Source colour
The single colour result appears in the output panel. Patch grid uses all loaded patches simultaneously.
Illuminants
Keyboard S swaps illuminants. Select "Custom CCT..." to reveal the temperature slider.
CAT method
Keyboard 1-8 cycles methods. CAT97 uses the CIECAM97s sharpened matrix; CAT94 uses HPE cone primaries with S=Z.
Adaptation degree - 100%
0% = no adaptation. 100% = full. CAT97 introduces moderate sharpening with cross-channel S coupling. Keyboard R resets to 100%.
Colour patches
Paste hex codes or select a built-in set. Click Apply to use custom patches.
Keyboard shortcuts
1-8 cycle CAT  |  S swap illuminants  |  R reset degree  |  E export CSV  |  C compare methods
White-point summary
Source illuminant
D65 - - -
→
Destination illuminant
A - - -
Single colour result
Source
#735244
→
Adapted
#735244
dE00-
dE76-
dE94-
Src XYZ-
Dst XYZ-
Src L*a*b*-
Dst L*a*b*-
Src LCH-
Dst LCH-
Gamut-
dE00 patch statistics
Min-
Max-
Mean-
Median-
Std Dev-
Patches-
Clipped-
Mean dE76-
Mean dE94-
Patch grid - source vs adapted
Source under D65
Adapted to A
Red borders indicate gamut-clipped patches.
SPD spectral power distribution (380-780 nm)
Blue = source SPD | Orange = destination SPD over 380-780 nm.
CIE 1931 chromaticity diagram
Spectral locus | sRGB gamut triangle (gold dashed) | Planckian locus | WP markers.
dE00 distribution histogram
Full adaptation matrix (M⁻¹ D M)
Loading...
det(M)-
k (condition)-
||M||_F-
Determinant near 1.0 confirms invertibility. Low condition number = numerically stable.
Transform matrices - M and M⁻¹
M (XYZ to LMS)
M⁻¹ (LMS to XYZ)
LMS cone-channel analysis
Scale factors (D)
L-
M-
S-
Src WP LMS
L-
M-
S-
Dst WP LMS
L-
M-
S-
CAT97 S row = [0.0357, 0.0469, 0.9174] — moderate cross-channel coupling unlike HPE's pure S=Z. Scale factors = dst/src cone ratios modulated by degree.
Per-patch results table
# Src Dst Src Hex Adapted dE00 dE76 dE94 Shift Gamut
Loading...
Export and share
CSV includes all patches with dE00/76/94 and gamut status. JSON includes full matrix diagnostics.
Multi-method comparison

Compare all 8 CAT methods against the same patch set and illuminant pair. Shows mean/median/min/max dE00, condition number, and determinant.

Method Mean dE00 Median Min Max Std Dev k(M) det(M) Best
Click "Compare all 8 methods" to run analysis.
Star marks lowest mean dE00. Keyboard C triggers comparison.
CIECAM97s, CAT97, and Chromatic Adaptation Standards
▶ CIECAM97s — CIE 1997 Colour Appearance Model

CIECAM97s (CIE TC 1-34, 1998; Luo & Hunt, 1998) was the first CIE-recommended colour appearance model for general use. It refined the earlier CIECAM94 with improved correlates of brightness (Q), colourfulness (M), saturation (s), lightness (J), and chroma (C).

Chromatic adaptation: CIECAM97s introduced the CAT97 matrix — a moderately sharpened transform that maps XYZ to a cone-like space with improved adaptation predictions compared to the pure HPE basis of CIECAM94.

Historical significance: CIECAM97s was the CIE's first standardised appearance model, bridging the gap between CIECAM94's research basis and the later CIECAM02 (CIE 159:2004) which replaced it with the more robust CAT02 matrix.

▶ CAT97 Matrix — Moderate Sharpening

CAT97 matrix: [[0.8562, 0.3372, -0.1934], [-0.8360, 1.8327, 0.0033], [0.0357, 0.0469, 0.9174]]. Row sums = 1.0000 (luminance-normalised).

Key difference from HPE: The third row is [0.0357, 0.0469, 0.9174] — not [0, 0, 1]. This means the S-channel response includes X and Y cross-talk, providing moderate sharpening without the extremes of Bradford or CAT02.

Conditioning: det ≈ 0.999, condition number ≈ 5.2. Well-conditioned and invertible for all standard illuminant pairs.

▶ CIECAM94 — Predecessor with HPE

CIECAM94 (Hunt, 1994) used the Hunt-Pointer-Estevez (HPE) cone primary matrix with the distinctive S=Z property (third row = [0, 0, 1]). CIECAM97s improved upon it by introducing CAT97's sharpened basis, better adaptation predictions, and refined appearance correlates.

▶ ICC Profile Connection Space (PCS)

ICC v4: The ICC mandates Bradford for converting between device colour spaces and the PCS (D50). CAT97 is for appearance model calculations, not ICC PCS conversion.

▶ CIEDE2000 — Colour Difference Standard

CIE 142-2001: CIEDE2000 (dE00) is the current recommended colour-difference formula with corrections for lightness, chroma, hue, and the rotation term RT.

Guideline ranges: dE00 <1 imperceptible, 1-2 perceptible by trained observers, 2-5 visible, 5-10 large, >10 very large.

▶ CIE Standard Illuminants
  • A (2856 K): Tungsten. Planckian radiator.
  • B (4874 K): Direct noon sunlight (deprecated).
  • C (6774 K): Average daylight (deprecated, superseded by D65).
  • D50 (5003 K): ICC PCS, printing.
  • D55 (5503 K): Mid-morning daylight.
  • D65 (6504 K): Standard daylight, sRGB.
  • D75 (7504 K): North sky daylight.
  • E: Equal-energy (theoretical).
  • FL series: Fluorescent lamp spectral types.
▶ Von Kries Coefficient Law

The Von Kries law models chromatic adaptation as independent gain control of each cone type. All CATs implement this in different spaces — HPE (physiological), CAT97 (moderate sharpening), Bradford (aggressive sharpening), CAT02/16 (optimised). The choice of space determines cross-channel coupling and prediction accuracy.

Mathematical Models and Formulas

CAT97 Transform Matrix (CIECAM97s sharpened):

M_CAT97 (XYZ to sharpened cone-like):
| 0.8562 0.3372 -0.1934 |
|-0.8360 1.8327 0.0033 |
| 0.0357 0.0469 0.9174 |

Key properties:
- Third row ≠ [0,0,1] — S-channel cross-talk
- Moderate sharpening (between HPE and Bradford)
- Row sums: 1.0000, 1.0000, 1.0000 (luminance norm)
- det ≈ 0.999, cond ≈ 5.2

Derivation: Optimised from CIECAM94 (Hunt & Luo, 1998)
to improve adaptation predictions while maintaining
physiological plausibility.

Used in: CIECAM97s (CIE TC 1-34, 1998)
Superseded by: CAT02 in CIE 159:2004
Research, Standards and Citations

CIECAM97s and CAT97 Primary References

[1] Luo, M.R., Hunt, R.W.G. (1998). The structure of the CIE 1997 colour appearance model (CIECAM97s). Color Res. App., 23(3), 138-146.

[2] CIE TC 1-34 (1998). The CIE 1997 Interim Colour Appearance Model (Simple Version), CIECAM97s. CIE Publication 131.

[3] Hunt, R.W.G. (1994). An improved predictor of colourfulness in a model of colour vision. Color Res. App., 19(1), 23-26.

[4] Hunt, R.W.G. & Pointer, M.R. (2011). Measuring Colour, 4th Ed. Wiley.

[5] Luo, M.R., Lo, M.-C., Kuo, W.-G. (1996). The LLAB(l:c) colour model. Color Res. App., 21(6), 412-429.

Cone Fundamentals and Physiological Models

[6] Estevez, O. (1979). On the Fundamental Data-Base of Normal and Dichromatic Colour Vision. PhD thesis, University of Amsterdam.

[7] Stiles, W.S. & Burch, J.M. (1959). NPL colour-matching investigation: final report (1958). Optica Acta, 6(1), 1-26.

[8] Stockman, A. & Sharpe, L.T. (2000). The spectral sensitivities of the middle- and long-wavelength-sensitive cones. Vision Research, 40(13), 1711-1737.

[9] CIE (2006). CIE 170-1:2006 Fundamental Chromaticity Diagram with Physiological Axes.

[10] Nayatani, Y. (1997). Simple estimation methods for the Helmholtz-Kohlrausch effect. Color Res. App., 22(6), 385-401.

Successor Models and Standards

[11] CIE (2004). CIECAM02. CIE Publication 159:2004.

[12] Li, C., et al. (2017). CAM16, CAT16. Color Res. App., 42(6), 703-718.

[13] Brill, M.H. & Susstrunk, S. (2008). Repairing gamut problems in CIECAM02. Color Res. App., 33(5), 424-426.

[14] Fairchild, M.D. (2013). Color Appearance Models, 3rd Ed. Wiley-Blackwell.

[15] CIE (2004). Colorimetry, 3rd Ed. CIE 15:2004.

[16] Sharma, G., et al. (2005). CIEDE2000 implementation notes. Color Res. App., 30(1), 21-30.

[17] Lam, K.M. (1985). Metamerism and colour constancy. PhD thesis, Bradford.

[18] ICC (2022). ICC.1:2022. International Color Consortium.

[19] IEC 61966-2-1:1999. sRGB.

About this tool

This tool implements 8 CATs with CAT97 (CIECAM97s) default, including CAT94 (HPE), full CIEDE2000/76/94, CIE 1931 chromaticity, SPD analysis, multi-method comparison — entirely client-side. Not a substitute for calibrated measurement.

Research Backend
Backend
CAT97 Sharpening Analysis — S-Channel Cross-Talk

CAT97 introduces moderate sharpening versus HPE's pure S=Z:

HPE (CAT94): S = 0·X + 0·Y + 1·Z = Z (pure)
CAT97: S = 0.0357X + 0.0469Y + 0.9174Z (sharpened)

For D65 white point (X=0.9505, Y=1.0, Z=1.0889):
HPE S = 1.0889 (= Z directly)
CAT97 S = 0.0357×0.9505 + 0.0469×1.0 + 0.9174×1.0889
= 0.03393 + 0.04690 + 0.99937 = 1.0802

Difference = 0.80% — small but meaningful for
high-accuracy chromatic adaptation predictions.

Sharpening spectrum (least to most):
HPE → Von Kries → CAT97 → CAT16 → CMCCAT2000 → CAT02 → Bradford → Sharp
CAT97 vs CAT02 — Why CIECAM97s Was Superseded
CAT97 S row: [0.0357, 0.0469, 0.9174]
CAT02 S row: [0.0030, 0.0136, 0.9834]

CAT97 cond ≈ 5.2 | CAT02 cond ≈ 4.6
CAT97 det ≈ 0.999 | CAT02 det ≈ 1.001

Key issues with CIECAM97s / CAT97:
1. Non-linear compression: CIECAM97s appearance
correlates had discontinuities near achromatic stimuli
2. Adaptation model: Less accurate for partial adaptation
3. Computational complexity: More steps than needed

CIECAM02 / CAT02 improvements:
+ Smoother appearance correlates
+ Better partial adaptation model
+ Simpler structure, fewer parameters
+ Sharper matrix → better large-shift predictions
Advanced Spectral Reconstruction

Spectral-domain adaptation comparison (coming soon).

Click Analyse to compute spectral-domain adaptation.
Batch Chromatic Adaptation

Enter hex colours (one per line or comma-separated). Adapts all using current settings with full dE analysis.

Click Run Batch to analyse colours.
Gamut Mapping Research
Gamut mapping is automatically shown in the per-patch table and JSON export (gamutClipped field).
Research backend: 8 CAT methods (CAT97/CIECAM97s default), CIEDE2000/CIE76/CIE94, CIE 1931 chromaticity, SPD analysis, multi-method comparison, matrix diagnostics, gamut clipping, batch adaptation, JSON/CSV export. All on-device, zero network.