JP2777509B2 - Color inspection method - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a color tone inspection method,
In particular, it can be used for color tone inspection of resin powder granules .

[0002]

BACKGROUND ART Conventionally, resin powders, pellets, etc.
Since the color is also important as one of the qualities in the resin-made granules described above, it was examined whether or not the color of each product was varied. In some cases, the color condition is measured as a quality inspection of a member or the like capable of inspecting the degree of deterioration by color change. A color difference meter was usually used for the color tone inspection of such resin-made granules .

[0003]

By the way, the color difference meter is
Since the color tone of an object to be measured is measured based on data obtained by measuring a standard white plate, calibration using the standard white plate is indispensable. However, since the calibration data using the standard white plate changes depending on the intensity, direction, quality, and the like of the light source, it is necessary to perform zero point and reference point alignment every hour, for example, and the calibration operation is complicated. For this reason, although it has been desired in recent years to perform color tone management at a manufacturing site, a color difference meter is not suitable for color tone inspection at a manufacturing site because calibration work is complicated.

Further, a general color difference meter uses a halogen lamp as a light source. However, a halogen lamp of a long life type has a maximum of only about 2000 hours, so that the number of lamp replacements increases. This point is also unsuitable for color tone inspection at a production site such as online / inline.

Further, the measurement data of the color difference meter is represented by values of an XYZ coordinate system, etc., but it is difficult to evaluate a sensible hue with this coordinate data. There is also a problem that it is difficult to evaluate the specific color of the inspection object because the color may be slightly different.

An object of the present invention is to provide a method for manufacturing a resin even at a manufacturing site.
An object of the present invention is to provide a color tone inspection method that can easily evaluate the color of a milled granule .

[0007]

According to the color tone inspection method of the present invention, transmitted light or reflected light from a resin granule irradiated with light is detected by a light detecting means, and an image of the resin granule is captured. Remove contour lines of resin granules from image signals
High and low brightness from the frequency distribution of the brightness of each image.
The luminance value is calculated such that the frequency of the color is the same as the RGB characteristic value.
Is performed, and the RGB signal and the Y signal of the image are converted into values in an XYZ coordinate system.
A measurement point on the XY chromaticity diagram is determined from the value of the coordinate system, and a chromaticity coefficient which is a ratio of a distance between the measurement point and the white reference point to a distance between the white reference point and the spectrum locus on the XY chromaticity diagram; And the deviation of the measured wavelength from the reference wavelength of each color is determined, and the color tone of the resin granules is inspected.

[0008]

According to the present invention, the image of the resin granules captured by the light detecting means is used to remove the contour lines of the granules.
Calculate the center of gravity of each image and Y signal
Image processing such as processing for extracting RGB characteristic values is performed.
Next , the RGB signals and the Y signals of the image are converted into values in the XYZ coordinate system, and measurement points on the XY chromaticity diagram are obtained. Next, the ratio of the distance between the measurement point and the white reference point to the distance between the white reference point on the XY chromaticity diagram and the spectrum locus at the measurement wavelength is determined as a chromaticity coefficient by a data processing device or the like. Further, the deviation of the measurement wavelength from the reference wavelength of each color is also obtained by a data processing device or the like. Then, the color tone of the resin granules is evaluated and inspected from the deviation of the chromaticity coefficient and the wavelength. At this time, the resin
Since the color tone inspection is performed based on the image data of the granular material , restrictions on the calibration frequency and light source life, such as in a color difference meter, are reduced, and the invention can be applied to the color tone inspection at a manufacturing site. Further, instead of the XYZ coordinate values, since it represents the color tone of the resin milling granular material by chromaticity coefficient and wavelength difference, it is easy to evaluate the specific hue of the resin grinding granules <br/>.

[0009]

An embodiment of the present invention will be described below with reference to the drawings. FIG. 1 shows a tree to which the color tone inspection method of the present invention is applied.
Color inspection apparatus 1 for fat made particulate material is illustrated. The color tone inspection apparatus 1 includes a sample container 2 in which a powdery or granular sample to be inspected is placed, a CCD color camera 3, illumination devices 4 and 5 for reflected light and transmitted light, and a color filter panel 6.
, A color decoder 7, an image processing device 8, a data processing device 9, and an external display device 10.

[0010] The sample container 2 holds a powdery or granular material which is a sample for which color tone is to be inspected.
At a predetermined position within the field of view. The sample container 2 is provided with a mechanism for periodically cleaning the container 2 in order to maintain reproducibility of measurement, so that the sharpness of an image can be maintained.

The CCD color camera 3 is a camera equipped with a CCD element used as a light detecting means for color tone inspection. A model which has a high detection resolution and operates stably even for long-term use has been selected. I have. Further, the color camera 3 is provided with a color filter board 6 in which several color filters of different colors are fitted on a rotating board in order to periodically measure a color reference signal at the time of color tone inspection. . Further, a stabilizing power supply is used as a power supply unit for the color camera 3 in order to improve detection accuracy.

The lighting equipments 4 and 5 are for measuring a color tone inspection image under a constant environment, and have a long life and a stable wavelength depending on the conditions of use, for example, high-frequency lighting such as a fluorescent lamp, strobe light, and the like. Light, light-emitting diodes and the like are used as light sources. Further, the illumination equipment 4 for reflected light and the illumination equipment 5 for transmitted light can be selectively used so that illumination with transmitted light or reflected light can be performed according to the transparency of the measurement sample. Further, measures such as stabilization of a power supply and prevention of disturbance light are taken to improve measurement accuracy, and a device for measuring the amount of received light is also provided in order to monitor a temporal change of a measured value due to long-term use.

The color decoder 7 separates a video signal sent from the CCD color camera 3 into RGB signals and inputs the RGB signals to an image processing device 8.

The image processing device 8 is a device capable of performing color image processing based on the RGB signal and the Y signal (luminance) from the color decoder 7, and is capable of continuous use after taking measures such as environmental protection. The model has been selected. In addition, an image processing apparatus is selected which is capable of high-speed image processing and capable of automatically processing the processing contents. In addition, the color image processing is
For example, when inspecting a sample having a large grain such as a pellet, the process of removing the contour line of each sample from the image signal and the frequency of the high luminance and the low luminance are the same from the frequency distribution (histogram) of the luminance of each image. That is, a process of calculating a luminance value that satisfies the following condition, that is, a process of calculating the center of gravity value of each of the RGB images and the Y signal and extracting the RGB characteristic values is performed.

The data processing device 9 is constituted by a personal computer or the like, and performs at least arithmetic processing of image processing results, display of the arithmetic results, control of the entire system of the color tone inspection device 1, and man-machine of the image arithmetic processing device 8. It has four functions with the interface.

The arithmetic processing of the image processing result is based on the barycentric value (characteristic value) of each of the RGB images obtained by the image processing device 8 and the barycentric value of the Y signal, and is a color serving as an index for actually managing a color tone. This is a process of performing numerical processing for calculating the deviation of the degree coefficient and the wavelength, and outputting the result to the external display device 10.

At this time, the chromaticity coefficient and the wavelength deviation are calculated in the following manner. First, the RGB characteristic value and the Y characteristic value are converted from the RGB coordinate system to XYZ using a conversion formula.
Was converted into the coordinate system, as shown in FIG. 2, obtaining the measurement points P ₁ represented in XY chromaticity diagram from the XYZ coordinates. On the other hand, a white reference point O obtained from an image without a sample, that is, a transparent image, and a color reference point S ₁ obtained by each color filter of the color filter board 6 are measured in advance, and each wavelength is measured. For example, the spectrum locus L is calculated according to JIS Z8701.
Each data of the white reference point O, the color reference point S ₁ , and the spectrum locus L is obtained from the values shown in Appendix Table 1 and stored in the data processing device 9 in advance as a database.

Next, a straight line passing from the white reference point O to the measurement point P ₁ is drawn to find an intersection P ₂ with the spectrum locus L.
Then, at the intersection point P ₂ 100, and sets the chromaticity coefficients such that 0 at the reference point O, obtaining chromaticity coefficient of measurement points P ₁ by a proportional calculation. Also, the deviation (λ ₀ −λ) between the reference wavelength λ _{0 at} the intersection S ₂ of the straight line passing through the color reference point S ₁ from the white reference point O and the spectrum locus L and the wavelength λ at the intersection P ₂ .
Ask for. The chromaticity coefficient and the wavelength deviation are obtained by performing the above numerical processing by the data processing device 9.

The data processing device 9 not only outputs the above calculation results to the external display device 10 but also displays the calculation results directly on a display unit provided in the data processing device 9. Is provided.

The data processing device 9 also has a function of controlling the timing of capturing an image into the image processing device 8, starting arithmetic processing, and outputting the measurement results to the external display device 10. Have been. Note that the color tone inspection apparatus 1 of the present embodiment can measure without setting up a sample in the sample container 2 without any manual operation. Therefore, if the process up to the unloading of the sample is automated, online measurement is also possible. In this case, if various interfaces such as a sample unloading device are added to the card throttle of the data processing device 9, the timing control of the entire system and the management of input / output signals can be performed by the data processing device 9, thereby facilitating the entire color tone inspection. Can be automated.

Further, the data processing device 9 is also provided with a man-machine interface function of the image processing device 8 so that an image processing program of the image processing device 8 can be created and various parameters can be set.

The external display device 10 is a device for finally displaying a color tone inspection result to a user. When the color tone inspection device 1 is incorporated in a chemical plant or the like, the instrumentation system of the plant is connected to the external display device 10. The color tone inspection data is fetched from the data processing device 9 and its numerical value is displayed or displayed as a trend graph. By using the external display device 10 as the instrumentation system in this way, the operation management of the color tone inspection device 1 can be performed simultaneously with the monitoring of other control loops.
However, when there is no instrumentation system serving as the external display device 10, the data can be directly displayed on the display unit of the data processing device 9 for monitoring.

Next, a color tone inspection using the color tone inspection apparatus 1 of the present embodiment will be described. First, an image input start signal is output from the data processing device 9 to activate the image processing device 8. In addition, the powdery or granular material to be a sample is placed in the sample container 2, and the lighting equipment 4 or 5 is selected according to the transparency of the sample to illuminate the sample.

A sample is photographed by the CCD color camera 3, and RGB signals of the photographed image are sent to the image processing device 8 via the color decoder 7. The image processing device 8 synthesizes a monochrome grayscale image based on the Y signal (luminance signal), and in the case of a sample having a large grain such as a pellet, removes a contour line of the sample in order to remove a disturbance component. Is performed. Then, the center of gravity of the RGB image is calculated to obtain the characteristic values of the RGB signal and the Y signal, and output to the data processing device 9.

In the data processing device 9, the RG
The B coordinate system is converted to the XYZ coordinate system, and the chromaticity coefficient and the wavelength deviation are obtained. Of the chromaticity coefficients, the XY chromaticity diagram shows the blueness coefficient, the yellowness coefficient, and the redness coefficient as shown in FIG. As can be seen from this figure, the value of the chromaticity coefficient enables the color purity of the sample to be recognized, and the change in the hue by the value of the wavelength deviation, that is, the sample of the sample if the blueness coefficient measurement area is taken as an example. Blue can be recognized as greenish or reddish.

The calculation results are displayed as numbers or points on a graph, and output to the external display device 10 as necessary. The above color tone inspection is repeated a predetermined number of times, or the sample is replaced and repeated, and the color tone inspection of each sample is performed.

According to this embodiment, the sample is
Since the chromaticity coefficient and the wavelength deviation are obtained as a color tone test for the resin-made granules, the specific color tone can be easily evaluated as compared with the conventional case where the color tone of the sample is evaluated with the XYZ coordinate values, and the XYZ coordinate values are obtained. The color tone of the resin granules can be inspected without a very professional worker who can judge the color condition from the results. Therefore, even a worker at a manufacturing site such as online or in-line can easily perform a color tone inspection.

Further, since the color tone inspection is performed based on the image data photographed by the CCD color camera 3, a fluorescent lamp or a strobe light can be used as an illumination light source. For this reason, the lamp life can be made several times longer than that of a color difference meter or the like using a halogen lamp as an illumination light source, and there is no need to replace the lamp for a long time. Therefore, it is particularly suitable for performing a color tone inspection at a manufacturing site.

Further, since the color tone inspection is performed based on the image data, even when performing the color tone inspection of the powder having irregular surface conditions, the image processing such as the contour line removal processing is performed to reduce the influence from the outside. Can be reduced. Further, a large number of samples can be measured at a higher speed than a color difference meter or the like.

Further, as shown in FIG. 3, the chromaticity coefficient for each coordinate value can be calculated in advance on the XY chromaticity diagram. Can also be stored.
In this way, the chromaticity coefficient can be obtained without performing the chromaticity coefficient proportional calculation at the time of measurement, and the processing can be performed at high speed.

Further, since the chromaticity coefficient and the wavelength deviation are calculated based on the white reference point O and the reference wavelength λ ₀ , if the zero point and the reference point are aligned only when the light source is updated, the same sample is always the same. Measurement data can be obtained. Therefore, it is not necessary to calibrate frequently like a color difference meter, and in particular, if the chromaticity coefficient data is made into a database, the correction value can be calculated from the zero point and the reference point alignment at the time of updating the light source. The wavelength deviation can be corrected, and it is possible to easily cope with the update of the light source. As described above, since the calibration is small and the data can be corrected only by the calculation, it can be easily used even at the manufacturing site.

Further, it is easy to control the entire color tone inspection device 1 by the data processing device 9 or the like. In this case, the color tone inspection can be automatically measured, and the inspection efficiency can be remarkably improved. Further, since the illumination equipments 4 and 5 for the reflected light and the transmitted light are provided, the inspection can be performed in an optimal illumination state according to the transparency of the sample.

The present invention is not limited to the above-described embodiment, but includes modifications and improvements as long as the object of the present invention can be achieved. For example, in the above-described embodiment, the CCD color camera 3 is used as the light detecting means. However, for example, an image pickup tube or the like may be used. In short, as the light detecting means, various photographing devices capable of photographing a color image of the inspection object can be used.

When the chromaticity coefficients are stored in a database in advance, as shown in FIG. 3, not only the measurement ranges of blue, red, and yellow but also the measurement ranges of other colors such as green may be stored in a database. These may be appropriately set in accordance with the type and color of the resin granules to be inspected .

Next, experimental examples and comparative examples performed to confirm the usefulness of the present invention will be described. In this experimental example and a comparative example, a resin granular material was used as a sample, and the degree of yellow, which is a measure of the degree of deterioration of the resin, and the degree of blue, which is in great demand, were measured.

As a sample A for inspecting the degree of yellowness in the resin deterioration test, Idemitsu polycarbonate and flakes (grade FN2500), which are fine powder samples, were prepared.
This sample was subjected to a Yamato Scientific Air Circulation Heating Incubator DK4
3, and the temperature of the heating incubator was set to 125 ° C., and each was untreated, taken out after 300 hours, 600 hours, and 900 hours, and taken out for color tone inspection. The powder sample A is an opaque milky white color, and the residence time of the heating incubator increases and the pale yellow color increases.

On the other hand, as a sample B for inspecting the degree of blue color, Idemitsu Polycarbonate FN2500 was used as a raw material, and Diaresint Blue N (Solvent Blu N), an anthraquinone dye, was used.
e 95) was not added, 0.5 ppm added, 1 ppm added, 2 p
pm was added, and a pellet sample prepared by granulation using a vented extruder in a usual manner was prepared. Irganox 1 which is an antioxidant to prevent thermal deterioration during granulation
010 was added at 500 ppm. Note that the pellet sample B is transparent, and the pale blue color becomes stronger as the dye increases.

The color tone inspection apparatus 1 of the experimental example is a CCD
The color camera 3 was composed of the devices shown in Table 1 such as using a SONY CCD camera having 380,000 effective pixels.

[Table 1]

The color camera 3, image processing device 8, data processing device 9, and color video copy device shown in Table 1
JEOL "Color Image Monitor JCI-3"
The product commercialized as "0" was used. For illumination, opaque samples such as the powder sample A are subjected to reflection illumination (epi-illumination) using the illumination equipment 4, and transparent materials such as the pellet sample B are colored with a dye. Transmitted illumination was performed using the illumination equipment 5.

On the other hand, as a comparative example, each of the samples A and B was inspected with a conventionally used color difference meter. At this time, the powder sample A was measured using a color difference meter # 90 manufactured by Nippon Denshoku Co., Ltd. The sample container was a # 90 powder cell having an inner diameter of 52 mm and a height of 50 mm. XYZ on the standard white plate of the color difference meter
Coordinate values of X = 95.26, Y = 93.44, Z = 113.09 were used, and as a light source, reflected illumination of a C light source specified in JIS was used.

The pellet sample B was measured using a color difference meter TC-8600A manufactured by Tokyo Denshoku Co., Ltd. The sample container used was a TC-8600A liquid cell having an inner diameter of 30 mmΦ and a height of 145 mm. Also, 0 for measuring the light intensity is used.
The / 100 setting was performed when the shutter was closed and when the sample container not containing a sample was inserted and the shutter was opened. As the light source, transmission illumination of a C light source was used.

In the pellet sample B measured by transmitted illumination,
Since the measurement results vary depending on the amount (thickness) of the sample, the measurement was performed by placing the sample in a sample container so as to have a thickness of 30 mm. In each measurement, the powder sample A and the pellet sample B were filled in each sample container so as to be full, and the measurement was repeated five times to find a standard deviation in order to detect a variation in each measurement.

Tables 2 and 3 show the measurement results of the powder sample A and the pellet sample B of this experimental example. Incidentally, a light reference point S ₁ color upon examination of the yellow was measured using a filter LA-140 of the color filter plate 6, the reference point S ₁ of the color during the light blue test color filter disc 6 Filter B
It was measured using -380. At this time, each reference wavelength λ
₀ is 580 nm for the filter LA-140 and 472 nm for the filter B-380.

[0044]

[Table 2]

[0045]

[Table 3]

Powder Sample A and Pellet Sample B of Comparative Example
Are shown in Tables 4 and 5. The powder sample A
Is measured by YI specified in JIS K7103, and pellet sample B is measured in JIS Z8729.
B ^* was measured according to L ^* a ^* b ^* specified in the above.

[0047]

[Table 4]

[0048]

[Table 5]

As can be seen from Tables 2 to 5, the standard deviation of the present experimental example is smaller than that of the comparative example, and there is little variation in each measurement, and the inspection can be performed with high accuracy. In the experimental example, each chromaticity coefficient changes according to the degree of color, and the amount of change substantially corresponds to a change in the degree of color obtained visually. Therefore, the degree of color of the sample is determined from the chromaticity coefficient and the wavelength deviation. Can be easily recognized, and the color tone inspection can be easily performed as compared with the comparative example. The usefulness of the present invention has been clarified from the above experimental examples.

[0050]

As described above, according to the color tone inspection method of the present invention , the tree which is the object to be inspected even at the manufacturing site.
There is an effect that the color evaluation of the fat-made granules can be easily performed.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a configuration of a color tone inspection apparatus to which the present invention is applied.

FIG. 2 is an XY chromaticity diagram for explaining a method of calculating a chromaticity coefficient and a wavelength deviation in the present invention.

FIG. 3 is an XY chromaticity diagram for explaining a method of calculating a chromaticity coefficient and a wavelength deviation in the present invention.

[Explanation of symbols]

 Reference Signs List 1 color tone inspection device 2 sample container 3 CCD color camera 4,5 lighting equipment 8 image processing device 9 data processing device

Claims (1)

(57) [Claims] 1. An image of a resin-made granule is captured by detecting transmitted light or reflected light from the irradiated resin-made granule by a light detecting means, and a ring of the resin-made granule is obtained from the image signal.
Removes contour lines and obtains high brightness from the frequency distribution of brightness of each image.
And a luminance value such that the frequency of low luminance is the same as that of R
Image processing for extracting the GB characteristic value is performed, and the RGB signal and the Y signal of the image are converted into values in the XYZ coordinate system.
A measurement point on the XY chromaticity diagram is determined from the value of the XYZ coordinate system, and chromaticity is a ratio of a distance between the measurement point and the white reference point to a distance between the white reference point and the spectrum locus on the XY chromaticity diagram. A color tone inspection method comprising: determining a coefficient and a deviation of a measurement wavelength of each color from a reference wavelength to inspect a color tone of the resin-made granular material.