JPH11271237A - Plate inspecting system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To perform an inspection of appearance quality on a line relative to a lengthy building board having a multicolor tone or irregular surface form by providing a signal processing means for processing the spectral reflectance data from an electronic camera for taking the image of a plate body. SOLUTION: A building board inspecting system is provided with a color electronic camera for taking the image of a building board 1 and an image processing means for performing a different image processing for inspection to the illuminance signal and color signal from the color electronic camera 20. Since the image processing means Fourier-transform the illuminance signal of the signal guided from the illuminance signal, the distribution of frequency components of the building board 1 can be detected, and the characteristic of texture can be determined from the result. According to this, the spatial information which is lost in concentration histogram can be utilized, and when the building board 1 having a highly irregular surface form is to be measured, particularly, its precise evaluation can be performed.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a board inspection system for inspecting a plate-like body such as a building board, and more particularly to an inspection board system capable of inspecting the appearance quality on a line. I do.

[0002]

2. Description of the Related Art Conventionally, in various production fields, the color of a colored object has been measured using a photoelectric colorimeter or a spectrophotometer.
It is used for production control and product quality (color etc.) control. In addition, as an alternative to the conventional color matching work by a skilled person, there is an example in which a color matching calculation, a correction combination calculation, and the like by a CCM (computer color matching system) are performed.

In the field of painting and printing, colorimetric methods using a photoelectric colorimeter or a spectrocolorimeter have been studied and implemented, and numerical evaluation of color differences (absolute value comparison and , Relative comparison with a standard product, etc. are performed). More specifically, a test piece whose coating or ink appearance quality (such as color and gloss as optical properties) is adjusted to a measurable size is measured and inspected. In some cases, a handy-type colorimeter is used to go to a production site or the like and check the product directly but manually. On the other hand, in the final process of a production line or an inspection line, a line colorimeter (often a desktop type colorimeter is used) is incorporated to perform color management. A method of performing color management by performing image processing on a computer has also been proposed.

[0004]

However, the latter TV
In the method using a camera, usually, the imaging conditions are not strictly defined, and further, since the object to be measured is imaged in a non-contact manner at a predetermined distance,
Although it is possible to obtain RGB data for valid pixels, it is not possible to obtain accurate colorimetric values.

On the other hand, in the former method using a line colorimeter, accurate colorimetry can be obtained because the colorimeter is used. FIG. 15 shows an example of a typical conventional spectral colorimetric device. FIG. 15A shows an example of an apparatus using a diffraction grating, in which a sample 71 is brought into contact with an integrating sphere 72 painted in white, which diffuses and reflects the inner surface, and guides a light beam from a light source 73 to the integrating sphere 72. , Sample 71
The light beam reflected from the surface and the inner surface of the integrating sphere 72 and exiting through the window is converged by the lens 73, passed through the slit 74, converted into parallel light by the # 1 concave mirror 75, and then separated by the diffraction grating 76 in different directions according to the wavelength of the light. Then, the light is converged by the # 2 concave mirror 77, and light having a different spectrum is received by each diode element of the photodiode array 78 as a light receiver, amplified by an amplifier 79 and converted into an analog signal by an A / D converter 80. The signal is converted into a digital signal, processed by the CPU 81, and displayed and printed. FIG. 15B shows an example of an apparatus using a prism, and a light source 91 corresponds to the integrating sphere 72 in FIG. 15A, in which reflected light from a sample is emitted, converged by a lens 92, and incident. After passing through a slit 93, the light is collimated by a # 1 lens 94, then separated by a prism 95 in different directions depending on the wavelength of the light, converged by a # 2 lens 96, and passed through an exit slit 97 that can slide horizontally.
A light receiver 98 receives light of each spectrum.

However, since a colorimeter is premised on performing colorimetry on a substantially flat object such as a plate or cloth having a monochromatic colored plane, a plate-like body having an uneven surface shape, In the case where it is necessary to measure a plate-like body having a mixed color tone of two or more colors, or a printed board having a repeated pattern, the colorimetric value is extremely ambiguous, and its application is almost impossible. I have to say that it is impossible. Further, in the ordinary color measurement method, a method of measuring a small area (for example, about φ10 mm) in a spot-like manner with the sensor portion substantially in contact with a test plate is adopted. Naturally, it is necessary to widen the measurement range for the object. Therefore, it is difficult to apply such a large object to the case of performing a visual inspection using a normal measurement method.

Therefore, when an attempt is made to introduce a colorimetric process on an inspection line of a building board or the like, the non-contact measurement and the fact that a relatively wide area can be imaged can be realized.
It can be said that it is indispensable from the viewpoint of line applicability. In other words, basically, it is assumed that a measurement mode is adopted in which a building board is imaged by a TV camera and the acquired image data is processed by a computer.

However, as described above, since the colorimeter is not used by the TV camera alone, accurate color measurement cannot be performed. For this reason, it is still necessary to rely on visual judgment by a skilled person, and a colorimetric system with line applicability has not yet been established. Further, in the case of the conventional method using a diffraction grating or a prism shown in FIG. 15, since the spectrophotometer is an extremely precise optical device, there is a problem in line applicability even if a protective device is provided. . The present invention has been made in view of such a situation, and even for a long building board having a multicolored tone or an uneven surface shape, which was conventionally impossible to measure color, inspection of appearance quality on a line. It is an object of the present invention to provide a plate inspection system that enables the inspection.

[0009]

SUMMARY OF THE INVENTION A plate inspection system according to the present invention is for inspecting a plate-like body, and a color electronic camera for imaging the plate-like body, and a luminance signal and a color signal from the color electronic camera. And image processing means for performing different image processing for inspection. Further, the image processing means performs Fourier transform on the luminance signal or a signal derived from the luminance signal, so that the distribution of frequency components of the plate-like body can be detected, and the characteristic of the texture can be obtained based on the result. Can be. This makes use of spatial information that is lost in the density histogram, and enables accurate evaluation of the image, particularly when a plate-like body having a surface with large irregularities is to be measured.

Further, the image processing means obtains a signal corresponding to the ratio between the absorption coefficient and the scattering coefficient of the plate from the color signal or a signal derived from the color signal. The color of the plate-like body can be evaluated on a scale suitable for the following. Further, the image processing means obtains an average value and a variance of the color signal or a signal derived from the color signal, so that the color of the plate-like body can be objectively evaluated. In addition, since the color electronic camera captures a still image, it is not necessary to capture an unnecessary image, so that the bandwidth can be widened by that much, and highly accurate spectral reflectance data can be obtained. be able to.

[0011]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Preferred embodiments of the present invention will be described below in detail with reference to the drawings. Referring to FIG. 1, there is shown an overall side view of a plate detection system according to the present embodiment. The plate-like body, for example, the building board 1 is conveyed on a conveyer arranged continuously. On the conveyor,
There is a carry-in conveyor 2, a first conveyor 4 in the measurement booth 3, a third conveyor 5, and an unloader 6 outside the measurement booth 3. In addition, the first conveyor 4 and the third conveyor
A measuring unit 10 is provided between the conveying unit 5 and a second conveying conveyor 11 for the measuring unit 10, and a reference white plate 8 and a table lifter 7 for vertical movement are provided below the measuring unit 10.
Is provided. On the table lifter 7, a white plate mounting table 9 on which a white plate 8 serving as a reference for color measurement rides is provided. The measurement booth 3 has an entrance shutter 3I and an exit shutter 3E to prevent light from entering the booth 3 from outside during measurement. The measurement booth 3 has an inner wall,
The ceiling and floor are painted black.

The building board 1 to be measured is moved from the carry-in standby position P1 to the measurement standby position P2 and the measurement point MP.
1, MP2, MP3, a measurement completion position P3, and an unloading completion position P4. Next, FIG.
The measuring unit 10 will be described in detail with reference to (a) and (b). FIG. 2A is a front sectional view of the measuring unit 10. A measuring box B whose lower surface is close to the measuring unit 10 where the building board 1 conveyed on the second conveyor 11 stops for color measurement, to such an extent that the running of the building board 1 is not hindered.
Is provided. Above the measurement box B, a CCD camera 20 that captures a still image of the building board 1 is arranged. D ₆₅ standard light source 22 with a color temperature conversion filter 21 is provided on both sides of the CCD camera 20. For example, the standard light source 22 tilts the building board 1 to be imaged at an angle of 45 °.
It is preferable to irradiate from. The lower part of the measurement box B is covered with the diffusion prevention cover 23, and the inner wall surface is painted black. In order to prevent light from leaking outside, only the tip of the CCD camera 20 is configured to enter the measurement box B and receive the reflected light emitted from the standard light source 22 and reflected by the building board 1. Have been.

FIG. 2B is a perspective view of the measuring unit 10. The building board 1 conveyed below the measurement box B is conveyed by the second conveyor 11, and the building board 1 is guided by a guide 24 in a traveling direction indicated by an arrow. On the other hand, below the measurement box B, a reference white plate 8 (shown in FIG. 1) having a white color serving as a color reference at the time of color measurement is arranged. The reference white plate 8 is a plate slightly larger than the imaging area of the CCD camera 20, has the same surface shape as the building board 1, and uses, for example, MgO coated with a predetermined coating film. The reference white plate 8 is mounted on a white plate mounting table 9 and is moved up and down by a table lifter 7. This elevating operation is performed by, for example, an electric hydraulic cylinder (not shown).

FIG. 3 is a configuration diagram of a CCD camera 20 serving as a colorimeter. The light reflected by the building board 1 is incident from the imaging lens 33 of the CCD camera 20, passes through the rotating plate with color filter RP, and further passes through the optical low-pass filter 34 in order to avoid aliasing distortion caused by sampling.
The image reaches the imaging device array matrix 35. This CC
The D imaging device array matrix 35 is driven by a CCD driving circuit 36, and data obtained by the CCD imaging device array matrix 35 is sent to an electronic circuit 37. The data received by the electronic circuit 37 is converted from an analog signal to a digital signal by an A / D converter 38 and stored in an image memory 39. On the other hand, the CCD camera controller CC sends the M1 stepping motor 42, M2 stepping motor 43, M
The three-stepping motor 44 is controlled. These motors can perform accurate position control by micro-step control.
Is used for auto-focusing of the imaging lens 33, the M2 stepping motor 43 is used for controlling the rotation of the rotating disk RP with the color filter, and the M3 stepping motor 44 is used for moving the CCD imaging device array matrix 35. Accordingly, the wiring between the CCD drive circuit 36 and the electronic circuit 37 is a flexible wiring. The data collected by the CCD camera 20 is transmitted to the image processing controller 46 via the input / output interface 40.
(Shown in FIG. 6). Further, a control signal from the main controller 45 is transmitted to the CCD camera controller CC via the input / output interface 40.

In the present invention, the spectral reflectance is obtained in place of the usual CCD camera specification in order to measure the color of the building board 1 having a multi-color tone and an uneven surface shape. And
The color is digitized based on the acquired spectral reflectance data.
For example, the value of X (also for Y and Z) in the CIE XYZ color system is given by the following equation.

[0016]

(Equation 1)

The present invention applies a multi-band photography method to the CCD camera 20. According to the multi-band photography, color information of a subject is recorded on a panchromatic film (a photographic film sensitive to light of all colors) through a plurality of channels having different spectral sensitivities. Thereafter, the densities of the plurality of panchromatic films are read by a color scanner, and the spectral reflectance of the subject is restored from the information. In this case, eight channels are sufficient as the number of channels (that is, to reproduce the original color). In the present invention, a matrix 35 composed of a large number of CCD image pickup devices built in the CCD camera 20 is regarded as a panchromatic film, and reflected light from the building board 1 entering through the image pickup lens 33 is reflected by eight color filters CH1 to CH8. Through the channels in order.

FIG. 4 is a front view of the rotating disk RP having the eight channels. The rotating disk RP is provided with eight channels CH1 to CH8 each having a color filter attached thereto, and a hole CH0 which is not provided with a color filter. Here, the eight channels to which the color filters are attached are configured as follows. Visible light wavelength range 380nm ~
At 780 nm, at 50 nm intervals, 380-430 nm (purple range) → 1CH 430-480 nm (blue range) → 2CH 480-530 nm (green range) → 3CH 530-580 nm (yellow range) → 4CH 580-630 nm (orange range) → 5CH 630-680 nm (first red range) → 6CH 680-730 nm (second red range) → 7CH 730-780 nm (third red range) → 8CH

Using these eight channels as the sampling wavelength range of the spectral reflectance data, a spectral image of one screen taken for each channel is taken as a CCD image pickup device array matrix 35 (for example, having an 800 × 500 pixel configuration). Image on top. The spectral image for one screen obtained here has information on luminance color data as the spectral reflectance of each pixel (having positional information in one plane on one plane) constituting one captured screen.

The CCD image pickup device 35 used is an all-pixel readout IT-CCD (interline transfer CC).
D). That is, in this device, three electrodes of the vertical transfer CCD correspond to one photodiode, and the signals of all the photodiodes are read at one time. Therefore, it has the feature that the vertical resolution of a still image can be increased. Thus, in the present invention, eight pieces of spectral image information are acquired. A multiplexed spectral image obtained by multiplexing these spectral images displays the measurement object.

Next, a measurement time chart will be described with reference to FIG. First, the shutter 3 at the entrance of the measurement booth 3
I and the exit shutter 3E are opened, and all five conveyors 2, 4, 5, 6, 11 are operated at the same speed,
First, the reference plate 1 as a reference building plate is transported from the carry-in standby position P1 to the measurement standby position P2. At this time, the carry-in conveyor 2 and the first conveyor 4 contribute to the conveyance of the reference plate 1. In the steady operation state, the building board 1, for which color measurement has already been completed, is carried out of the measurement booth 3 from the measurement completion position P3 to the unloading completion position P4 by the third conveyor 5 and the unloading conveyor 6. Reference plate 1
Enters the measurement booth 3 and reaches the measurement standby position P2, the five conveyors 2, 4, 5, 6, and 11 are stopped, and the entrance shutter 3I and the exit shutter 3E are closed. When the shutters 3I and 3E are closed, the inside of the measurement booth 3 is shielded from the outside to form a dark room. In order to measure each of the front part, the central part, and the rear part of the building board 1, the first transport conveyor 4, the second transport conveyor 11, and the third transport conveyor 5 are intermittently operated at the same speed and intermittently. During this time, the reference plate 1 has the measurement points MP1, M
It stops at each position in the order of P2 and MP3, and measurement (1), (2), and (3) is performed at each position. That is, the CCD camera 2
0 is used to determine image data. When the imaging at each position is completed, the reference plate 1 reaches the measurement completion position P3, opens the entrance shutter 3I and the exit shutter 3E, and operates the five conveyors 2, 4, 5, 6, and 11 at the same speed. , Reference plate 1
Is transported to the unloading completion position P4. In this case, the third
The transport conveyor 5 and the unloading conveyor 6 contribute to the transport of the reference plate 1. As described above, at this time, the inspection plate 1, which is a building plate to be measured next, is carried into the measurement booth 3 from the carry-in standby position P1 to the measurement standby position P2.

Next, the process shifts to the measurement of the inspection plate 1, and the inspection plate 1 is moved to the measurement points MP1 and MP1 according to the same time chart.
By transporting the test plate 1 to MP2 and MP3, the front, center, and rear portions of the inspection plate 1 are imaged. FIG. 6 is a block diagram showing the basic configuration of the present invention. The image signal captured and stored by the CCD camera 20 is sent to an image processing controller 46 that processes the image signal. The main controller 45 controls the CCD camera 20 and the image processing controller 46. The main controller 45 is connected to a keyboard 47 for inputting necessary data, a display 48 for displaying required data, and a printer 49 for printing required data. The transport conveyors 2, 4, 5, 6, 11, the entrance and exit shutters 3I and 3E, and the table lifter 7 are respectively a transport conveyor controller TC, a shutter controller SC, And a table lifter controller RC. Further, the main controller 45 is connected to a standard light source controller 50 that detects a change in the spectral energy distribution of the standard light source 22 and performs a correction calculation later to use for color evaluation.

FIG. 7 is a diagram showing a recording data structure in the image memory 39 shown in FIG. This data area has, for example, a storage capacity of 500 pixels × 800 pixels. The calibration data obtained from CH0 of the CCD camera 20 is stored in the area 60. Similarly,
Data of the reference plate 1 obtained in CH0 to CH8 are stored in the areas 51 to 59, respectively. As described above, the areas 61, 62, and 63 respectively include CH0 to CH8 corresponding to the sampling data at the front part of the test plate, the sampling data at the center part of the test plate, and the sampling data at the rear part of the test plate.
Is stored. The measurement data stored in the data area 60 is updated for each calibration and the data area 51 is updated.
The measurement data stored in .about.59 is updated each time the type of the inspection plate 1 is changed and the measurement of the reference plate 1 is performed. The measurement data stored in the data areas 61 to 63 is Updated for each measurement.

Referring to FIG. 8, various possible surface shapes of the building board 1 are shown. In the present invention, the surface features of the building board 1 having various surface designs, patterns, and irregularities are evaluated by classifying them into large, statistical textures and structural textures. That is, the present invention provides surface features that are difficult to evaluate in normal color evaluation,
Clarify by texture analysis. Here, the texture refers to a fine pattern or pattern uniformly distributed on the screen, and specifically, a sandy ground or a lawn is given as a statistical texture. have. Examples of the structural texture include a print pattern of a fabric and the like, and have a regular pattern structure.

That is, the surface characteristics of the building board 1 include:
A thing having a surface shape such as a brick pattern, a thing having a surface shape having large irregularities such as a molded article, such that a mountain part and a valley part are mixed and mixed, and fine sand and beads having different sizes are formed. Some have a natural stone-like surface shape that is scattered and has a rough surface, and others have a regular pattern like a printed pattern. In the present invention, in order to evaluate such various characteristics of the surface of the building board 1 from the color plane, images of the measurement surface taking various aspects of the building board 1 are subjected to data processing using the concept of “texture”. I do.

FIG. 8 shows examples of "statistical texture", "structural texture" and "intermediate texture". FIG. 8A shows an example of the surface of a “statistical texture” having a random speckle pattern, FIG. 8B shows a random fine irregular concave and convex granular pattern, and FIG. 8C shows a directional fine pattern. ) Is a brick pattern divided by vertical and horizontal grooves, (e)
Is a regular print pattern, (f) is a mosaic pattern,
(G) is an example of a “structural texture” having a surface shape having relatively large uneven portions, (h) is a spot pattern in a portion divided by a horizontal groove, and (i) is a fine uneven portion. It is an example of "intermediate texture" having a surface shape having.

Among these, regarding the analysis of “statistical texture”, the average value (μ) and the variance (σ)
² ), the degree of distortion (D: shows how much the shape of the density histogram H (z) is distorted from the symmetric shape), and the degree of kurtosis (K: how much the distribution of the density histogram H (z) is To show similarity of statistics). When a pigment or pigment having a high scattering property is used, the α / S index is used in place of the concentration according to the following equation using Kubelka-Munk theory.

Α / S = (1−R) ² / 2R (2) α: absorption coefficient S: scattering coefficient R: spectral reflectance

Equation (2) applies only to monochromatic light. Expression (2) is a basic expression, and an optimal relational expression corrected according to each case should be used. Therefore, the obtained spectral reflectance for each channel may be converted into an α / S index, and the above-described statistical quantities may be obtained and evaluated. Among such statistical quantities, the degree of distortion D and the degree of kurtosis K are obtained by the following equations.

[0030]

(Equation 2)

It should be noted that in the density histogram H (z), the fact that spatial information of pixels is lost is a precautionary item that must be remembered in the evaluation. On the other hand, the analysis of “structural texture” is as follows. That is, since the periodic information on the brightness distribution of the surface of the building board 1 that has been captured has reached the light receiving surface (imaging surface), the acquired image data having the brightness information on one screen is displayed. By performing the Fourier transform, the distribution of the frequency component can be detected, and the feature of the texture is obtained from the result. In this method, spatial information that is lost in the density histogram is alive. Especially when a building board 1 or the like having a highly uneven surface shape is to be measured, a correct evaluation of the image is required. It becomes possible.

The Fourier transform is based on the theory that various waveforms can be created by superimposing a sine function and a cosine function. In fact, considering the infinite sum of the sine function and the cosine function, the Fourier series All functions can be expressed. The reflected light from each part of the subject reaches corresponding parts of the light receiving surface and forms an image. Therefore,
If an image signal obtained by capturing the image is subjected to Fourier transform, the reflected light from each part of the subject that has reached the light receiving surface can be subjected to spatial frequency analysis. However, since the image has a two-dimensional spread, the frequency also has two frequencies, the horizontal direction and the vertical direction. In performing such a two-dimensional Fourier transform, after performing the Fourier transform in the horizontal direction, May be Fourier-transformed. In this case, the Fourier transform is performed by a high-speed Fourier transform called FFT.

As described above, the Fourier transform can be said to be a method of expressing an image with a frequency. Specifically, an image with a low frequency becomes a “rough image”, and conversely, an image with a low frequency A high image can be considered a “fine image”. If the result after the Fourier transform is represented by an image (that is, a frequency coordinate space image), the result is as shown in FIG. 9 and the frequency distribution state can be grasped. In FIG. 9, the vertical axis is the vertical frequency, and the horizontal axis is the horizontal frequency. The outer circle indicates the high frequency region HR, and the inner circle indicates the low frequency region LR.

In addition to the above-described color filter configuration of 8CH, in order to obtain a Fourier transform image that requires only luminance information, a hole without a color filter is provided as CH0, and CH0 to CH8 are placed on the disc RP. Formed. When obtaining the luminance information, the rotation of the disk RP is controlled to match CH0,
In order to adjust the optical path length as much as there is no color filter,
By moving the CCD image pickup device array matrix 35 by the M3 stepping motor 44, color information and luminance information can be obtained in a time-division manner. With such a configuration, it is possible to analyze both the “statistical texture” and the “structural texture”. Finally, a multiplexed spectral image in which spectral images of each channel are collected is obtained, and is displayed on the display 48 or printed out so as to be visually confirmed.

Accordingly, it has become possible to evaluate a building board 1 having a multi-color tone and an uneven surface shape characteristic, which has been conventionally difficult to evaluate by colorimetry. FIG. 10 is a flowchart showing a control operation of the main controller 45 shown in FIG. First, it is determined whether or not the system power is ON (step S1). If NO, the process waits until the system power is turned ON. If YES, initial settings such as data including settings for receiving measurement data from the CCD camera 20 are performed (step S2).

In step S3, the CCD camera 20
Perform calibration. That is, measurement booth 3
The entrance shutter 3I and the exit shutter 3E are closed, and the reference white plate 8 is raised by the table lifter 7,
The reference white plate 8 is positioned at an intermediate position (determined in advance from a design value) between the top and the bottom of the reference plate 1 in the thickness direction with respect to the conveyance back surface position of the reference plate 1. And measure the white reference level.
Upon receiving the sampling data from the camera 20, the data is transferred in order and recorded in the image memory 39. At this time, the standard light source 22 detected by the standard light source controller 50 is used.
The correction calculation value obtained based on the variation of the spectral energy distribution is taken into account. When this measurement is completed, the reference white plate 8 is lowered.

This calibration is performed before the start of the measurement. After that, since the characteristics of the CCD camera 20 may change depending on the temperature, the calibration is frequently performed at the beginning of the measurement when the temperature rises. It is good to do it gradually. As shown in FIG. 10, in the colorimetric procedure, the timing at which the type of the inspection plate 1 is changed is appropriate, but it is not always necessary to perform the timing at this timing.
You may go at other times.

In step S4, the colorimetry of the reference plate 1 (described in detail in FIG. 11) is performed. In step S5, the colorimetry of the inspection plate 1 is performed. In step S6, data processing for color evaluation (described in detail in FIG. 14) is performed. Then, the color evaluation result is displayed on the display 48 in step S7. In step S8, it is determined whether or not the colorimetry of the same type of inspection plate 1 has been completed. If the colorimetry of the same type of inspection plate 1 is still continued with NO, the process returns to step S5 to return to the next inspection plate 1. If the color measurement of the test plate 1 of that type is completed with YES,
Proceeding to step S9, it is determined whether or not to terminate the measurement itself. If the determination is NO, the color of the next type of test plate 1 is measured, the process returns to step S3, and if the measurement is terminated, the determination is YES. After finishing the measurement (step S10),
End the flow.

These controls are performed by the main controller 4
5, and each controller 46, 50, TC, SC, RC
And control data is exchanged between them, and timing control of color measurement is performed according to the time chart described with reference to FIG. FIG. 11 shows the “color measurement (S
4) is a flowchart showing the content of "4)." First, step S
At 100, it is determined whether the reference plate 1 has reached the measurement standby position P2. If NO, the control waits until the reference plate 1 reaches the measurement standby position P2. YES
If so, the entrance shutter 3I and the exit shutter 3E are closed in step S101.

Next, it is determined whether the reference plate 1 has reached the measurement point MP1 (step S102). N
If it is O, it waits until the reference plate 1 reaches the measurement point MP1. If YES, the flow advances to step S103 to instruct the CCD camera controller CC to perform measurement (1), and the CCD camera 20 measures the front portion of the reference plate 1
Perform (1). Next, it is determined whether the reference plate 1 has reached the measurement point MP2 (step S104). N
If it is O, it waits until the reference plate 1 reaches the measurement point MP2. If YES, the flow advances to step S105 to instruct the CCD camera controller CC to perform measurement (2), and the CCD camera 20 measures the center of the reference plate 1
Perform (2).

Next, it is determined whether the reference plate 1 has reached the measurement point MP3 (step S106). N
If it is O, it waits until the reference plate 1 reaches the measurement point MP3. If YES, the flow advances to step S107 to instruct the CCD camera controller CC to perform measurement (3), and the CCD camera 20 measures the rear portion of the reference plate 1
Perform (3). Next, in step S108, it is determined whether the reference plate 1 has reached the measurement completion position P3. If NO, the control waits until the reference plate 1 reaches the measurement completion position P3. If YES, the entrance shutter 3I and the exit shutter 3E are opened in step S109,
The reference plate 1 is carried out of the measurement booth 3, and the "color measurement of the reference plate 1 (S4)" ends.

"Color measurement of the inspection plate 1 (S5)" shown in FIG.
Also, the control operation may be performed according to the same flow as “color measurement of reference plate 1 (S4)”. Next, a control flow of the CCD camera controller CC will be described with reference to FIG.

In step S200, it is determined whether the system power has been turned on. If NO, the control waits until the power is turned on. If YES, initialization is performed in step S201. In this initial setting, normal imaging conditions, the CH0 position of the disk RP, and the like are also set. Next, step S
At 202, it is determined whether or not there is a calibration instruction from the main controller 45. At this time, information indicating whether the measurement target previously input as a key is a “statistical texture”, a “structural texture”, or an “intermediate texture” is also indicated. If NO and there is no calibration instruction, the process waits until the instruction is issued. If YES and there is a calibration instruction, the process proceeds to step S203. In step S203, the calibration data for the reference white plate 8 is sampled and recorded in the area 60 of the image memory 39.
In step S204, it is determined whether or not there is a measurement (1) to (3) instruction from the main controller 45. NO
If so, wait for an instruction from the main controller 45, and
If ES, the data of CH1 to CH8 is sampled in order in step S205,
Record at ~ 59. Next, in step S206, it is determined whether the measurement target is a “statistical texture”.
If YES, it is determined in step S207 whether there is a data transmission request from the image processing controller 46. If there is no request in NO, the process waits until there is a request for data transmission, and if YES, the process proceeds to step S208 to transmit print data to the image processing controller 46. In step S209, it is determined whether the recording data has been received. If NO, the process returns to step S208, and YE
If S, the process returns to step S204.

If NO in step S206 and it is determined that the object to be measured is not a "statistical texture", the flow advances to step S210 to determine whether the object to be measured is a "structural texture". In the case of YES, the process proceeds to step S211 and the FFT is performed on the data of CH0. Proceeding to step 212, the result of FFT execution is recorded, and the process returns to step S207.

If NO in step S210 and it is determined that the object to be measured is not the "structural texture" but the "intermediate texture", the flow advances to step S213 to store the recorded data of CH1 to CH8. When the data shows an abnormal value that is clearly different from the surface part, such as a groove part (composed of a slope and a bottom), the data in that part is deleted so as to be blank On the other hand, the original data (data before deletion) is also stored as it is, and the process returns to step S211.

FIG. 13 shows the image processing controller 4.
6 is a control flowchart of FIG. In step S300, it is determined whether the system power has been turned on. If NO
Wait until the power is turned on. If YES, initialization is performed in step S301. Next, step S302
It is determined whether there is an instruction from the main controller 45 for correction calculation on the standard light source 22. N
If O, skip to step S304. YE
If S, in step S303, the correction calculation is performed based on the fluctuation of the spectral energy distribution of the standard light source 22 detected by the standard light source controller 50, and the calculation is recorded. In step S304, the main controller 4
It is determined whether or not there is a color evaluation calculation instruction from 5.
If there is no instruction, that is, if NO, step S3
Return to 02. If there is an instruction, and if YES, the flow advances to step S305 to request the CCD camera 20 to transmit necessary data. In step S306, it is determined whether data reception is completed. If the reception has not been completed, and if NO, the process stands by until the data reception is completed. If YES, the process advances to step S307 to perform a color evaluation calculation. Next, in step S308, it is determined whether the calculation is completed. If the calculation has not been completed, the process returns to step S307. If the calculation is completed with YES, the process proceeds to step S309, and the calculation result is transmitted to the main controller 45. Next, in step S310, it is determined whether the calculation result has been received by the main controller 45. If no calculation result has been received, ie N
If O, the process returns to step S309; if YES, the process returns to step S302.

FIG. 14 is a flowchart of the color evaluation calculation. S
At 400, the type of texture is determined. If the texture is determined to be “statistical texture”, the recording data of CH1 to CH8 is subjected to α / S conversion in step S401 according to equation (2). In step S402, a density (α / S) histogram of CH1 to CH8 is created. In step S403, the average value μ, variance σ ² , skewness D, and kurtosis K are calculated for CH1 to CH8. In step S404, comparative evaluation with the reference plate 1 data is performed. Next, in step S405, CH
A multiplex spectral image is obtained by calculation from the recording data of α1 to CH8 before α / S conversion, and the process returns to step S308 in FIG.

If the type of the texture is “structural texture”, in step S406, C
A frequency distribution is obtained for the F0 data of H0. Next, in step S407, comparative evaluation with the reference plate 1 data is performed. Next, in step S408, CH1 to CH1
A multiplex spectral image is obtained by calculation from the recording data of CH8, and the process returns to step S308 in FIG.

If the type of texture is "intermediate texture", the recording data of CH1 to CH8 from which abnormal values have been deleted in step S409 are converted to α /
Perform S conversion. Next, in step S410, the occupation ratio of the blank data pixel portion to all pixels constituting one screen is calculated. In step S411, the C whose abnormal value has been deleted
Steps S402 to S404 are executed for H1 to CH8 (Evaluation 1). Next, in step S412, it is determined whether the occupancy ratio is smaller than a predetermined value. If smaller, that is, if YES, step S
At 413, the recorded data of CH1 to CH8 (the original data stored before the data indicating the abnormal value is deleted, and the occupation ratio of the corresponding portion is small, so its influence on the whole needs to be particularly considered. Multiplex spectral image is calculated by calculation, and the process returns to step S308 in FIG. If “NO” in the step S412, the steps S402 to S404 are performed for each of the pixels CH1 to CH8 for the pixel having the abnormal value in the step S414.
(Evaluation 2). In step S415, steps S406 to S407 are executed, a frequency distribution is obtained for the FFT data, and a comparison evaluation with the reference plate 1 data is performed.

Although the preferred embodiment of the present invention has been described above, the present invention is not limited to this embodiment, and various improvements and designs of the present invention can be made without departing from the gist of the present invention. Changes are possible. For example,
The CCD camera may capture moving images,
It is also possible to change the number to two or more or to determine the positioning of the installation in accordance with the properties and characteristics of the plate-like body.

Further, the measurement positions of the plate-like body are the front part, the center part and the rear part, but the measurement positions can be appropriately changed depending on the surface shape and size. Further, in this specification, the term “color measurement” means not only measurement of hue but also measurement of lightness. Therefore, it is assumed that the color signals from the electronic camera include those capable of recognizing brightness.

[0052]

As described above, according to the present invention, the plate detection system measures the color of a plate-like body, and includes an electronic camera for imaging the plate-like body, and a spectral reflection from the electronic camera. Since the signal processing means for processing the rate data is provided, the colorimetric evaluation can be performed more reliably. In addition, the present invention can be applied particularly when the object to be measured has a plate-like body having various uneven shapes, and highly accurate colorimetric evaluation can be performed without stopping the production line.

[Brief description of the drawings]

FIG. 1 is an overall side view of an embodiment of the present invention.

FIGS. 2A and 2B are views for explaining a measurement unit of the present invention, wherein FIG. 2A is a front view and FIG. 2B is a perspective view.

FIG. 3 is a configuration diagram of a CCD camera of the present invention.

FIG. 4 is a front view of a rotating disk according to the present invention.

FIG. 5 is a time chart showing the operation of each part of the present invention.

FIG. 6 is a block diagram showing a basic configuration of the present invention.

FIG. 7 is a diagram showing a recording data configuration in the image memory of the present invention.

FIG. 8 is a view showing characteristics of the surface shape of a building board.

FIG. 9 is a diagram showing an optical Fourier transform image.

FIG. 10 is a flowchart showing the control operation of the main controller of the present invention.

FIG. 11 is a flowchart showing a colorimetric control operation of the reference plate of the present invention.

FIG. 12 is a flowchart showing a control operation of the CCD camera controller of the present invention.

FIG. 13 is a flowchart showing a control operation of the image processing controller.

FIG. 14 is a flowchart showing a control operation of color evaluation calculation.

FIG. 15 is a configuration diagram of a conventional apparatus for measuring spectral reflectance.
1A is a configuration diagram of an apparatus using a diffraction grating, and FIG. 2B is a configuration diagram of an apparatus using a prism.

[Explanation of symbols]

B Measurement box 1 Building board (reference board, inspection board) 2 Loading conveyor 3 Measurement booth 3I Inlet shutter 3E Exit shutter 4 First transport conveyor 5 Third transport conveyor 6 Unloading conveyor 7 Table lifter 8 Reference white board 9 White board mounting Table 10 Measurement unit 11 Second transport conveyor 20 CCD camera 21 Conversion filter 22 Standard light source 23 Diffusion prevention cover 24 Board guide 33 Imaging lens 34 Optical low-pass filter 35 CCD imaging device array matrix 36 CCD drive circuit 39 Image memory 40 Input / output interface 45 Main controller 46 Image processing controller 47 Keyboard 48 Display 49 Printer 51, 52, 53, 54, 55, 56, 57, 58, 5
9,60,61,62,63 Sampling data storage area

────────────────────────────────────────────────── ───

[Procedure amendment]

[Submission date] March 31, 1999

[Procedure amendment 1]

[Document name to be amended] Statement

[Correction target item name] Claims

[Correction method] Change

[Correction contents]

[Claims]

2. The image processing means for obtaining a signal corresponding to a ratio between an absorption coefficient and a scattering coefficient of the building board from the color signal or a signal derived from the color signal. 1. The inspection plate system according to 1 .

Wherein said color electronic camera according to claim 1 or 2 test plate system wherein a is for capturing a still image.

[Procedure amendment 2]

[Document name to be amended] Statement

[Correction target item name] 0001

[Correction method] Change

[Correction contents]

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a board inspection system for inspecting building boards , and more particularly, to provide a board inspection system capable of inspecting appearance quality on a line.

[Procedure amendment 3]

[Document name to be amended] Statement

[Correction target item name] 0009

[Correction method] Change

[Correction contents]

[0009]

According to the present invention, there is provided a board inspection system for inspecting a building board, comprising: a color electronic camera for imaging the building board; and a luminance signal or the luminance signal from the color electronic camera. Signal derived from
Rier transforms the color signal from the color electronic camera or
Image processing means for obtaining an average value and a variance of a signal derived from a color recording signal; and a result of the Fourier transform, an average value and a variance.
And display means for displaying

[Procedure amendment 4]

[Document name to be amended] Statement

[Correction target item name] 0010

[Correction method] Change

[Correction contents]

Further, the image processing means that is intended to determine the signal corresponding to the ratio of the absorption coefficient and scattering coefficient of the plate-shaped body from a signal derived from the color signal or the color signal, the human visual The color of the plate-like body can be evaluated on a scale suitable for the following. Further , since the color electronic camera captures a still image, it is not necessary to capture an unnecessary image, so that the bandwidth can be widened by that much, and highly accurate spectral reflectance data can be obtained. be able to.

[Procedure amendment 5]

[Document name to be amended] Statement

[Correction target item name] 0052

[Correction method] Change

[Correction contents]

[0052]

As described above, according to the present invention, the measurement object
The luminance signal of the building board or the signal derived from the luminance signal
The frequency distribution of the building board is detected by the Fourier transform signal of
And display them, and use the results to determine texture features.
Can be This gives the density histogram
Utilizing spatial information that is lost, especially large irregularities
When a building board with a critical surface shape is to be measured
Allows correct evaluation of the image. And architecture
Average value and minute value of the color signal of the board or the signal derived from the color signal
The color of the building board can be objectively evaluated from
You.

Claims (5)

[Claims] In a plate inspection system for inspecting a plate-shaped object, a color electronic camera for imaging the plate-shaped object, and different image processing for inspecting a luminance signal and a color signal from the color electronic camera for inspection. And an image processing means for performing the following. 2. The inspection board system according to claim 1, wherein said image processing means performs Fourier transform on said luminance signal or a signal derived from said luminance signal. 3. The image processing means according to claim 1, wherein a signal corresponding to a ratio between an absorption coefficient and a scattering coefficient of the plate is obtained from the color signal or a signal derived from the color signal. Item 3. The inspection plate system according to Item 1 or 2. 4. The inspection board system according to claim 1, wherein said image processing means calculates an average value and a variance of said color signal or a signal derived from said color signal. 5. The plate detection system according to claim 1, wherein the color electronic camera is configured to capture a still image.