CA1059338A

CA1059338A - Textile color analyzer calibration

Info

Publication number: CA1059338A
Application number: CA263,204A
Authority: CA
Inventors: Paul A. Schumann (Jr.); Alvin H. Tong
Original assignee: International Business Machines Corp
Current assignee: International Business Machines Corp
Priority date: 1975-10-10
Filing date: 1976-10-08
Publication date: 1979-07-31
Also published as: IT1077026B; JPS5634050B2; JPS5247788A; GB1530947A; DE2638398A1; US4029419A; FR2327524A1; FR2327524B1

Abstract

TEXTILE COLOR ANALYZER CALIBRATION
Abstract of the Disclosure A spectrophotometer is operative to measure or analyze the colors of materials such as textiles. A test sample is held by a holder and is illuminated by polychromatic light. Diffuse light reflected from the sample is collected and converted to variable monochromatic light and detected. The system is calibrated by using a secondary white standard calibrated against a primary white standard, by using a black standard to compensate for dark current and internal reflectance from a pressure plate, and by using a mirror to obtain internal wall re-flectance used in a correction factor.

Description

Field of the Invention ~.
This invention relates to spectrophotometry and~ more particularly, to the calibration of a spectrophotometer system particularly adapted for analyzing the colors of materials such as textiles.

DLM/Tl ~.

5~3 ~3~3 Prior Art In spectrophotometry, it has been customary to use white and -black standards to establish 100% and 0% reflectance values. After being thus calibrated, re-flectance of a test sample is measured and determined or calculated relative to such values.
Also known in the prior art are spectrophotometers that include or are connectable to a computer to provide a high speed system for automatically measuring the sample, storing calibration factors, and calculating the reflectance o-f the sample. An example of such a -system is disclosed in U.S. Patent 3,75l~643 which discloses a spectro-photometer primarily intended for measuring the thickness of thin film on silicon substrate by measuring reflectance. The present invention includes a similar system. ~ , Summary of the Invention One of the objects of the invention is to provide a spectrophoto-meter for analyzing colors that is fast, automatic and accurate.
Another object is to provide a spectrophotometer having apparatus of the type disclosed in the above related applications, with means to calibrate the system to provide measurement results that are highly accurate and reproducible. '~
A further ob~ect is to provide a novel method for calibrating an instrument for measuring or analyzing the color of a textile sample. ;~

PO~-75-028 - 2 -~i 33~3 :
:

1 Still another object is to calibrate a color analyzer

2 to compensate for internal reflections in an instrument head.

3 Another object is to calibrate a spectrophotometer

4 initially on a primary standard and to thereafter calibrate the apparatus using a secondary standard.
6 A still further object is to provide a textile color 7 analyzer having a measuring head using a glass plate for 8 compressing a textile sample, wherein calibration measurements 9 are made to compensate for the reflectance of the glass plate.
Another object is to provide a spectrophotometer which 11 calculates the reflectance of a test sample allowing com-12 pensation to be made for specular reflections from the sample 13 and from a standard.
14 Brie~ly, the invention comprises a spectrophotometer havin~ a measuring head adapted to press a test sample between 16 a glass plate and back-up pad. The sample is illuminated and 17 diffuse reflections are collected, the light being of variable 18 wavelength across the ranye of wavelengths of visible 19 light. Calibration measurments are made on a black standard to compensate for dark current and reflectance of the glass 21 plate, on a primary white standard to determine a 100~
22 value, on a secondary white plaque to determine a scaling 23 factor relative to the 100~ line, and on a test sample the 24 color of which is to be analyzed. A mirror is used to de-termine the internal wall reflectance. The reflectance of the 26 test sample is calculated using the above factors.

~t)S9338 1 Other ob,iects and advantages of the invention will be apparent from the following detailed description of a pre~erred embodiment of the învention taken in connection with the accompanying drawing where-in:
Fig. 1 is a schematic diagram of a spectrophotometer embodying the invention, Fig. 2 is an illustrative graph showing the output of the detec~
tor shown in Fig. 1, for one revolution of the monochromator wheel while making a measurement;
Fig. 3 is an illustrative graph showing the various signal ratios obtained during operation of the spectrophotometer; and Fig. 4 is an illustrative graph of the reflectance of a test sample at different wavelengths.
Detailed Description Referring now to the drawing, and first to Fig. 1~ the preferred embodiment of the invention there shown comprises a spectrophotometer 10 including a transducing or instrument section 11 and a computer or data processing section 12. The instrument section comprises a head 13; the remaining portions of the instrument section are similar to those disclosed in the above mentioned patent.
A polychromatic light source 15 emits light towards a reflector 16 and lens 17. The reflector directs the light received thereby towards the lens and the lens concentrates the light received directly from source 15 and from reflector 16 into the entrance of a fiber optic bundle or light pipe 18.

;
PO9-75-a28 4 593;~

, .
1 The exit end of light pipe 18 is bifurcated to form part of 2 a reference path and a sample path, the reference path including 3 a light pipe 19 whose exit end is fanned out into a narrow 4 rectangular face aligned with a similar face of a fiber optic bundle 20 disposed on the other side of a monochromator 6 wheel 21. Wheel 21 includes an opaque portion 22 and a 7 wedge type variable interference filter 23 that is ring 8 shaped and extends for 180 around and concentric to the 9 axis of wheel 21. A~ wheel 21 is rotated, light from bundle 19 will be transmitted to bundle 20 during one half 11 of the revolution when filter 23 is located therebetween.
12 During the other half, opaque portion 22 blocks the trans-13 mission of light. The exit end of bundle 20 is connected 14 to a detector such as a photo multiplier tube (PMT) 24.
The sample path includes a light pipe 26 having an 16 exit end that terminates in head 13 and functions to 17 transmit light therealong so as to illuminate a test sample 18 28 pressed between a glass plate 27 and a polyethylene back-19 up pad 30. This pad is movable between the dotted position and full position so as to allow the textile sample 28 to 21 be inserted and removed so that other samples or standards 22 may be placed therebetween. Light reflected from the sample 23 is collected by four ends 32 of a fiber optic bundle 33 that ;
24 transmits such light therealong. The end of bundle 33 is 25 fanned out similar to the manner described above with `
26 reference to the end of bundle 19 and is in alignment with a 27 similarly shaped end of a bundle 34 disposed on the other side pog-75-028 -5-:

,- ~

1~59338 .
1 of wheel 21 and connected to PMT 24. The ends of bundles 33 and 34 are diametrically opposite to the ends of bundles 19 and 20 so that light is alternately transmitted along the reference path and the sample path. It is to be appreciated that Fig. 1 is merely a schematic diagram.
Filter wheel or monochromator wheel 21 is rotated by a motor 36 in a predetermined direction and a conventional shaft encoder 38 is connected to rotate with the motor and provide output signals at fixed angular displacements of the wheel. Filter 23 is uniformly graduated from one end to the other to transmit a narrow band of mono-chromatic light throughout the range of visible light and because of the uniform gradation, the light being transmitted at any position of wheel 21 is a function of the angular displacement which can be ob-tained by signals from the shaft encoder. PMT 24 provides an output signal proportional to the intensity of light received thereby, and an illustrative diagram oF the output signal thereof is shown in Fig. 2 for one revolution of monochromator wheel 21. In Fig. 2, it is as-sumed that the zero degree position occurs at the blue end of filter 23 and that that light is first transmitted along the sample path or test path. As wheel 21 rotates through the first 180 of revolution, the output of PMT 24 relative to the angular displacement may vary as shown in the drawing. During the second 180 of revolution, light is transmitted along the reference path and the output of PMT 24 is proportional to the intensity of the light received thereby. This system pO9-75-~28 - 6 -~LM/T4 , ;, , i ~ "

~ S~3 3~3 `~
l is initially set up so that an equal amount of light passes along both the sample and reference paths. The ratio of the signal St at a given wavelength ~ n to the reference signal ~r corresponding to wavelength ~ n provides a signal ratio that is a measure of the reflectance of the sample. That is, it is approximately the ratio of the light reflected by the sample to the incident light received by the sample. - . ..
It should be obvious that the output of PMT 24 is an analog signal that can be used in an analog system to drive a plotter or recorder to provide graphs from which measurements could be taken manually and the final reflectance calculations done manually. How- ;
ever, in the preferred embodiment of the invention, PMT 24 is con-nected to a conventional analog-to-digital converter (A/D) 40 which in conjunction with signals from shaft encoder 38 provide a series .
of discrete digital values corresponding to different angular posi-tions or wavelengths. Converter 40 may be connected to the I/0 bus 42 of a computing system including a CPU 44 and a memory 46 where the data is transferred under the control of CPU 44 and stored :
in memory 46. A keyboard printer 48 is preferably connected to the system and acts as a terminal allowing an operator to control operation of the system and obtain a printed output, and a display S0 may be connected to the system to allow an operator to view a d1splay of the results of a measurement. Quite obviously, such a system runs under the control of programs stored in memory 46 and the programs may be conventional or standard in nature. Relative to the computer, the instrument section appears as another I/0 P09-75-028 - 7 _

5~

r~ .lLI( ~59 3 3~ ;

1 device.
In the operation of the system, the output oAf the PMT as shown in Fig. 2 would be converted into a series of discrete digital values each of which would b~ associated with a different nominal wavelength.
The computer would then compute the signal ratios of St/Sr and a plot of such ratios could be then calculated or made where a typical one would look such as is shown for line T (test) in Fig. 3. The number of discrete points to be calculated is somewhat a function of the sig~
nals provided by the shaft encoder and the resolution desired. Com-mercially available shaft encoders provide signals every 1 of rota-tion so that 180 separate discrete points could be obtained. However, we found that 16 such discrete points are sufficient to provide readings every 20 nanometers over the range of ~00 to 700 nanometers.
Thus, the graphs shown represent making determinations at such dis-crete points and then drawing a curve through the points.
In accordance w;th the calibration technique of the invention, five different measurements are made for a primary standard, a secon-dary standard, a black standard, a mirror and a test sample and, during each of the measurements, the signal ratios P, S, B, M, and T
respectively are determined. Fig. 3 shows an illustrative chart in which these various signal ratios have been plotted it being recalled that the signal ratio at any given wavelength ~n is a ratio St/Sr for the ltem being measured. When such measurements are made and the signals stored in the memory, calculations can ~0593313 1 be performed to determine the true or corrected reflectance R
2 of the test sample, an illustrative graph of a typical example 3 being shown in Fig. 4. The significance of these measure-4 ments and the calculations will now be discussed.
The primary standard signal ratio P for establishing the

6 100% reflectance line is obtained by measuring a standard

7 having a coating of barium sulfate. As is known, it is neces-

8 sary to prepare such standards using barium sulfate powder and g that once prepared, such standards are sub~ect to deteriora-tion so that it becomes necessary to prepare new ones from time 11 to time. To avoid such preparation, a secondary standard is 12 used which is a white ceramic plaque that doe not deteriorate 13 with time and usage. Thus, the primary standard signal ratio 14 P is determined one time only, the secondary standard signal ratio S is determined at the same time to provide a 16 scaling factor that compensates for any later variations ln 17 the system, and the secondary standard can be used at later 18 time to periodically calihrate the system. ~ `
19 A black ceramic plaque is measured to provide a black siynal ratio B that represents the zero percent (0%) 21 reflectance line. Such measurement compensates for the dark 22 current of the detector, PMT 24, and for the reflectance of 23 plate 27 which is coated with an anti-reflective coating but 24 nevertheless exhibits a low level of reflectance.
A surface coated mirror is measured to provide a signal 26 xatio M that provides a measure of the internal wall reflec-27 tance of head 13.
28 When such signal ratios have been determined, the 29 calculations below can be made. The calculations are '' ' , , 1 made at each of the discrete wavelengths, eg, every twenty nanometers.
The following symbols are used:
P = primary standard signal ratio at a given wavelength ~n S = secondary standard signal ratio at a given wavelength ~ n M = mirror standard signal ratio at a given wavelength ~ n B = black standard signal ratio at a given wavelength ~ n T = test sample standard signal ratio at a given wavelength ~ n Rs = uncorrected reflectance of secondary standard Rp = reflectance of primary standard Rt = uncorrected reflectance of test sample Rt = TS B x RRp (1) Rs = F~ (2) By definition, Rp = 1 so that by making the measurements above, the uncorrected reflectance Rt of the test sample can be calculated.
This uncorrected value compensates for the dark current and plate re-fiectance and determines the reflectance of the test sample between the 0 and 100% lines of the black and white standards. The ratio Rs/Rp (Equation 1) is the scaling factor which, with Rp = 1, reduces to Rs representing the scaling factor.
It can be shown that the corrected or true reflectance R ~n of a test sample is given by the formula: ~ i R ~n = {E2 ~, ~ As Rp - Aw Rw) ~ [ 2 As B

P09-75-02~ - 10 -~ ~ '.
. . ., . : ': , ~: . : ..

35~ 8 r r, ~ As Rp - Aw Rw 1 l 1 - ¦1 4~ ~s Rt (1 - Aw Rw) Ll -~ As Rp +r - Aw) Rw~ 1l ~ l [~ As Rt ~ ~1 - r As Rp - Aw Rw ~h + (~ - Aw) Rw3~1 5 ~1 - r As Rp ~ (r - Aw) Rw ~ ~
Rw = reflectance of wall (3) Rw = 1 - As Rl-T + Aw - 1 (4) where As = fractional area of sample Aw = fractional area of internal wall of head and As ~ Aw = 1 The fractional areas are dependent on the geometry of the instru-ment and have to be determined beforehand and the values stored.
R~T - uncorrected reflectance of mirror ;
RMT = MS B x RRp Equations (4) and (5) define terms in (3) and the only terms un-deflned so far~re ~ and r .
~ is the specular multiplier for a test sample. If the specular component of reflections from a sample is known, the~ factor allows it to be compresentated for. For a specular component, eg, of .2, ~ = 1.2.
If ~ is not known, assume the specular component is low or negligible and ~ = 1 and thus may be dropped from equation (3). This assumption is valid because the instrument head is designed to minimi~e the specu-lar component.
r iS the specular multiplier for the primary standard and allows the specular reflectance of such standard to be compensated for. r iS
assumed to be 1 for the same reasons as ~ .
In summary, the corrected reflectance Rln represents the true reflectance of the test sample and it includes P0~-75-028 1 DLMlT5 . -~ .' . .

1 factors that account for the internal wall reflectance, the 2 head geometry, and the specular multipliers for the 3 sample and standard. These multipliers may be neglected 4 as pointed out above in which case equation ~3) can be reduced or simplified.
6 It should be obvious that various changes can be made in 7 the details and arrangements of parts and steps without de~
8 parting from the spirit and scope of the invention as defined

9 in the appended claims.
What is claimed is:

.' ., , ~

Claims

The embodiments of the invention in which an exclusive property or privilege is claimed are defined as follows:

1. The method of calibrating a spectrophotometer for measuring the reflectance of a test sample, said spectrophotometer comprising an instrument section including a light source, a detector, an opaque head having blackened inner walls defining a chamber and an opening adapted to be positioned next to a test object, first fiber optic means terminating in said chamber for directing light from said light source outwardly through said opening to illuminate a test object threat, second fiber optic means terminating in said chamber for collecting light reflected from a test object through said opening and conducting such reflected light to said detector whereby said detector provides an output signal proportional to the intensity of light reflected from said test object, the steps comprising:
a. placing a mirror next to said opening as a test object and measuring the amount of light reflected from said inner wall;
b. placing a white standard next to said opening as a test object and measuring the amount of light reflected there from;
c. calculating the reflectance of said inner wall using the amounts of light measured in steps (a) and (b);
d. placing a test sample next to said openings as a test object and measuring the amount of light reflected there-from;
e. and calculating the true reflectance of said test sample using the amounts of light measured in steps (b) and (d) and said reflectance of said inner wall.

2. The method of claim 1 further comprising the steps of:
f. placing a black standard next to said opening and measuring the output of said detector, and steps (a) and (d) each further comprise subtracting the output of said detector obtained in step (f) from the detector out-puts during steps (a) and (d) to determine the amounts of light measured therein.

3. The method of claim 1 wherein said white standard is a secondary standard calibrated against a primary white standard to provide results representing values of 100% reflectance.

4. The method of claim 1 wherein:
the true reflectance calculated is R .lambda.n.

5. In a spectrophotometer for measuring the color of a sample, said spectrophotometer having an instrument head for illuminating a test object, and means for collecting diffuse light reflected from said test object and providing an output signal proportional thereto, the combination comprising:
first means for measuring light reflected from a white primary standard test object and storing a value P indicative thereof;
second means for measuring a black standard test object and pro-viding and storing a value B indicative of the light reflected there-from;
third means for measuring a white secondary standard test ob-ject and storing a value S indicative thereof;
fourth means for calculating the reflectance Rs of said secondary standard and storing the result thereof according to ;
fifth means for measuring a sample test object and storing a value T indicative thereof;

sixth means for calculating the reflectance Rt of said sample according to ;
seventh means operative to illuminate a mirror and detect light reflected from inner walls of said head, eighth means for calculating the reflectance of said inner walls, and ninth means for calculating the true reflectance of said sample using the factor of said reflectance of said inner walls.

6. The combination of claim 5 comprising tenth means for storing values representing the relative areas of said sample and of said inner walls;
said ninth means being operative to use said relative areas to calculate said true reflectance.

7. The combination of claim 5 wherein said ninth means is operative to calculate the true reflectance as R .lambda. n.