JPH08297099A - Method and apparatus for accurately measuring gloss irregularity - Google Patents

Abstract

PURPOSE: To accurately and quantitatively measure the gloss irregularity. CONSTITUTION: A visible light is emitted from one light source 21 to the outer surface of an object 23, its reflected light is received by a television camera 24, and a two-dimensional image having obtained bright and dark gradation distribution is Fourier-transformed by an image analyzer 20. Then, the brightness and darkness of the image are bipolarized by using emphasizing coefficient in a specific wavelength range in the case of reversely Fourier-transforming it, the bright or dark image region for forming a closed section is calculated in area as gloss irregular part, and the product of the mean area and the standard deviation of the area is output from the distribution parameter of the area as the measured result.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a high-precision measuring method and apparatus for uneven gloss, which quantitatively measures uneven gloss on the surface of a product as one item of quality.

[0002]

2. Description of the Related Art In recent years, in addition to the strength of gloss on the surface of a product, uneven gloss has become more important in terms of product quality. For example, the "feel" of coated paper, the "flow mark" on the surface of molded plastic, and the "vividness" of automobile coating.

The surface feeling of the coated paper will be described. That is, it is said that the more the product has a uniform glossiness, the better the surface feeling. The evaluation has hitherto been made by an inspector visually observing the unevenness of gloss in the submillimeter region of the coated paper surface. In addition, the following inspection method has also been proposed, in which the appearance of a product is imaged by an imaging device such as a television camera, and the distribution of the brightness level (reflected light intensity) of the product appearance is calculated from the imaging result.

One of them is a modification of a chromatographic scanner, in which a luminous flux of 0.4 mmφ is scanned on the surface of coated paper at an incident reflection angle of 75 °, and the standard deviation of the intensity of reflected light is related to the visual order. (H. Fujiw)
ara et. al. , 1990 TAPPI Coa
t. Conf. Proc. , 209 (1990)).

The other is to analyze the surface reflection two-dimensional image of the printing paper input from the CCD camera to examine the correlation with the average visual order, and to find the correlation with the intensity of reflected light, that is, the variation coefficient of gradation. It's a good thing. But,
It is said that there were many cases where the uneven glossiness of white paper did not correlate <M. A. MacGregor et. a
l. , 1991 TAPPI Coat. Conf. P
rc. , 495 (1991)>.

In the proposal of Japanese Unexamined Patent Publication No. 6-22002, unlike the conventional method, uneven gloss is defined as the size (area and distribution) of a part (bright part) where the reflected light from the glossy surface is strong or a weak part (dark part). As a result, the glossy coated paper has a correlation in which the average area or area standard deviation of the dark area matches the visual evaluation.

[0007]

However, in the inspection of uneven gloss by visual inspection, individual subjectivity easily enters the evaluation and lacks quantitativeness. Therefore, a large number of specialized inspectors are required to objectively judge pass / fail. And

On the other hand, the optical inspection method for uneven gloss has the following drawbacks.

For the measurement of the optical unevenness of gloss, quantification has been attempted by the reflected light intensity distribution, but there is a drawback that it does not always correspond to the result of visual judgment. In particular, the blank surface feeling (surface feeling before printing) is strong in diffuse reflected light, and the uneven glossiness due to specular reflected light is weakened, and the optical inspection method for examining the reflected light intensity distribution has a large error.

In the proposal of Japanese Patent Laid-Open No. 6-22002, it is desired to measure with higher accuracy.

In view of the above points, the present invention is capable of quantifying and expressing the measurement result of uneven glossiness based on the size and distribution of uneven glossiness rather than the intensity distribution of light, thus improving the measurement accuracy. An object of the present invention is to provide a highly accurate measuring method and apparatus for unevenness.

[0012]

In order to achieve such an object, the invention of claim 1 takes an image of an object to be inspected by an image pickup device, and an image analysis device closes the image obtained as an image pickup result. An image area of a bright part or a dark part forming a section is detected as the uneven gloss portion of the inspection target, the area of each of the detected closed section image areas is calculated, and the distribution of the calculated closed section image area is calculated. The product of the calculated average area of the closed divided image areas and the standard deviation of the area of the closed divided image areas is used as the measurement result of the uneven gloss.

According to a second aspect of the present invention, the lightness or darkness of the image pickup result is emphasized when detecting the image area of the bright portion or the dark portion forming the closed section.

According to a third aspect of the present invention, an image pickup means for picking up an image of an object to be inspected, and an image area of a bright part or a dark part forming a closed section in the image obtained as a result of the image pickup is defined as a gloss of the object to be inspected. Detection means for detecting the uneven portion,
A first calculating unit that calculates the area of each of the detected closed section image regions, and a distribution of each of the calculated areas is calculated by calculating the distribution of each of the calculated areas. It is characterized by further comprising a second calculating means for outputting a product of an average area of the divided image area and a standard deviation of the area of the closed image area as a measurement result of uneven glossiness.

[0015]

According to the first and third aspects of the invention, the inventor of the present invention has found that the unevenness of the gloss of the product appearance causes a difference in lightness and darkness, so that a bright portion or a dark portion forms a closed section in the captured image. Then, the area distribution of the image of the closed section is calculated, and the product of the average area of the closed section image area and the standard deviation of the area of the closed section image area is obtained to accurately quantify the measurement result of uneven gloss. .

According to the second aspect of the present invention, the uneven brightness portion can be accurately identified by emphasizing the lightness and darkness of the image pickup result.

[0017]

Embodiments of the present invention will be described below in detail with reference to the drawings.

FIG. 1 shows the basic configuration of the embodiment of the present invention.

In FIG. 1, reference numeral 1000 denotes an image pickup means for picking up an image of an inspection object.

Reference numeral 1100 is a detecting means for detecting an image area forming a closed section from the image obtained as a result of the image pickup, as the uneven glossy portion of the inspection object.

Reference numeral 1200 is a first calculating means for calculating the area of each of the detected closed section image regions.

Reference numeral 1300 denotes a product of the calculated average area of the closed divided image area and the standard deviation of the area of the closed divided image area, by calculating the distribution of the calculated area, and the calculated result of unevenness in glossiness. Is a second calculation means for outputting as.

As will be described later, the functions of the detecting means 1100 and the first and second calculating means 1200, 1300 are realized by the arithmetic function of the image analysis processing device.

FIG. 2 shows a schematic structure of a measuring apparatus to which the present invention is applied.

In FIG. 2, a light source 21 in the image pickup device (image pickup means 1100 in FIG. 1) 10 for an inspection object 23.
Light is projected through the polarization filter 22. The light reflected by the object 23 is received by the television camera 24 via the polarization filter 25. The television camera 24 converts the received light into an image signal for each pixel, and the image analysis device 2
Output to 0.

Visible monochromatic light or white light is preferably used for the light source 21 used in this embodiment. Further, the incident light on the surface of the object 23 is preferably parallel light or light in a state close to parallel light. If the light is not parallel light or light close to parallel light, uneven gloss may be difficult to capture over the entire image. The incident angle is preferably 15 ° to 35 °, but is not limited to this.

In this embodiment, the reflected light received by the television camera 24 is -10 ° to +1 with respect to the value of the incident angle.
It is preferable to capture at a reflection angle of 0 °. If it deviates from this range, it becomes difficult to capture the specularly reflected light, which is the main component of gloss.

By passing the incident light and the reflected light through the polarization filters 22 and 25 having the same phase, the diffuse reflected light is blocked,
Transmits specularly reflected light. Therefore, uneven gloss is emphasized more.

The television camera 24 used in the present embodiment refers to a device for obtaining a two-dimensional (planar) image of the object 23, and what is used is one that can capture a change in brightness and darkness with respect to a position in the plane by an electric signal. Good. C as commercially available
Although there are a CD camera, a video camera, and the like, it is preferable to use a CCD camera in which light and dark and electric signals have a linear relationship. The focal length of the lens of the TV camera is 0.5 mm or less when the image resolution is expressed by the size of the surface of the object,
And the measurement area of the object corresponding to the whole image is 2 x 2 mm
It is preferable to select the above.

The present invention is characterized by an image signal analysis method, and the image analysis processing relating to the present invention will be described.

The image signal for one screen obtained by the television camera 24, that is, the brightness value (gradation expression) for each pixel is stored in the memory in the image analysis device 20 and then read by the central processing unit. The image is analyzed in the following processing steps.

(1) Fourier transform is applied to the two-dimensional image by the following equation.

[0033]

[Equation 1]

The result of the Fourier transform shows the intensity distribution for each frequency band and is called a power spectrum.

(2) When an inverse Fourier transform of the Fourier transform result is performed, the portion corresponding to the specific wavelength range that can be recognized by the naked eye is multiplied by the enhancement coefficient to obtain an enhanced image in which the lightness and darkness of the two-dimensional image at the time of image pickup is enhanced. . The inverse Fourier transform formula is represented by the following formula.

[0036]

[Equation 2]

(3) In the obtained emphasized image, an image area portion in which a white portion corresponding to a light portion or a black portion corresponding to a dark portion forms a closed partition is detected as a gloss unevenness portion, and the closed partition for detection is detected. The area, average area, standard deviation, coefficient of variation, or number of white or black areas per unit area for each of the regions is calculated as a gloss unevenness determination parameter corresponding to visual evaluation. Further, these parameter values quantitatively represent the degree (distribution) of uneven gloss.

Here, the white part and the black part refer to independent closed partition regions having gradations corresponding to white and black, respectively.

For reference, the equations for calculating the above parameters are used.
The formula is shown below. Further, it goes without saying that the image analysis apparatus when executing this process operates as the detecting means 1100, the second calculating means 1200, and the second calculating means 1300 of FIG.

[0040]

(Equation 3)

[0041]

[Equation 4]

[0042]

(Equation 5)

The lower limit of the specific wavelength range is 0.
It is preferable to set the wavelength range such that the wavelength range is 004 to 2.0 mm and the upper limit is 2.0 to 200 mm. The lower limit is 0.
If it is out of the range of 004 to 2.0 mm, it is different from the resolution of human vision, and it becomes impossible to obtain an effective measurement value. Also, if the upper limit is outside the range of 2.0 to 200 mm, it is different from the size perceived by humans as a difference, and an effective measurement value cannot be obtained.

The emphasis coefficient in the equation (2) is preferably 2 to 50 times. If it is less than 2 times, the polarization of light and darkness cannot be obtained, and if it exceeds 50 times, the uneven shape does not match the visually recognized one, and it becomes difficult to obtain an effective measurement value.

When the above arithmetic processing is calculated by a computer, a two-dimensional image is divided into a group of pixels, and the gradation of each pixel is input by digitizing an image signal.

In the image analysis processing described above, the operation of Fourier transforming and inverse transforming a two-dimensional image to obtain a white part or a black part is a region setting which is conventionally called a "threshold" and which tends to be arbitrary. Since it can be set under the same conditions for all samples, measurement values with good reproducibility and reliability can be obtained.

The method and apparatus for quantitatively measuring the unevenness of gloss of the object to be inspected are suitable as a method for replacing the conventional method for visually judging the unevenness of gloss of the product surface, and do not require any arbitrary operation. It has the feature that quantitative values with good reproducibility and reliability can be easily obtained without being influenced by the experience and judgment of the person.

Therefore, according to the present invention, the minute gloss unevenness of the coated paper, the white surface and the printed surface, as well as the minute gloss unevenness of the molded plastic surface, various coated surfaces and other articles where importance is attached to the aesthetics and smoothness of the surface are quantified. The present invention provides a method and device useful for clinical evaluation.

The results of the actual measurement of uneven glossiness will be introduced for reference. The coated paper samples are commercially available glossy fine coated papers for sheet offset (4 types of A0 to A3,
5 groups, 33 samples) were used.

The apparatus shown in FIG. 2 was used as the apparatus for measuring the minute gloss unevenness of white paper. White visible parallel light (light irradiation device HT-
2: Nikon) is polarized and irradiated onto the surface of the coated paper at an incident angle of 25 ° with respect to the normal line, and the reflected light of the same 25 ° is passed through a polarization filter of the same phase as the incident light and a CCD camera. (SONY: CE-75).

The photographing area of the surface of the object per pixel is 4
The measurement area was 0 × 40 μm and 10 × 10 mm.
The data were taken at 4 points per sample, and the arithmetic mean thereof was used as the value.

The brightness of each pixel was converted into a digital gradation by an A / D converter. 256 tones from black to white
The level was divided into 0 and black was set to 0 and white was set to 255. The amount of light was set so that the gradation distribution of the sample with the highest glossiness in each group of coated paper samples does not reach the gradation level 255.

Image analysis (Image analysis device IP-1000:
Asahi Chem. Ind. ) Conditions,
After the two-dimensional Fourier transform of the original image, the specific wavelength range that can be recognized by the naked eye at the time of the inverse transform is multiplied by an emphasis coefficient to emphasize the uneven brightness of the image. In the gradation of the image in which unevenness is emphasized, the range of 0 to 1 is <black part>, 254-2
The range of 55 was defined as <white part>. After the area of each of the black part (dark part in the original image) and the white part (bright part in the original image) was obtained, its standard deviation was calculated.

With respect to the visual level, 7 to 18 experts involved in paper quality evaluation inside and outside the company gave a rank, and the arithmetic mean value of the rank was taken as the visual average rank. The order of ranking was given as integers such as 1, 2, 3, ...

Table 1 shows the results of the visual evaluation ranking of commercially available A0 to A3 gloss papers. Table 1 shows the average rank of the visual evaluation, the standard deviation of the highlighted white and black areas corresponding to the uneven gloss of the coated paper, and the average area. In this case, the emphasis wavelength range at the time of inverse Fourier transform was 0.4 to 8.0 mm, and the emphasis coefficient was 10 times. The correlation coefficient of the standard deviation of the black portion showed a value of 0.90 to 0.97, and the correlation coefficient of the average area of the black portion was 0.89 to 0.97. It was the product of the average area of the black part and the area standard deviation of the black part that showed a better correlation coefficient than those, and the correlation coefficient was 0.94 to 0.9.
8 (see Table 1).

[0056]

[Table 1]

In addition to this embodiment, the following example can be realized.

(1) In this embodiment, the processing up to the measurement of the uneven gloss is explained, but the appearance inspection can be automatically carried out by using the measurement result. In this case, the presence or absence of abnormality is determined by comparing the parameter value obtained as the measurement result with a predetermined threshold value. At this time, it is advisable to display the inspection result and an image emphasizing the imaging result on the display.

(2) To visually output the measurement result of uneven glossiness, the following forms can be used.

(A) The calculated parameter value is numerically output by a display device or a printer.

(B) The calculated parameter value is graphically output by a display device or a printer.

(C) The degree of uneven glossiness is converted into a numerical value by a numerical operation expression using a plurality of parameter values, and the degree of unevenness of glossiness is classified into "large", "medium", and "small" according to the numerical value. Display a message like this.

(3) In the present embodiment, detailed processing for detecting the image area of the closed section from the image has not been described, but this processing can be performed by using various well-known methods as contour line extraction processing. Good.

(4) When it is desired to make the uneven glossiness or the boundary of the portion clear, it is preferable to emphasize and display the contour line forming the closed section.

[0065]

As described above, according to the present invention, it is possible to accurately measure the uneven glossiness quantitatively, and thus it is possible to further know the degree of uneven glossiness.

[Brief description of drawings]

FIG. 1 is a functional block diagram showing a basic configuration of an embodiment of the present invention.

FIG. 2 is a structural diagram showing the structure of an embodiment of the present invention.

[Explanation of symbols]

 10 Imaging Device 20 Image Analysis Device 21 Light Source 22 Polarizing Filter 23 Object 24 Television Camera

Claims (3)

[Claims] 1. An image of an inspection target is picked up by an image pickup device, and an image area of a bright part or a dark part forming a closed section is detected as an uneven glossy part of the inspection target by an image analysis device. Then, the area of each of the detected closed partition image areas is calculated, the distribution of the calculated closed partition image areas is calculated, and the calculated average area of the closed partition image areas and the standard of the area of the closed partition image areas are calculated. A high-accuracy measuring method for uneven gloss, wherein the product of the deviation is used as the uneven gloss measurement result. 2. The high-precision measurement method for uneven glossiness according to claim 1, wherein when detecting an image region of a bright portion or a dark portion forming the closed section, the lightness and darkness of an imaging result is emphasized. 3. An image pickup device for picking up an image of an inspection target, and an image region of a bright portion or a dark portion forming a closed section is detected as a gloss uneven portion of the inspection target from an image obtained as a result of the image pickup. Detecting means, first calculating means for calculating the area of each of the detected closed partition image regions, and calculating each distribution of the calculated area, thereby calculating each distribution of the calculated area, The gloss unevenness is characterized by comprising a second calculation means for outputting the product of the calculated average area of the closed partitioned image areas and the standard deviation of the area of the closed partitioned image areas as the measurement result of the gloss unevenness. High precision measuring device.