(a) Clipping to the colour on the reproduction gamut boundary, which has the smallest distance in XYZ.For the compression methods the author suggests the use of piece?wise linear functions whereby there is no compression for the first segment, some compression for the second segment and clipping for the third segment.(b) Clipping along lines towards the centre of the lightness axis.
(c) Clipping along lines of constant lightness and hue.
(d) Clipping, whereby gamut mapping is subjectively defined for 26 points. For each of these points a centre-of-gravity is defined as the intersection of the lightness axis and the line connecting the original and gamut-mapped colours. The centre-of-gravity towards which a given colour is clipped is then calculated as a weighted average of the centres-of-gravity for the 26 points.
(e) Compression towards the centres-of-gravity as suggested in method (d).
(f) Compression as suggested by Johnson (1979).
Reproductions made with these six GMAs were subjectively evaluated by
the author alone who judged the clipping to the closest colour (method
(a)) to give the best results, but noted objectionable hue shifts with
this method. Clipping towards the centre was judged to darken highlights
and lighten shadows too much and clipping along lines of constant lightness
and hue angle was judged as giving too light reproductions.