A Method and Apparatus for Measuring Lustre
Field of the Invention
The present invention concerns a method of measuring the lustre of a surface having a metallic appearance and to apparatus suitable for performing the method. The method is of particular benefit in the surface coating industry e.g. for printing and packaging.
Background of the Invention
The use of metallic printing inks, particularly silver in the packaging industry, is becoming increasingly popular as a high impact visual effect. New types of metallic pigment are being developed in the inks and coating industries as an alternative to expensive metallic foil. As a result, the need to be able to measure the lustre or brilliance of a metallic finish in a reliable and consistent way has become essential.
An example in practical terms could be taken by the need of corporate packaging to maintain the graphic consistency of its brand when using metallic finishes to achieve equal lustre throughout a series of point-of-sale packaging despite the use of different application processes and substrates.
It became apparent during a pre-development investigation through producers of metallic pigment, universities and spectrophotometer manufacturers that there was no established method or equipment specifically designed to measure metallic lustre. In some industries, the use of a gloss meter has been adopted as an appropriate method of measurement. In this method the specular gloss is measured at a specidic angle. This does not represent the scattered reflection of the metallic pigment and therefore a very glossy surface with a low metallic effect would show an unrepresentative result in terms of lustre. In some cases panels of visual assessors are set up to judge metallic effect, but these are of course heavily subjective. The invention overcomes or at least mitigates the inaccuracies of the above known procedures by providing a method of measurement that will assess numerically the lustre of a surface having a metallic appearance in a representative, repeatable and reliable way, and provide the basis for the creation of an index that can be adopted by various markets to specify a level of metallic effect.
Brief summary of the Invention
In one aspect the invention provides a method of measuring the lustre of a surface having a metallic appearance which method comprises a) determining the intensity of light reflected from an area of the surface at various points across the visible spectrum; b) producing from the reflected light measurements a set of CIE L*a*b* coordinates for the surface with the specular content of the reflected light included and a set of CIE L*a*b* coordinates for the surface with the specular content of the reflected light excluded; c) converting each set of CIE L*a*b* coordinates into its tristimulus values; d) converting the tristimulus values for each set of CIE L*a*b* values into spectral reflectance data for each set of tristimulus values; and, e) converting the spectral reflectance data for each set of tristimulus values into a numerical value that represents the level of lustre for the surface which is a function of the difference between the sum of spectral reflectance data for the set wherein specular content is included and the sum of spectral reflectance data for the set wherein specular content is excluded.
In another aspect the invention provides apparatus for measuring the lustre of a surface having a metallic appearance which apparatus comprises a) means for determining the intensity of light reflected from an area of the surface at various points across the visible spectrum; b) means for producing from the reflected light measurements a set of CIE L*a*b* coordinates for the surface with the specular content of the reflected light included and a set of CIE L*a*b* coordinates for the surface with the specular content of the reflected light excluded; c) means for converting each set of CIE L*a*b* coordinates into its tristimulus values; d) means for converting the tristimulus values for each set of CIE L*a*b* values into spectral reflectance data for each set of tristimulus values; and, e) means for converting the spectral reflectance data for each set of tristimulus values into a numerical value that represents the level of lustre for the surface which is a function of the difference between the sum of spectral reflectance data for the set
wherein specular content is included and the sum of spectral reflectance data for the set wherein specular content is excluded.
In a further aspect the invention provides a computer programme product directly loadable into the internal memory of a digital computer comprising software code portions for performing steps c), d) and e) of the method of the invention.
In another aspect of the invention there is provided a computer programme product stored on a computer usable medium, comprising computer readable programme means for performing steps c), d) and e) of the method of the invention.
Brief Description of the Drawings
Figure 1 is a schematic representation of a measuring guide suitable for use in the method of the invention.
Figure 2 is a chart showing the results of using the measurement method of the invention as a way of monitoring the lustre of point-of-sale products.
Detailed Description of the Invention
Surfaces that display a metallic appearance are usually metal, finished or polished in a particular way, or metallic flakes processed from metal and dispersed in a medium that enables the resulting composition to be applied to whichever product or substrate is required to attain the desired metallic finish.
In the case of cast or formed metal surfaces, it is the process of polishing or surface treatment that determines the level of reflection, hi basic terms, the smoother the surface of the metal the greater the reflectance and increase in the specular content of light reflected, ultimately resulting in a mirror finish. Conversely, the more uneven the surface of the metal is, the more the reflected light is scattered away from the specular angle resulting in a lower level of lustre or surface reflection.
Broadly, the same principles apply where metal is atomised and processed into flakes that are used in the manufacture of paints, inks and coatings. In basic terms, the dispersed flakes of metal are required to be applied and to lie on the substrate in an even film and form as near as possible, a continuous film of dispersed metallic flakes mimicking a solid metal effect. The film thickness of such a coating, depending on which application process is used, will vary from 3 to 50 microns. Consequently, the
size of the metallic flake is crucial to its particular application process and the associated film thickness. The size and shape of the metallic flake is an important factor in the resulting lustre of the coating together with the way that the flake orientates itself within the film. The method of the invention takes these factors into account when producing a numerical measurement of the lustre of the surface.
The first step of the method of the invention is determining the intensity of light reflected from an area of the surface at various points across the visible spectrum, preferably in the range 400nm to 700nm.
Advantageously, the reflected light measurements are made using a sphere spectrophotometer. The spectrophotometer may view the sample diffusely by the use of a barium coated integrated sphere. It is a feature of the sphere spectrophotometer that the light from its light source reflects from the metallic surface at all angles befoτe being passed to the viewing optics. Advantageously, the spectrophotometer comprises a specular exclusion port or gloss trap. This feature allows the spectrophotometer to obtain data from the sample with the specular gloss removed from the measurement and also with the specular gloss included. In this way, the two separate sets of CIE L*a*b* coordinates required by the method of the invention can be produced, namely, one set for which specular content is included and another set for which specular content is excluded. The spectrophotometer reads an area of the sample through an aperture having a particular size. Some spectrophotometers have a variable aperture size. The size of the aperture used will affect the measurement taken. Therefore, for accurate comparisons to be made between the lustre of different samples, it is important that the same size aperture is used when taking readings. Preferably, the spectrophotometer obtains raw spectral data from the surface of the sample and delivers colorimetric data in the form of CIE L*a*b* colour space using the CIE Standard Observer, CIE Daylight Illuminant.
For convenience, a portable sphere spectrophotometer may be used. An example of a suitable commercially available instrument is the X-Rite™ Model SP64. The instrument has d/8° measuring geometry, a viewing aperture of 8mm and illumination of 12mm under a gas filled tungsten light source. The instrument has a spectral range from 400 to 700nm measuring at lOnm intervals.
In measuring metallic surfaces, the surface profile is contributory to the result as is the metallic flake orientation. Irregularities in the surface will affect the reflection of light and to maximise the consistency of reading it is advantageous to measure the surface at different angles, preferably set angles, in the same area of the sample. The resulting measurements can be averaged.
It has been found that taking measurements of the area of the surface through 144° at intervals of 36°, i.e. at 0°, 36°, 72°, 108° and 144° provides excellent results.
In practice, this can be achieved by positioning the spectrophotometer aperture over the area of the surface to be measured, taking the measurements, rotating the spectrophotometer by 36° over the same area, taking further measurements, and repeating these steps until the required five sets of measurements have been taken. The data from the measurents are averaged and processed within the spectrophotometer.
To maintain consistency while measuring samples, it is preferred to use preset viewing angles. A measuring guide as illustrated in Figure 1 may be used to standardise the method. The measuring guide 10 comprises a sheet of material 11 e.g. plastics material having a circular aperture 12 large enough to accommodate the spectrophotometer aperture, typically having a diameter of 20mm. The circumference of the aperture 12 is marked at the various angles by lines 13 to enable the spectrophotometer to be orientated in the correct way.
In use, the measuring guide is positioned on the sample to be measured. The spectrophotometer is positioned over the measuring guide so that it can measure the reflectance from the area exposed by the aperture 12 and is aligned with one of the lines 13 on the measuring guide. After measurements have been taken, the spectrophotometer is rotated so that it is aligned with another of the lines 13. Further measurements are taken and the process repeated until five sets of measurements at intervals of 36° have been taken.
When a metallic coated surface is on a transparent or translucent substrate, the surface on which the sample is laid to be measured can affect the lustre measurement. In this case, it is advantageous for the sample to be laid on a surface that gives enough reflectance and opacity for the spectrophotometer to obtain accurate colorimetric data. For example, a bright white card having a thickness of 400μm and having CIE
L*a*b* coordinates of L*94.50 a*0.70 b*1.50 is suitable. Employing such means provides consistency of measurement leading to repeatable results.
In the method of the invention, the reflectance measurements recorded by the spectrophotometer may be converted into the desired lustre readings by any suitable means e.g. by manual computation or by the use of appropriate software.
Each of the two sets of CIE L*a*b* coordinates produced by the spectrophotometer is converted into its tristimulus values. Tristimulus values are mathematically derived from the colour measurements and represent the amounts of red, blue and green light recorded by the eye of a theoretical Standard Observer. In order to generate the tristimulus values, both sets of CIE L*a*b* readings obtained from the spectrophotometer may be incorporated into the transformed CIE 1976 CIELAB colour space equation referenced in CIE Publication 15.2 (1986) Colorhnetry (second edition) Section 4.2.2. the transformed equation is shown below
Y = Y0 [ (L* + 16)/116 ]3
X = X0 [ a*/500 + (Y/Yo)1/3 ]3 Z = Z0 [-b*/200 + (Y/Yo)I/3]3
wherein X0 = 94.811, Y0 = 100.00 and Z0 - 107.304. The conversion of tristimulus values for each set of CIE L*a*b* coordinates into the respective spectral reflectance data may be achieved by the use of linear models and basis functions as referenced in Computational Colour Science Using MATLAB by Stephen Westland and Caterina Ripamonti, Section 10.5.3.
The computation of the tristimulus values T for a given reflectance spectrum P can be represented by the linear system
T = MP wherein M is a 3 x 31 matrix whose rows contain the wavelength-by-wavelength product of the illuminant with one of the three colour-matching functions. By using basis functions representing the smoothness of the reflectance spectra a more accurate prediction of T can be obtained. For example, if a linear model with three basis functions is used, then the 31 x n matrix of reflectance spectra P can be replaced by BA where B is a 31 x 3 matrix of basis functions and A is a 3 x n matrix of coefficients, producing the equation T = MBA
MB is a 3 x 3 matrix and therefore the above equation represents a linear system with three constraints and three unknowns and can easily be solved using the equation
A = (MB)-1T Since P = BA, an equation can be written to recover the spectral reflectance data as follows
P = B(MBy1T
The spectral reflectance data for each set of tristimulus values is converted into a lustre measurement for the surface which is a function of the difference between the sum of spectral reflectance data for the set wherein specular content is included and the sum of spectral reflectance data for the set wherein specular content is excluded.
Preferably, the lustre measurement for the surface is the difference between the sum of spectral reflectance data for the set wherein specular content is included and the sum of spectral reflectance data for the set wherein specular content is excluded divided by the number of wavelengths at which the intensity of the reflected light is determined.
More preferably, the spectral reflectance data for each set of tristimulus values is converted into a lustre measurement for the surface in accordance with the equation (A-B)/31 wherein the intensity of the wavelengths of light reflected from an area of the surface is measured at IOnm intervals in the range from 400nm to 700nm and A is the sum of spectral reflectance data for the set wherein specular content is included and B is the sum of spectral reflectance data for the set wherein specular content is excluded. The apparatus required for measuring the lustre of a surface having a metallic appearance comprises a) means for determining the intensity of light reflected from an area of the surface at various points across the visible spectrum; b) means for producing from the reflected light measurements a set of CIE L*a*b* coordinates for the surface with the specular content of the reflected light included and a set of CIE L*a*b* coordinates for the surface with the specular content of the reflected light excluded; c) means for converting each set of CIE L*a*b* coordinates into its tristimulus values;
d) means for converting the tristimulus values for each set of CEE L*a*b* values into spectral reflectance data for each set of tristimulus values; and, e) means for converting the spectral reflectance data for each set of tristimulus values into a numerical value that represents the level of lustre for the surface which is a function of the difference between the sum of spectral reflectance data for the set wherein specular content is included and the sum of spectral reflectance data for the set wherein specular content is excluded.
As discussed above, the means for determining the intensity of light reflected from an area of the surface at various points across the visible spectrum is preferably a sphere spectrophotometer.
Preferably, means b) to e) comprise one or more computer systems for performing the required steps.
Typically, commercially available sphere spectrophotometers having a specular exclusion port include means for producing from the reflected light measurements a set of CIE L*a*b* coordinates for the surface with the specular content of the reflected light included and a set of CIE L*a*b* coordinates for the surface with the specular content of the reflected light excluded by way of an appropriate computer programme comprising software code portions for performing the required steps stored in the computing means in the spectrophotometer. The remaining means c), d) and e) may also be provided in the form of a computer programme comprising software code portions for performing the required steps. Clearly, the computer programme present in the spectrophotometer could include the software code portions representing means b), c), d) and e). Alternatively, a computer programme comprising software code portions for performing the steps c), d) and e) could be run on a separate computer.
A lustre index can be created that operates on a linear scale with 100 as the highest value representing a surface equivalent to a perfect mirror and 0 as the lowest value representing a surface having no metallic appearance or specular gloss.
The method of the invention is suitable for establishing comparative lustre levels on surfaces with a similar profile. Arbitrary standards can be set although it must be recognised that different metallic materials can have different surface profiles and consequently this can affect the visual perception of lustre.
The use of the CIE L*a*b* coordinates will give an evaluation of colour difference when comparing a range of coloured metallic samples. By combining
lustre measurement according to the invention with CIE L*a*b* coordinates, users are able to effectively and accurately evaluate and control coloured metallic surfaces.
The use of a lustre index in the active working situation of monitoring point- of-sale products of a major retail chain is shown by Figure 2. This demonstrates the ability of the measurement method of the invention system to record the level of silver lustre on a series of packaged goods achieved by several different print producers using various processes and printing substrates.
The tolerance levels have been set at an upper limit of 11.5 and a lower limit of 7.5. This allows the variability in lustre that can be caused by the substrate and by the printing process to be monitored, and, in cases where the lustre falls out of the tolerance band, advice given to the printer as to how to adjust his process to comply with the set tolerance.
The graph shows the lustre results of random samples purchased from the store at the time of the pre-monitoring situation. The wide differences shown are those that can occur due to a number factors such as: different silver inks, low ink film weight and excess over-varnishing. These were accepted by the retail group as uncontrollable factors prior to the use of a lustre index.
The graph also shows the lustre results of samples produced after the quality control system described above had been applied. With this particular monitoring programme, a visual assessment comparison of the samples was carried out by experienced personnel to ensure that the lustre numbers complied with the visual perception which in all cases agreed.