERPOLATION OVER A HEXAGONAL
 Inventor: Steven J. Harrington, Holley, N.Y.  Assignee: Xerox Corporation, Stamford, Conn.
 Appl. No.: 598,741
 Filed: Feb. 8,1996
Related U.S. Application Data
 Continuation of Ser. No. 297,473, Aug. 29, 1994, abandoned.
 Int. CI.6 G03F3/08
 U.S. CI 358/518; 358/525; 382/276;
 Field of Search 382/276; 358/525,
358/518; 355/38, 77; 348/45; 395/131
 References Cited
U.S. PATENT DOCUMENTS
2,790,844 4/1957 Neugebauer 178/5.2
4,275,413 6/1981 Sakamoto et al 358/525
4,477,833 10/1984 Clark et al 358/80
4,500,919 2/1985 Schreiber 358/78
5,241,373 8/1993 Kanamori et al 358/518
5,305,119 4/1994 Rolleston et al 358/522
5,377,041 12/1994 Spaulding et al 358/518
5,390,035 2/1995 Kasson et al 358/518
Research Report "Tetrahedral Interpolation Technique for Color Space Conversion" James Kasson Jan. 1993 IBM Research Division.
Input device signals Rs, Gs, Bs generated by an image input terminal are converted to colorimetric values Rc, Gc, Bc. The colorimetric values processed by an image processing unit to generate address entries to a lookup table which stores a set of transform coefficients with which the Rc, Gc, Bc values may be processed to convert them to C^, Mx, Yx, Kx colorant signals or any multi-dimensional output color space, which include but are not limited to CMYK or spectral data. Values which are not directly mapped from the table are determined using tetrahedral interpolation over a hexagonal lattice. The hexagonal lattice is formed by offsetting by a half unit every other row of one of its dimensions relative to another dimension. The offset provides closer packing of sample points that define more regular tetrahedrons to reduce relative interpolation errors. The packing also allows for easy lookup table access and simple tests to determine which tetrahedron contains a desired value.