The measurement of chroma and its uncertainty

0 Introduction:
color difference is a commonly used detection item in paint detection, that is, the difference in color. In the national standard GB/T 3181 "paint film color standard", the color difference is defined as: quantitatively expressed color perception differences, expressed in e.
1 color and its
color is the perception of light emitted and reflected by an object through
the
. The three-factor hue (hue, also known as hue), brightness (value, also known as brightness) and saturation (chroma, also known as color) represent the characteristics of different colors, the degree of light and dark color, the degree of brightness of colors.
commonly used colors are HV/C color spaces based on Menzel colors, and three stimulus values XYZ and L*a*b?color spaces as defined by the International Lighting Commission (CIE).
HV/C color space is to mark hues, brightness, and saturation with numbers to indicate their color to varying degrees. CIE1964 XYZ Three Stimulus Value Color Space: specifies that color coordinates consist of three vectors perpendicular to each other, each of which can be represented by a point in the coordinates. CiE1976 L*a*b*color space is one of the most common color spaces for measuring colors, which overcomes the problem that the equal distance on the CIE1964 XYZ coordinates does not correspond to the observed equal color difference.
addition to the above color space, there are Hunter Lab color space, FMC II. color space, CIE1931XYZ color space, L.C.h color space, and so on.
2 chromic achance
because of the establishment of color space, we can use numbers to represent color and color difference. It is the most commonly used color difference calculation formula to calculate the total chromance between the two colors by using the geometric distance between the two positions in the CIE1976 L*a*b*color space, which can better match the visual match of the human eye. The calculation formula is as follows:
color detector is based on the principle of color space, instead of light sources and observers to measure the color difference of objects. The commonly used spectrographs fall into two categories: diffuse reflection and directional reflection. The most commonly used in the coatings industry are 45/0 (45-degree directional lighting, 0-degree direction observation) and d/8 (diffuse lighting, 8-degree directional observation), and the others are 0/45, 0/0, 0/d, 8/d, d/0 and other different conditions.
depending on the observer's perspective, the color meter can determine the chromosome in the case of 2 degrees field of view and 10 degrees field of view, such as CIE1931XYZ color space is suitable for 2 degrees field of view, and CIE1964 XYZ color space is suitable for 10 degrees field of view.
colorometers can simulate different light source conditions: commonly used with D65 light sources with a color temperature of 6,504 K for normal daylight (including ultraviolet wavelength zones), C light sources for normal daylight with a color temperature of 6,774 K (excluding ultraviolet wavelength zones), and A light sources with a color temperature of 2,856 K for incandescent light.
addition, there are F2 cold white light, F11 three narrow band cold white light and so on. People are influenced by some specular reflections when looking at colors, and color detectors use two measurement modes, Specular Company Excluded and Specular Component Index. SCE is to exclude the specular reflected light of an object from the measurement and measure only diffuse reflected light, which is comparable to what the human eye actually observes. SCI is a measure of an object's specular and diffuse light, which is closer to the object's true color than to the object's surface conditions.
3 Assessment of Uncertainty
3.1 Uncertainty Definition
Uncertainty Definition: the parameters that are indicatively and reasonably give dispersion to the measured values and are associated with the measurement results. It is a measure of the range of measured true values, giving the range in which the true value may be located based on a certain confidence probability. It is not a specific true error, but a quantitative representation of the part of the error range that cannot be corrected in the form of parameters. It comes from the imperfect correction of chance effect and system effect, and is a decentralized parameter for the measurement value given reasonably. Uncertainty is divided into categories A and B: Category A is an uncertainty assessment of statistical analysis of observed columns;
In practice, uncertainty may stem from incomplete or imperfect definitions of measurements, unsatisfactory methods for implementing measurement definitions, insufficient representation of sampling, inadequate understanding of the environmental impacts of the measurement process, or inadequate measurement and control of environmental conditions, and artificial readings of analog instruments Deviations, insufficient resolution or identification of the measuring instrument, the inability to give the standard and reference substance (standard substance) values for measurements, the non-compliance of constants and other parameters quoted from data calculations, the approximation and hypothesis of measurement methods and procedures, and changes in repeated observations that appear to be identical on the surface.
3.2 uncertainty and error
is the difference between the measured result and the measured true value. It is usually divided into two categories: system error and accidental error.
the difference between uncertainty and error:
first, the difference between the purpose of evaluation. Uncertainty indicates the dispersion of the measured value, and error indicates the degree to which the measurement deviates from the true value.
second, the difference in assessment results. Uncertainty is an unsigned argument;
third, the difference in factors. Uncertainty is obtained by people's analysis and evaluation, which is related to people's understanding of the measurement process.
fourth, the difference between the correction of the measurement results. Uncertainty does not mean a specific, exact error value, which can be estimated but cannot be used to correct a measure, and if the estimate of the error is known, the measurement results can be corrected.
3.3 Uncertainty Assessment
Extended Uncertainty: U-kuc (k=2, 3)
In order to facilitate user comparison of laboratory capabilities and levels, for general applications, the extended uncertainty response should be 95% confidence level, k=2.
4 Instrument Uncertainty Assessment
4.1 Instrument Parameter Setting and Sample Selection
Instrument: Color-Eye7000A Romometer, Gretag Macbeth, USA;
lighting conditions: d/8;
observer perspective: 10 degree field of view;
light source: D65;
calculation formula: CIE1976 L x a x b* ;
measurement mode: SCI and SCE mode, respectively;
sample selection: select the most commonly encountered color of the test - white palette.
4.2 Chroma measurement
Open the instrument and computer, open the measurement software, according to the operating procedures first with the black standard board and white standard board to correct the instrument, and then set the parameters of the instrument. Secure the sample to the measuring port and measure its chromation. In order to obtain the standard uncertainty weight of repetitive determination, 10 repeated measurements were made on the same sample, and the results were found in Table 1.
4.4 Uncertainty Report
Instrument determines the uncertainty of the white schrome color difference (E* ):
SCI mode, extended uncertainty U-0.044
(k=2);