Optimal inter-instrument agreement and communication of color values

To communicate color values effectively, completely specify the 6 parameters of your measurement method and verify that your color measurement is repeatable. These are the key factors to ensure inter-instrument agreement.

The best inter-instrument agreement in color measurement at multiple locations occurs when a product is measured with the same instrument model from the same manufacturer combined with the same measurement method at all sites.

80.86, 1.03, 58.81 ?

In a physical measurement of weight or volume, there is only one parameter that can affect the measured values – the expression of units in American Standard or metric.

In a psychophysical measurement that quantifies color as a person perceives it, a complete quantification is 3 numbers with 6 parameters. Changing any of these measurement parameters will affect the expression of your measured color values.To effectively communicate your color values to a customer or another site, the following 6 parameters must be provided along with your values to ensure effective color communication and inter-instrument agreement.

  1. Color Scale: Is it Hunter L, a, b or CIE L*, a*, b*? Both are L, a, b – type color scales in use by industry but color values will be different depending on the color scale used. Are these numbers expressed in absolute values, or differences (delta L*, delta a*, delta b*, delta E*) relative to a product standard? Sometimes, only a single specialized color index is being reported such as Whiteness Index, Yellowness Index or correlated visual color scales like APHA. Color scales are like color languages; you must be clear in what scale your values are expressed in in order to communicate effectively. Our recommendation is to use CIE L*, a*, b* unless a customer, test method or industry specification requires another.
  2. CIE Illuminant: There are a choice of several – A (representing home light), C or D65 (representing daylight) and F2 (representing office light) are the most common illuminant choices. Illuminant C (6770 degrees Kelvin) representing average daylight is near but not identical to Illuminant D65 (6500 degrees Kelvin) representing noon daylight. Our recommendation is to use illuminant D65 unless a customer, test method or industry specification requires another.
  3. CIE Standard Observer: You have choices of the 1931 2 Degree or the 1964 10 Degree Standard Observer. Both are very similar but not the identical. The 1964 10 Degree Observer is recommended.
  4. Instrument Geometry: This parameter is fixed by the choice of instrument purchased and the nature of the sample. There are two general categories of instrument geometries for color measurement – directional 45°/0° or 0°/45° and diffuse 8°/d or d/8° sphere geometry. Instruments of different geometries view the sample differently, which affects the measured values. It is also best to further identify further conditions of the instrument mode or optical path – specular inclusion (RSIN)/exclusion (RSEX) and UV filter position.
  5. Sample Preparation: How is the sample prepared prior to measurement? Is the sample backed by another material? Is it measured in a container such as a transmission cell or sample cup? Is it a powder pressed into a plaque? Is the sample cleaned in a defined way prior to measurement? The key purpose of sample preparation is to make the sample more uniform to help ensure a repeatable measurement. As preparation affects the nature of the sample presented for measurement, the sample must be prepared the same way each time.
  6. Sample Presentation: What is the sample area being measured and number of readings averaged for each measurement? Is there a specific measurement pattern for the readings? Presentation techniques use optical (largest area of sample view) and statistical (multiple readings with replacement per measurement) to ensure a repeatable and representative color measurement of the sample. This must be the same at all sites to attain good inter-instrument agreement.

These 6 parameters defining color measurement are supported in ASTM, ISO and JIS industrial test methods.

Along with specifying your method of 6 parameters, it is recommended that you provide an estimate of the repeatability for this method. Typically color values are reported to two decimal places but given the non-uniform characteristics of the sample and replacement of the sample between readings, measurement repeatability for an application is typically in tenths of a unit.

Using the method above, take several readings of the same sample (10 is good; 30 is better) and calculate the standard deviation for your color values from sampling. Round the standard deviation up to a single decimal place and multiply x 2 for an estimate of measurement repeatability to a 95% confidence level. This is also an estimate of the MDL or mean detection limit of your measurement method, as you can never detect a smaller color difference than the repeatability of your measurement method.

At HunterLab we use a little Excel spreadsheet to work this out. Contact us if you would like a copy.

FAQ: “Communication of color measurement seems complicated. Is there a simpler way?”

In color measurement you are quantifying a perception which is a little more complicated than weights or volumes. If you want effective color communication and inter-instrument agreement, it is necessary to specify your color values plus the 6 parameters that define them. A measurement repeatability statement is a valued extra.

Color communication of “lemon yellow” measurement.

For this lemon color, my color values are CIE L* = 80.86 a* = 1.03, b* = 58.81 D65/10°; directional 45°/0° geometry; gently wash the lemon skin before measurement; measure a 25 mm sample area; and average 3 readings with rotation around the center of the lemon. My measurement repeatability is +/- 0.2 in L*; +/- 0.1 in a*; +/- 0.1 in b* to a 95% confidence level.  These 6 elements of colorimetric communication are also found in most industrial color measurement methods.

For more information, see the HunterLab Application Note – AN 1031 Identification of Measurement Parameters.

FAQ: “Do you have information of LabScan XE versus ColorFlex EZ? What is the different or vary of measure one sample? If LSXE use 1” Aperture same as CFEZ will it have same measure color result?”

Critical to a discussion is the type of sample but I can speak to the general rules of inter-instrument agreement.

Agreement in absolute color values between CFEZ 25 mm and LSXE 25 mm area of sample view will be excellent on opaque, solid, uniform color tiles, and even better on color differences between a lot sample and target color.

Inter-instrument agreement should also be very good on less uniform product color if the 6 parameters that define the color measurement method match.

Samples that have optically active characteristics such as translucency, light trapping, UV brightening, gonioapparent color, prismatic, photochromism, thermochromism, or hygrochromism, represent more challenging cases for agreement. In these cases, the best option is to use the same model of instrument from the same manufacturer in combination with the same measurement method.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s