Colour Difference

One measure of the difference in colour

between two stimuli is the Euclidean
distance E between the two points in the
three-dimensional space.
E = { (X)2 + (Y)2 + (Z)2 }1/2
The term E is derived from the German
word for sensation Empfindung. E
therefore literally means difference in
sensation.

Differences in rectangular coordinates L *, a* and b* in the CIELAB space

Tolerances in rectangular coordinates L*, a* and b* in the CIELAB space

Tolerances in cylindrical co-ordinates L *, C* and h in the CIELAB space

Some Anomalies
Major non-uniformity in colour space
was caused by an increase in the
numerical colour difference as the
chroma value of the standard is
increased.
Lightness tolerance was biased in
the tolerance to the full side of the
standard was about twice that to the
thin side of standard.

Towards Single Number

Shade Passing
Depending on the position of the
standard in colour space the
Lightness, Chroma and Hue
tolerances limits could be different.
Tolerance ellipsoid may be created,
where the semi-axes Lt units for
Lightness, Ct units for Chroma and Ht
units for Hue may be assessed and
used.

CMC (I:c)

hue

lightness

JPC79 Equation

It was found that the Chroma tolerance

increased rapidly as the Chroma of
standard increased; the Hue tolerance also
increase in a similar fashion to Chroma
tolerance, but Hue tolerance (Ht) was
about half the Chroma tolerance (Ct). The
lightness tolerance (Lt) increased only as
the Lightness of the standard increased.

Hunter L,a,b
L = 100 (Y/Yn)1/2
a = Ka [(X/Xn Y/Yn)/ (Y/Yn)1/2]
b = Kb [(Y/Yn-- Z/Zn)/ (Y/Yn)1/2]
X,Y,X and Xn, Yn, Zn are the tristimulus
values of sample and source.
Ka and Kb are the chromacity coefficients of
the source.

JPC79 Equation
E = [(L/Lt)2 + (C/Ct)2 + (H/Ht)2]1/2
Where Lt = 0.08195 L1/(1+0.01765 L1)

Ct = 0.0638 C1/(1+0.0131C1) + 0.638

Ht = TCt
T = 1 if C1 <0.638, otherwise
T = 0.36 + 0.4 cos (1 + 35)

Unless 1 is between 1640 and 3450 when

T = 0.56 + 0.2 cos (1 + 168)
L1, C1 and 1 refer to standard

Calculate Total Colour Difference and

difference in Chroma, hue between
A1 &A2 and between B1&B2
L*

a*

b*

A1

56.70 37.10 46.90

A2

57.25 36.50 47.40

B1

61.70 -37.23 -20.15

B2

61.10 -38.40 -21.95

C*

H*

L*

a*

b*

C*

h0

A1

56.70 37.10 46.90 59.8

A2

57.25 36.50 47.40

B1

61.70 -37.23 -20.15 42.33 208.40

51.60

(28.40+1800)

B2

61.10 -38.40 -21.95

L*

a*

b*

C*

H *

A1

56.70 37.10 46.90

A2

57.25 36.50 47.40 -0.02 0.77

B1

61.70 -37.23 -20.15

B2

61.10 -38.40 -21.95 -1.9

1.0

0.95

2.23

Limitations of JPC79 Equation

For dark colours L1 approaches zero and Lt also
approaches zero. L/Lt becomes very large. For L1
below 16 Lt = 1.022 solves the problem
Anomalies with near-neutral samples

STD 1

50

0.35

0.50

SAM1

50

-0.50

0.35

STD 2

50

0.367

0.525

STD

50

-0.525

0.367

E
1.3

3.7

Acceptability tolerances in CMC in the CIEIAB space

Group2
Sample

K/S

L*

a*

b*

c*

h*

E 76

E CMC
1:1

E CMC
2:1

2.2

51.61

16.41

1.05

16.44

3.68

1.8

54.22

15.50

2.24

15.66

8.23

3.01

2.73

1.80

2.0

52.79

15.64

1.19

15.68

4.37

1.42

1.20

0.76

2.2

52.09

16.25

1.29

16.30

4.55

0.56

0.51

0.34

2.0

52.99

15.81

2.49

16.00

8.94

2.08

1.95

1.62

1.9

53.47

15.77

0.95

15.80

3.44

1.97

1.74

0.95

2.4

20.71

16.45

0.99

16.48

3.44

0.90

0.82

0.41

1.6

56.66

15.07

1.70

15.17

6.45

5.26

4.70

2.55

2.4

50.69

16.59

0.83

16.61

2.86

0.96

0.87

0.49

10

2.8

48.47

16.60

0.79

16.62

2.73

3.15

2.85

1.45

Group2
Sample

K/S

L*

a*

b*

c*

h*

E 76

E CMC
1:1

E CMC
2:1

2.2

51.61

16.41

1.05

16.44

3.68

1.8

54.22

15.50

2.24

15.66

8.23

3.01

2.73

1.80

2.0

52.79

15.64

1.19

15.68

4.37

1.42

1.20

0.76

2.2

52.09

16.25

1.29

16.30

4.55

0.56

0.51

0.34

2.0

52.99

15.81

2.49

16.00

8.94

2.08

1.95

1.62

1.9

53.47

15.77

0.95

15.80

3.44

1.97

1.74

0.95

2.4

20.71

16.45

0.99

16.48

3.44

0.90

0.82

0.41

1.6

56.66

15.07

1.70

15.17

6.45

5.26

4.70

2.55

2.4

50.69

16.59

0.83

16.61

2.86

0.96

0.87

0.49

10

2.8

48.47

16.60

0.79

16.62

2.73

3.15

2.85

1.45

Group3
Sampl
e

E 76

E CMC
1:1

E CMC
2:1

K/S

L*

a*

b*

c*

h*

2.2

52.67

16.12

2.02

15.55

3.42

2.1

53.00

15.97

2.22

15.43

3.23

3.08

1.98

1.78

1.9

54.12

15.07

1.94

15.42

3.75

2.34

2.10

0.76

2.1

52.21

15.87

1.29

16.12

4.21

2.22

1.76

0.92

2.1

52.32

15.45

2.12

16.01

2.24

1.43

0.78

1.20

2.0

52.66

15.39

2.07

15.98

3.12

1.44

2.12

0.43

2.0

53.38

15.21

1.97

15.48

2.54

2.10

1.90

0.89

2.3

52.12

16.23

1.76

16.87

3.44

2.34

2.10

1.10

1.9

54.43

15.09

2.00

15.76

2.54

1.95

1.10

0.22

10

1.9

53.12

15.19

1.88

15.34

3.20

1.85

0.98

0.78

11

2.1

52.34

15.67

2.34

15.99

3.54

0.95

1.30

1.11

12

1.9

53.01

15.39

1.98

15.21

3.12

4.40

1.40

2.10

Group3
Sampl
e

E 76

E CMC
1:1

E CMC
2:1

K/S

L*

a*

b*

c*

h*

2.2

52.67

16.12

2.02

15.55

3.42

2.1

53.00

15.97

2.22

15.43

3.23

3.08

1.98

1.78

1.9

54.12

15.07

1.94

15.42

3.75

2.34

2.10

0.76

2.1

52.21

15.87

1.29

16.12

4.21

2.22

1.76

0.92

2.1

52.32

15.45

2.12

16.01

2.24

1.43

0.78

1.20

2.0

52.66

15.39

2.07

15.98

3.12

1.44

2.12

0.43

2.0

53.38

15.21

1.97

15.48

2.54

2.10

1.90

0.89

2.3

52.12

16.23

1.76

16.87

3.44

2.34

2.10

1.10

1.9

54.43

15.09

2.00

15.76

2.54

1.95

1.10

0.22

10

1.9

53.12

15.19

1.88

15.34

3.20

1.85

0.98

0.78

11

2.1

52.34

15.67

2.34

15.99

3.54

0.95

1.30

1.11

12

1.9

53.01

15.39

1.98

15.21

3.12

4.40

1.40

2.10

EXAMPLE
The CIE L*a*b* values of a standard
colour and a batch of material dyed to
match it are given in the table:
L*

a*

b*

Standard

47.22

-18.22

17.64

Batch

46.09

-17.88

16.83

Question
1. Calculate the colour-difference between the
standard and the batch using CMC(l:c)
equation, for both acceptability and
perceptibility conditions, given that the
tolarance ellipsoid semi-axis values of SL,
SC,SH are 1.064,1.852 and 1.400 respectively.
2. Comment on the two values of E obtained,
given that the tolarance limit of acceptability
is = 1.000 unit

Solution
Calculate values of C* and h0 and then the
difference L* , C* and H*
L* = -1.130, C* = - 0.85, H* = 0.351
Perceptibility (1:1) E = 1.83
Acceptability (2:1) E = 0.734

Samples
STD
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18

L*

a*

b*

C*

h0

31.79
33.24
30.96
31.85
30.90
32.79
39.09
31.89
32.43
31.02
32.92
32.23
32.06
32.85
32.08
32.28
33.12
31.45
32.07

43.83
45.53
44.03
44.92
41.96
44.32
44.21
44.82
45.31
44.10
45.87
44.79
45.61
45.55
45.07
44.52
46.06
45.19
44.82

24.8
26.89
25.92
26.02
24.07
25.51
24.52
26.11
26.70
25.21
27.15
26.01
26.82
26.33
27.07
26.32
27.10
26.08
26.02

50.36
52.87
51.09
51.91
48.38
51.13
50.55
51.87
52.59
50.80
53.30
51.80
52.91
52.61
52.57
51.72
53.44
52.17
51.82

29.51
30.57
30.49
30.09
29.84
29.93
29.02
30.22
30.51
29.76
30.62
30.14
30.45
30.03
30.99
30.59
30.47
29.99
30.13

E
E
(CIE76) (CMC1:1)
3.11
1.41
1.63
2.20
1.32
0.56
1.64
2.49
0.91
3.31
1.61
2.71
2.53
2.60
1.74
3.49
3.47
1.90

2.21
1.21
0.71
1.33
1.27
0.46
0.76
1.33
0.95
1.92
0.87
1.22
1.58
1.35
1.08
2.08
2.09
0.87

E
CMC2:1)
1.52
0.86
0.71
0.96
0.72
0.37
0.75
1.16
0.52
1.52
0.74
1.18
1.13
1.31
0.93
1.59
1.57
0.80

Samples
STD
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18

Visual

08.33
50.00
25.00
50.00
08.33
75.00
58.33
00.00
58.33
08.33
66.66
41.66
33.33
16.66
41.66
00.00
58.33
58.33

555

368
588
767
664
578
366
678
799
679
488
678

b*

26.89
25.92
26.02
24.07
25.51
24.52
26.11
26.70
25.21
27.15
26.01
26.82
26.33
27.07
26.32
27.10
26.08
26.02

c*

52.87
51.09
51.91
48.38
51.13
50.55
51.87
52.59
50.80
53.30
51.80
52.91
52.61
52.57
51.72
53.44
52.17
51.82

30.57
30.49
30.09
29.84
29.93
29.02
30.22
30.51
29.76
30.62
30.14
30.45
30.03
30.99
30.59
30.47
29.99
30.13

E
E
(CIE76) (CMC1:1)
3.11
1.41
1.63
2.20
1.32
0.56
1.64
2.49
0.91
3.31
1.61
2.71
2.53
2.60
1.74
3.49
3.47
1.90

2.21
1.21
0.71
1.33
1.27
0.46
0.76
1.33
0.95
1.92
0.87
1.22
1.58
1.35
1.08
2.08
2.09
0.87

E
CMC2:1)
1.52
0.86
0.71
0.96
0.72
0.37
0.75
1.16
0.52
1.52
0.74
1.18
1.13
1.31
0.93
1.59
1.57
0.80

The CIE 1994 Colour Difference Model

The CIE have defined reference conditions under
which the new metric with default parameters
can be expected to perform well.
1.The specimens are homogeneous in colour.
2. The colour difference (CIELAB) is <= 5 units.
3. They are placed in direct edge contact.
4. Each specimen subtends an angle of >4 degrees
to the assessor, whose colour vision is normal.
5. They are illuminated at 1000 lux, and viewed
against a background of uniform grey, with L*
of 50, under illumination simulating D65.

where C* = value of
standard's Chroma. If neither
can be reasonably termed the
standard, the geometric mean
of the C values is used:
C* = (C*1.C*2)1/2

The CIE 94 formula and acceptability

University of Bradford - BFD (l:c)

EBFD = [(LBFD/l)2 + (C*/cDc)2 + (H*/DH)2 +
RT(C*. H*/Dc DH)]1/2
The new formula is similar in structure to the
CMC(l:c). However, it was found that a new term
was required to take into account the fact that
when chromaticity ellipses calculated from
experimental data are experimental data are
plotted in the CIELAB a* b* diagram, they do not all
point towards the neutral point as assumed in the
CMC(l:c) formula

CIEDE 2000
E00 = [(L/klSl)2 + (C/ kcSc)2 + (H/ kHSH)2 +
RT(C/ kcSc .H/ kHSH)]1/2
L = L* , a = (1+G)a* , b = b*
C = (a2 + b2)1/2 , h = tan-1 (b/a)

CIELAB to CIEDE 2000

CIE2000 formula includes 5 corrections to
CIELAB:
1. A lightness weighting factor (SL)
2. A chroma weighting factor (SC)
3. A hue weighting factor (SH)
4. An interactive term (RT) between chroma
and hue differences for improving the
performance for blue colours, and
5. A factor (1 + G) for rescaling the CIELAB
a* scale for improving the performance
for gray colours