Delta E
Navigation: Colour and Dispense > FAQ's
Delta E is the overall numerical colour distance between two samples, where a Delta E of 0 is a perfect match. Therefore the higher the Delta E, the further the two samples are from one another.

Colour Difference Measurement I - L*,a*,b*
Once the position of the colour in the CIELab Colour Space is established, one can describe the difference between colours in a numerical and objective way.

Consider the orange sample, 1. Let us assume the following CIELab coordinates:

Sample 1: L* = 50  a* = 20  b* = 60

Sample 2: L* = 50  a* = -10  b* = 50

 Colour Difference:

Difference in Lightness:
DL*  = (L*2 - L*1) = 50 - 50 = 0
Difference in a* value:
Da* = (a*2 - a*1) = -10 - 20 = -30
Difference in b* value:
Db* = (b*2 - b*1) = 50 - 60 = -10

In visual terms, we would say that the difference between these colours is that sample 2 is greener and less saturated than sample 1. 
In mathematical terms, the overall colour difference between the samples, DE,  can be expressed in the following way:
                                                 DE = √ (DL*)2 + (Da*)2 + (Db*)2

where DL*, Da* and Db* are the differences in lightness, a* value and b* value respectively.
So for the example above, the overall colour difference, DE, is:

                                                  02 + (-30)2 + (-10)2 = √ 1000 = 31.6

So, again using the example above, DL* is 0, Da* is negative, so we say the sample is greener than the standard, and Db* is negative, so we say that the sample is bluer than the standard. Note that the sign of the colour difference value gives us important information about the direction of the change, whilst the number itself gives the magnitude of the difference

Colour Difference Measurement II - L,C,h
Using a* and b* values enables us to calculate colour differences easily. We can use the LCh system to calculate colour difference in a similar way, however, due to the nature of the h value, we cannot simply subtract one h value from another to get a difference in hue. Rather we must define a new expression DH* from the equation for DE on the previous slide.
                                            DE = √ (DL*)2 + (Da*)2 + (Db*)2 

                                            DH* = √ (DE*)2 - (DL*2 + DC*2)

Consider sample 1. With a hue angle of around 45 degrees, the sample is an orange. Assuming a lightness value of around 50 and a chroma of 70, we have a saturated, mid-shade orange.
Now look at sample 2:
                                      L = 50
                                      C = 50
                                      h = 105 degrees, which makes it a green shade yellow.

In visual terms, we would say that the difference between these colours is that sample 2 is greener and less saturated than sample 1.

Using the equations shown above for colour difference, we can see that DL* = 0, i.e. no difference in lightness,
DC* = 50 - 70 = -20, i.e. a negative chroma value meaning that the sample is less saturated than the standard, and DH* has a numerical difference of 105 - 45 = 60; the direction of the change is determined by the direction in which colour is measured around the circle in an anti-clockwise direction from the red at 0. Since the sample is anti-clockwise from the standard, we say that the hue difference is POSITIVE, and the sample is greener than the standard.

The direction of hue differences is important and dependant on the colour of the standard. For example, a positive (anti-clockwise) difference on a red means it is yellower; the same positive difference on a blue would make it redder.
Conversely a negative (clockwise) hue difference on a green would make it yellower; on a red, a negative hue difference would mean that the sample is bluer than the standard.