JPH0712538A - Selection method of image sensing device in image processing apparatus - Google Patents

Abstract

PURPOSE:To automate the selection of an image sensing device, to reduce the number of selection processes, to eliminate a selection error and to quickly and precisely process an image by a method wherein an image sensing device which includes an image sensing region inside its image sensing field-of-view region is selected by an image sensing device corresponding to the image sensing region of an object to be measured. CONSTITUTION:An image sensing device 13 is installed on a stand, and it takes in an image in every measuring part on a sheetlike body 1 positioned by a sheetlike-body positioning device 12. A total of 21 devices corresponding to measuring regions is installed on the stand. An image processing apparatus 14 images a region including a measuring point on the sheetlike body 1, it processes an image, and it computes the size of every measuring point. At this time, the apparatus 14 selects and uses the device 13 corresponding to an image sensing region on the sheetlike body 1 from the total of 21 devices 13. A control device transfers, to the apparatus 14, information on the size measurement of the sheetlike body 1 carried in by a sheetlike-body conveyance device 11. Which image sensing device out of the devices 13 is used is selected automatically by the apparatus 14 on the basis of information on a type transferred by the control device.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for selecting an image pickup device in an image processing device.

[0002]

2. Description of the Related Art In a plate-like product manufacturing line for wall panels or the like, the size of a measurement point defined on the plate-like member is measured and a pass / fail judgment is made with respect to a standard size.

Incidentally, Japanese Patent Laid-Open No. 2-77607 proposes a shape dimension measuring method using an image processing technique. In this conventional technique, a measurement point of an object to be measured is imaged by an imaging device, and the dimension of the object to be measured can be measured by binarizing the image signal.

Therefore, it is conceivable to measure the size of a plate-like body such as a wall panel by using the above-mentioned image processing technique. At this time, the overall size of the plate-like body is far beyond the imaging visual field range of one image pickup device. Therefore, as the image pickup device, (a) the image pickup device is movably used, or (b) ) It is used in the form of an imaging device group in which a large number of imaging devices are fixedly arranged. In order to speed up the measurement speed of a continuous processing line for plate-like bodies, it is preferable to use the large number of image pickup devices described in (b) above.

In the image processing apparatus that uses a large number of image pickup devices as described above, the objects to be measured of various types are sequentially positioned at the measurement positions in the measurement area, and the measurement points of the objects to be measured are set. When performing image processing by picking up an image pickup area including the image pickup area, an image pickup apparatus corresponding to the image pickup area of the object to be measured is selected and used from a large number of image pickup apparatuses installed corresponding to the measurement area. Become.

[0006]

However, when selecting an image pickup device corresponding to the measurement point of the object to be measured this time from among a large number of image pickup devices, the image pickup device must be manually selected. Can be considered.

However, when the plate-like object as the object to be measured has many kinds and has many measurement points, the manual selection time of the image pickup device becomes long, and there is a possibility that an improper image pickup device is erroneously selected due to a selection error. There is.

It is an object of the present invention to automate the selection of an image pickup device in an image processing device using a large number of image pickup devices, reduce the number of selection steps, and eliminate the selection error to achieve quick and accurate image processing. .

[0009]

According to a first aspect of the present invention, each of various types of objects to be measured is sequentially positioned at a measurement position within the measurement area, and an imaging area including the measurement points of each object is formed. An image pickup apparatus in an image processing apparatus that selects and uses an image pickup apparatus corresponding to the image pickup area of the object to be measured from a large number of image pickup apparatuses installed corresponding to the measurement area when the image is picked up and image processed. In the selection method, while predetermining the imaging visual field range in the measurement area of each imaging device, the imaging area of the measured object positioned at the measurement position in the measurement area is obtained based on the type information of the measured object, An image pickup device including the image pickup region of the object to be measured within its image pickup field range is selected as an image pickup device corresponding to the image pickup region of the object to be measured.

According to a second aspect of the present invention, in addition to the first aspect of the present invention, the object to be measured is a patterned plate-like object, and the plate-like object is based on the type information of the plate-like object. The image pickup area for measuring the ear size, the total height, the total length, the diagonal length, etc. is obtained.

[0011]

With the imaging visual field range within the measurement area of each imaging device being determined in advance, the imaging area of the measured object positioned in the measurement area is determined based on the type information of the measured object, An image pickup device including the image pickup region within the image pickup field range can be automatically selected as an image pickup device corresponding to the image pickup region of the object to be measured. Therefore, as compared with manual selection, it is possible to reduce the number of selection steps, eliminate selection mistakes, and achieve quick and accurate image processing.

When the object to be measured is a patterned plate-like object, the ear size, the total height of the plate-like object,
It is possible to obtain an imaging region for measuring the total length, diagonal length, etc., and automatically select the imaging device corresponding to the imaging region as described above.

[0013]

EXAMPLE FIG. 1 is a schematic diagram showing a dimension measuring system, FIG.
Is a schematic diagram showing a plate-like body manufacturing line, FIG. 3 is a schematic diagram showing a control system of a dimension measuring system, FIG. 4 is a schematic diagram showing dimension measuring portions of the plate-like body, FIG. 5 is a schematic diagram showing an image pickup device, Figure 6
7 is a schematic diagram showing an image processing procedure, FIG. 7 is a schematic diagram showing a shadow formation state, FIG. 8 is a schematic diagram showing a defective portion measurement state, FIG. 9 is a flowchart showing an imaging region determination procedure, and FIG. 10 is an imaging device determination procedure. 2 is a flowchart showing

The dimension measuring system 10 measures the dimensions of a plurality (16 places) of measuring portions defined on the plate-like body 1 by image processing.

The plate-like body 1 is a wall panel manufactured through a dryer 3 and a trimming device 4 on a molding line in a wall panel manufacturing line as shown in FIG. Then, the plate-shaped body 1 is provided with the dimension measurement system 10 provided in the inspection line 5.
The size is measured by and the product is delivered to the sorting line 6.

As shown in FIG. 4, the plate-shaped body 1 has a convex portion 1A at the center and ears 1B at the periphery, and the total height H ₁ , H ₂ on the left and right, and the total length B ₁ , B _{2 on the} upper and lower sides. , Diagonal length C ₁ , C
_2. All 16 measurement sites of ear dimensions a _{1 to} a ₆ and b _{1 to} b ₄ are defined.

The dimension measuring system 10 is capable of measuring the dimensions of the plate-shaped bodies 1 of all 500 types. In the plate-shaped body 1, the standard dimensions of the above 16 measurement sites are different for each product type.

Therefore, the dimension measuring system 10 is shown in FIG.
As shown in FIG. 3, it is configured to include a plate-shaped body conveying device 11, a plate-shaped body positioning device 12, an image pickup device 13, an image processing device 14, and a control device 16.

The plate-shaped member conveying device 11 is a belt conveyor device driven by a motor 17, and sequentially carries the plate-shaped member 1 into and out of the measurement area.

The plate-shaped body positioning device 12 comprises a W-direction positioning device 12A and an L-direction positioning device 12B.
Both positioning devices 12A and 12B can be moved up and down on the transport surface of the plate-shaped transport device 11. The W direction positioning device 12A is moved in the W direction (short direction) by the pusher 18.
The plate-shaped body 1 is pushed in the W direction by the pressing portion 19 made of the urethane roller which is linearly moved to press the plate-shaped body 1 against the urethane-roller stopper 20 to position the plate-shaped body 1 in the W direction. To do. The L-direction positioning device 12B positions the plate-shaped body 1 in the L-direction by sandwiching the plate-shaped body 1 by a pair of right and left urethane roller sandwiching portions 22 which are opposed to each other in the L-direction (longitudinal direction) by the pusher 21. That is, the plate-shaped body positioning device 12 positions the plate-shaped body 1 carried into the measurement area by the plate-shaped body transport device 11 at the measurement position.

In FIG. 1 (B), W0 is the W-direction reference position (stopper surface), L0 is the L-direction reference position (center line), and O is the origin of the measurement area.

The image pickup device 13 is installed on the pedestal 23 and captures images of each measurement site of the plate-shaped body 1 positioned by the plate-shaped body positioning device 12. A total of 21 image pickup devices 13 corresponding to the measurement area are installed on the gantry 23. A CCD camera is used as the image pickup device 13. The image pickup device 13 includes a dustproof case 24, a power supply 25, and an illumination 31, as shown in FIG.

The image processing device 14 picks up an image of the image pickup area including the measurement points of the plate-shaped body 1 and performs image processing to calculate the dimension of each measurement point. At this time, the image processing device 14 selects and uses the imaging device 13 corresponding to the imaging region of the plate 1 from all 21 imaging devices 13 installed corresponding to the measurement region.

The control device 16 is a host computer, and transmits the size measurement information (product type, standard size, etc.) of the plate-shaped body 1 carried in by the plate-shaped body conveying device 11 to the image processing device 14.

The image processing device 14 determines whether the calculated size of each measurement site is acceptable or not, based on the size measurement information of the plate-shaped body 1 transmitted from the control device 16.

The dimension measuring operation of the plate-shaped body 1 by the dimension measuring system 10 is performed as follows.

(1) The plate-like body 1 cut into a predetermined size by the trimming device 4 is carried in by the plate-like body conveying device 11. The trimming device 4 is controlled by the trimming control device 4A to which the cutting information is transmitted from the control device 16.

(2) Plate-like body 1 carried into the dimension measuring position
Is the W-direction positioning device 12 of the plate-shaped body positioning device 12.
Positioning is performed by the A and L direction positioning device 12B.

(3) The image of the measurement site of the plate 1 is captured by the image pickup device 13. Which of the image pickup devices 13 is used among all the image pickup devices 13 is automatically selected by the image processing device 14 based on the type information transmitted by the control device 16 as described later.

(4) Image processing is performed by the image processing device 14,
Calculate the dimensions of each measurement site.

(5) Plate-shaped body 1 transmitted from the control device 16
On the basis of the dimension measurement information, the calculated dimensions of each measurement site are compared with their standard dimensions to make a pass / fail judgment.

The image processing procedure by the image processing device 14 will be described below. As an example of image processing, the plate-shaped body 1
A case of obtaining the ear size of will be described.

(1) Raw image capture (FIG. 6 (A),
(B), see FIG. 7) Ear portion 1B defined at the stepped boundary portion with the convex portion 1A of the plate-shaped body 1.
The dimension a of is calculated. An illumination (high-frequency fluorescent lamp) 31 is applied from diagonally above the convex portion 1A constituting the step-shaped boundary of the plate-shaped body 1, and an imaging region including a shadow 32 formed by this illumination 31 (including the ear portion 1B which is a measurement point). The raw image of the imaging area) is captured by the imaging device 13.

The shadow 32 can be made linear by adjusting the installation state of the illumination 31.

When the shadow 32 is planar, the width dimension a _w of the shadow 32 is previously obtained. Then, the image processing device 14 adds a _w as a correction value to the dimension a ₀ obtained from the number of images of the ear 1B by the image processing described later.

(2) Search for Binarization Level (See FIG. 6C) The threshold value is determined from the density difference of the shadow 32 generated at the stepped boundary portion of the raw image picked up by the image pickup device 13. Specifically, it is as follows.

Creation of Grayscale Histogram The raw image captured in the image pickup device 13 in the above (1) is cut out with a constant width (Y) and divided into 512 quadrangles (unit: pixel).
Create a density histogram (density level: 256 levels)
(See FIG. 6C).

Determination of Threshold Value From the above histogram, the level for image binarization (distinguishing between white and black) is determined by the Yamatani method (binarization level = threshold value).

In (2), since the concentration distribution is checked for all the objects to be measured, it is possible to flexibly cope with the change in illumination illuminance.

(3) Optimal binarization (see FIG. 6 (D)) The raw image captured by the image pickup device 13 in (1) above is
Binarization processing is performed with the threshold value obtained in (2).

The measurement target image (white) 33 including the measurement site (ear 1B) divided by the binarization process is obtained, and the area per pixel which is predetermined in the number of pixels of the measurement target image 33. The area (A ₁ ) of the measurement target image 33 is calculated by multiplying
From the contour line shape of the area of the measurement target image 33 (A ₂ )
Is calculated, and the above-mentioned threshold value is changed until A ₁ and A ₂ substantially match to obtain the optimum threshold value.

Here, the above is specifically described in the following (a)
Or as in (b). (a) When the inclination θ of the main axis (short axis and long axis) of the measurement target image 33 is obtained by the moment of inertia theorem

For the image 33 to be measured, X is passed through the center of gravity.
A coordinate system with the center of gravity as the origin, by obtaining the moment of inertia Mxx about a straight line parallel to the axis, the moment of inertia Myy about a straight line passing through the center of gravity and parallel to the Y axis, and the moment of inertia Mxy about a straight line passing through the center of gravity and parallel to the X and Y axes The crossing angle (tilt) θ between the X-axis and the long axis of the inertia ellipse equivalent to the measurement target image 33 is obtained.

[0044]

[Equation 1]

XY coordinates having the above-mentioned tilt direction as the principal axis direction are set in the measurement target image 33, and A ₂ = XY is obtained from the length X of the short axis and the length Y of the long axis. Then, the above threshold value is changed by 5 (in 256 steps) until 0.9 × A ₂ ≦ A ₁ ≦ 1.1 × A ₂ (5) (the maximum number of changes is 20 times).

(B) When the main axis of the measurement target image 33 has no inclination: A ₂ = XY is obtained from the maximum length X of the measurement target image 33 and the cutout width Y of the raw image. And

The threshold value is changed by 5 (in 256 steps) until A ₁ ≉A ₂ (6) (maximum change number is 20 times).

In this (3), the threshold changing time for obtaining the optimum binarization level is changed 20 times.
It is about a second.

(4) Calculation of the ear size (see FIG. 6 (E)) The above-mentioned raw image is set to 2 with the optimum threshold value obtained in (3) above.
Quantize.

The number of pixels of the rectangular portion 34 having one side of the measurement site (ear 1B) calculated by the binarization process is measured.

The area (A) of the rectangular portion 34 is calculated by multiplying the number of pixels of the rectangular portion 34 by a predetermined area per pixel, and the area (A) is calculated as the cutout width of the rectangular portion 34. By dividing by Y, the dimension a _{0 of the} measurement site is measured.
Is calculated (see FIG. 8).

As a result, the measuring portion (ear 1B) is captured by the rectangular portion 34, so that the missing portion 35 as shown in FIG.
It is possible to prevent the appearance of abnormal dimensions due to the presence of the like.

When the shadow 32 of the raw image has the width dimension a _w , as described above, the value obtained by adding a _w to the a ₀ as a correction value is used as the measurement portion (ear 1B). (X = a).

A method of selecting the image pickup device 13 corresponding to the image pickup area to be image-processed in the image processing device 14 will be described below (FIGS. 9 and 10). It should be noted that an example of the imaging area including the ear portion 1B of the plate-shaped body 1 will be described.

(1) Determination of the imaging visual field range of the imaging device 13 The imaging visual field range within the measurement area of each imaging device 13 is determined in advance. It is as shown in Table 1.

[0056]

[Table 1]

(2) Determination of imaging area (FIG. 9) The imaging area of the plate 1 positioned at the measurement position in the measurement area by the plate positioning device 12 is obtained based on the product type information of the plate 1. . The procedure for determining the imaging region for the measurement of the ear size on the a side will be described below.

The plate-like body 1 designated as the image processing target
Read the product type information (master beta). It is the L dimension, the W dimension, each ear dimension (a, b), and their respective dimension allowable values shown in FIG. 4.

The maximum value of the ear size is calculated. aMmax = aM + aM allowable value bMmax = bM + bM allowable value Note that M is the ear size, aMmax is the maximum ear size on the a side, bM
max is the maximum value of the ear size on the b side.

Necessary condition values R and S for determining the imaging area
To calculate. R = bMmax + α S = aMmax + β R and α are values for removing the b-side ear from the imaging region. S and β are values for slightly including the convex portion 1A in the imaging region.

The position of the edge of the plate-shaped body 1 can be calculated. Lmax = L + L allowable value Wmax = W + W allowable value Note that Lmax is the maximum value in the L direction and Wmax is the maximum value in the W direction.

The absolute coordinates of the starting point A (x, y) of the imaging area are obtained from the origin O of the measurement area. A (x, y) = (Lmax-R, Wmax + γ)

Γ is a background portion for making the edge of the plate-shaped body 1 easy to see.

The absolute coordinates of the end point B (x, y) of the image pickup area are obtained from the origin O of the measurement area. B (x, y) = (Lmax- (R + Y), Wmax-S)

An image pickup device in which the points A and B coexist is selected. Specifically, it is based on (3) below.

(3) Determining the image pickup device 13 (FIG. 10) The image pickup device 13 including the start point A and the end point B of the image pickup area of the plate-like body 1 within the image pickup visual field is picked up for the image pickup of the platelike body 1. The image pickup device 13 corresponding to the region is selected. Specifically, it is as follows.

The No. of the image pickup device 13 is selected from No. 1, and the image pickup visual field range of the image pickup device 13 is read from the data shown in Table 1. It is determined whether or not point A exists in the read imaging visual field range.

By the above, the above is repeated until the image pickup device 13 in which the point A exists in the image pickup field range is found, and the image pickup device 13 corresponding to the point A is determined.

It is determined whether or not point B exists in the image pickup visual field range of the image pickup device 13 that corresponds to point A as described above, and if the determination result is Yes, the image pickup device 13 is set as the image pickup region of this time. It is determined as the imaging device 13 to be operated. N
If o, the two image pickup devices 13 are selected and used.

The operation of this embodiment will be described below. The plate-shaped body 1 positioned in the measurement region while predetermining the imaging visual field range in the measurement region of each imaging device 13
The imaging area of is obtained based on the type information of the plate 1,
The image pickup device 13 including the image pickup region of the plate-shaped body 1 within the image pickup field range can be automatically selected as the image pickup device 13 corresponding to the image pickup region of the plate-shaped body 1. Therefore, as compared with manual selection, it is possible to reduce the number of selection steps, eliminate selection mistakes, and achieve quick and accurate image processing.

When the plate-shaped body 1 is a patterned plate-shaped body, an imaging area for measuring the ear size, total height, total length, diagonal length, etc. of the plate-shaped body 1 based on the product type information of the plate-shaped body 1. And the image pickup device 13 that should correspond to this can be automatically selected as described above.

The embodiment of the present invention has been described in detail above with reference to the drawings. However, the specific structure of the present invention is not limited to this embodiment, and changes in design within the scope not departing from the gist of the present invention can be made. Even if it exists, it is included in the present invention. For example, the present invention
Not only when the image pickup device 13 corresponding to the image pickup region for measuring the ear size of the plate-shaped body 1 is selected and used, but the total height of the plate-shaped body 1 is
It can be widely applied to select and use the image pickup device 13 corresponding to the image pickup region for measuring the total length, the diagonal length and the like.

[0073]

As described above, according to the present invention, in an image processing apparatus using a large number of image pickup apparatuses, the selection of the image pickup apparatus can be automated, the selection process can be reduced, and selection mistakes can be eliminated to speed up the image processing. Can be accurate.

[Brief description of drawings]

FIG. 1 is a schematic diagram showing a dimension measuring system.

FIG. 2 is a schematic view showing a plate-like body production line.

FIG. 3 is a schematic diagram showing a control system of the dimension measuring system.

FIG. 4 is a schematic diagram showing a dimension measurement site of a plate-like body.

FIG. 5 is a schematic diagram showing an imaging device.

FIG. 6 is a schematic diagram showing an image processing procedure.

FIG. 7 is a schematic diagram showing a shadow formation state.

FIG. 8 is a schematic diagram showing a measurement state of a defect portion.

FIG. 9 is a flowchart showing a procedure for determining an imaging area.

FIG. 10 is a flowchart showing a procedure for determining an imaging device.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Plate-like object (object to be measured) 10 Dimension measuring system 11 Plate-like body conveying device 12 Plate-like body positioning device 13 Imaging device 14 Image processing device 16 Control device

Claims (2)

[Claims] 1. Each of various types of measured objects is sequentially positioned at a measurement position within a measurement area, and when an image pickup area including a measurement point of each measured object is imaged and image-processed, the measurement area is corresponded to the measurement area. In an image pickup device selection method in an image processing device that selects and uses an image pickup device corresponding to the image pickup region of the object to be measured from a large number of installed image pickup devices, an image pickup is performed in the measurement region of each image pickup device. While predetermining the visual field range, the imaging area of the measured object positioned at the measurement position in the measurement area is obtained based on the product type information of the measured object, and the imaging area of the measured object is within the imaging visual field range. An image pickup apparatus selecting method in an image processing apparatus, comprising selecting an image pickup apparatus including the image pickup apparatus as an image pickup apparatus corresponding to the image pickup area of the object to be measured. 2. The object to be measured is a patterned plate-shaped body, and based on the type information of the plate-shaped body, the ear size, the total height of the plate-shaped body,
The imaging area for measuring the total length, the diagonal length, etc. is obtained.
A method for selecting an image pickup apparatus in the image processing apparatus described.