JP5350082B2 - Accuracy determination device for shape measuring device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a method, apparatus and reference jig for discriminating measurement precision of a shape measuring instrument, for facilitating confirmation of the precision with a small amount of man-hours. <P>SOLUTION: For precision determination of imaging an object to be measured positioned in a visual field of a fixed breadth by irradiating the object with linear light 4 from linear light generation means 3, the apparatus includes jig drive means 30 retaining and driving a reference jig 10 so that the reference jig 10 formed with recesses and protrusions 10j continuous to position an irradiating surface irradiated with linear light 4 along the extension direction of the linear light 4 is positioned at a target position and a holding means 40 for holding a linear light generating means 3 so that the irradiating surface of the reference jig 10 is irradiated with linear light 4. With this configuration, the precision of the shape measuring instrument is discriminated by imaging the recesses and protrusions 10j irradiated with the linear light 4 with imaging means 5, monitor-displaying the recesses and protrusions 10j imaged by display means 7 on a conversion table 6f as an image cross-sectional figure converted into actual dimensions, and comparing the displayed image cross-sectional figure and an actual recesses and protrusions figure to each other. <P>COPYRIGHT: (C)2011,JPO&amp;INPIT

Description

The present invention relates to a precision determination equipment you determine the accuracy of the shape measurement device.

Conventionally, in order to check whether a die or the like (measurement object) formed by precision machining has been machined according to the specified dimensions, the shape of the machined object is measured by performing three-dimensional shape measurement etc. doing. For this three-dimensional shape measurement, for example, an optical cutting method is widely used.
The light cutting method irradiates a measurement target existing within a predetermined imaging field of view with slit light and images the measured object with an imaging means, obtains an image cross-sectional figure on the measured object, and analyzes the image This is a method for acquiring information on the three-dimensional shape of the object to be measured.

That is, as shown in FIG. 15, the shape measuring apparatus generates laser light as linear light generating means on a measured object 2 such as a tire mold positioned in a camera field of view (virtual) 1 having a certain width. A slit light 4 as linear light extending up and down is irradiated from the device 3, and the reflected light of the slit light 4 is received by a CCD camera 5 as an imaging means, and the received light signal is a computer 6 as a control means. The computer 6 causes the monitor 7 serving as a display means to display the cross-sectional real figure 8 corresponding to the portion irradiated with the slit light 4 of the measurement object 2 as an image cross-sectional figure 8 '. is there.
The image cross-sectional figure 8 'is obtained by converting the captured cross-sectional real figure 8 into an actual size by a conversion table provided in advance in the computer 6, and as an image in the camera image field 1' of the monitor 7. Is displayed. Whether or not the detailed dimensions of the display content of the image cross-sectional figure 8 ′ are displayed with high accuracy is an important issue in, for example, molding a high-quality tire, and this accuracy error should be confirmed. Is important.
That is, the measurement object 2 placed at a predetermined position in the camera visual field 1 needs to be accurately displayed on the monitor 7 as an image cross-sectional figure 8 '(maintaining a one-to-one relationship).
However, a one-to-one relationship is not necessarily maintained at each of the positions 1a, 1b, and 1c due to an error in the lens system of the CCD camera 5, a processing error in the image processing system captured by the lens system, or the influence of the imaging position of the camera. The actual cross section 8 is not always displayed as the image cross section 8 'on the output screen. For this reason, a method for discriminating an error between the cross-sectional real figure 8 and the image cross-sectional figure 8 'is required.
As a first method of determining accuracy, in FIGS. 16 and 17, a reference jig 9a having a predetermined size is installed at each of the positions 1a, 1b, and 1c at least above, in the center, and below the field of view 1 of the camera. The reference jig 9 a is irradiated with the slit light 4 from the laser light generator 3, the reflected light is picked up by the CCD camera 5 and processed by the computer 6, and the cross-section actual figure 9 of the reference jig 9 a is displayed on the monitor 7. It is displayed as an image cross-sectional graphic 9 ', and this image cross-sectional graphic 9' is compared with the actual dimension of the cross-sectional actual graphic 9 of the reference jig 9a, and the accuracy is closely examined. The reference jig 9a is a rectangular parallelepiped made of a mold in which the vertical dimension a and the depth dimension b are accurately set. As an example, the reference jig 9a is molded with high accuracy to a dimension of 100 mm and a dimension of b to 50 mm. Yes.
The reference jig 9 a is processed by the computer 6 through the CCD camera 5 and displayed on the monitor 7. The image cross-sectional figure 9 ′ has a dimensional error relative to the cross-sectional real figure 9 of the reference jig 9 a. By checking whether or not, an accuracy error occurring in the CCD camera 5 can be determined. That is, the image cross-sectional figure 9 'displays the result processed through the CCD camera 5 and the computer 6, and the vertical dimension a' and depth dimension b 'of the image cross-sectional figure 9' as a display result are displayed. By checking whether the vertical dimension a and the depth dimension b of the reference jig 9a shown in FIG. 16 are within the allowable range, the reference jig 9a at each position 1a, 1b, 1c is checked. It is possible to know whether or not the image is accurately processed. If it is out of the allowable range, the accuracy error is set to an allowable value by correcting the lens system and the image processing system.
For example, when the reference jig 9a is arranged at the central position 1b, it can be seen that an error has occurred in the imaging system of the CCD camera 5, so that this error is corrected by correcting the lens system or the camera position. Thus, a camera that does not cause an error is obtained.
However, in the conventional method, the reference jig 9a covers each of the positions 1a, 1b, and 1c, but at each position according to the survey positions of the upper and lower positions 1a, 1c, and the center position 1b. The work which has to relocate the reference jig 9a is required.
Further, in this conventional method, for example, when it is desired to image and determine a range of a smaller area of each position 1a, 1b, 1c with the CCD camera 5, as shown in FIG. A small reference jig 9b (having known dimensions a and b) is prepared separately, arranged so as to cover only a small area, and captured by the CCD camera 5 and displayed on the monitor 7 as an image cross-sectional figure. Thus, this image sectional figure is compared with the actual size of the reference jig 9b.

Therefore, in the first method, in addition to the above-described troublesome work, a plurality of reference jigs having different dimensions are prepared in accordance with the size of the measurement position. In addition, a plurality of reference jigs must be replaced and placed in front of the CCD camera 5, and the measurement is troublesome as described above.
Further, in the conventional method, the accuracy can only be discussed within a certain range of the field of view captured by the CCD camera 5. That is, only the accuracy in the left and right center and the vicinity of the top and bottom center in the field of view can be obtained, and the accuracy may be different for each place of the field of view. For this reason, in order to calibrate the accuracy of the entire visual field, it is necessary to perform the calibration by moving the reference jig within the visual field. However, even if it tries to obtain the accuracy of the edge of the field of view, it cannot be compared with the actual size because it protrudes from the field of view with a large standard jig and the horizontal dimension of the standard jig is not measured. .
Further, in the case of an optical (laser irradiation type) shape measurement sensor, the accuracy of the side surface side of an object that is difficult for the laser to hit deteriorates, and the accuracy of accuracy evaluation decreases. That is, the front side (dimension a) of the reference jig 9a is correctly irradiated because the laser is well irradiated, but measurement is possible because the side surface side (dimension b) of the reference jig 9a is not easily irradiated with laser. Accuracy will deteriorate.
In such an accuracy evaluation, it is desirable to perform multiple measurements in consideration of measurement variations, but since the same reference jig is measured, the same result can be obtained even if multiple measurements are simply performed. Is highly likely to be measured, and measurement variations that occur when measuring the actual measurement object 2 are difficult to reproduce. Thus, there is a method of performing measurement again by re-installing the reference jig, but the operation becomes complicated.
Conventionally, when the arrangement of the reference jig is changed, the positional relationship between the reference jig and the camera changes, so the size of the image cross-sectional figure 9 'changes, and comparison with the actual cross-sectional figure 9 is easy. There was a drawback that could not be done.

As a second method, as shown in Patent Document 1, the plane is positioned so as to protrude from the field of the sensor that measures the target as a calibration reference jig formed in a step shape so that the height is different. Then, irradiate each plane of this target with a light pattern, measure the distance between the light patterns by scanning with the sensor, and measure the height from the sensor to each plane of the target. A method to quickly calibrate is published.
However, according to the second method, since the plane formed so that the target used for calibration has a predetermined height is formed in a step shape, for example, when the size of the measurement object changes, Since it is necessary to create a target corresponding to the size of the measurement target and perform measurement, it is necessary to check the measurement accuracy every time the target is changed, which is troublesome as in the first method. .

JP 2008-170279 A

In this invention, provided in order to solve the above problems, the accuracy determination equipment that can be easily confirmed by the small number of working steps the measurement accuracy of the shape measurement device.

As a first configuration of the present invention, linear light from the linear light generating means is irradiated to a measurement object positioned within a field of fixed width, and reflected light of the linear light is captured by the imaging means. In the accuracy determination device of the shape measuring apparatus that receives the light and images the object to be measured and displays the captured image as a cross-sectional shape on the display means, the linear light is irradiated at a predetermined position in the field of view. A reference jig is formed, and an irradiation surface on which the linear light of the reference jig is irradiated is formed as a continuous concavo-convex part along the extending direction of the linear light, and the concavo-convex part is imaged by an imaging means, and display means It is possible to discriminate the accuracy error of the shape measuring device by comparing the displayed image cross-sectional figure with the actual uneven figure, and the display means is divided into small fields for each field of view. an input screen for displaying the region to display so, this target position And it can be input by the designation of a small area.
According to the present invention, the unevenness portion of the reference jig irradiated with linear light is imaged by the image pickup means, and the image cross-sectional figure displayed on the monitor by the display means is compared with the actual uneven figure, thereby It is possible to discriminate the accuracy error of the shape measuring device at a predetermined position in the field of view, and the uneven part of the reference jig imaged at the target position in the field of view is displayed on the monitor as an image cross-sectional figure, which is intuitive and it is possible to obtain a precision error of accurately measuring easier it is to work discrimination accuracy error, also simply select the small region of the field of view displayed on the display unit, accurate measurement of the small region since the determination is performed automatically as possible out to obtain a determination result of easily and accurately measuring the respective small regions.

As a second configuration of the present invention, the shape measuring apparatus includes a jig driving unit that holds the reference jig and drives it to a target position in the field of view, and a linear shape that irradiates the irradiation surface of the reference jig with linear light. And holding means for holding the light generating means, and positioning the reference jig at a plurality of target positions in the field of view so that the image cross-sectional figure of the reference jig at the plurality of target positions can be displayed on the monitor. .
According to the present invention, in a state where the linear light is irradiated, the reference jig is moved to the plurality of target positions in the field of view by the jig driving means and imaged. The accuracy can be checked without replacing the reference jig when determining the details in each field of view or the accuracy error of a wide range by comparing the dimensions of the uneven pattern of the actual reference jig over a wide range.
Furthermore, if a plurality of target positions in the field of view are set and the measurement accuracy in the entire field of view is examined in detail, the number of samplings for discriminating the accuracy error at the position in the field of view increases and the reliability is improved.
Furthermore, by combining the concave or convex portions of the concavo-convex figure, the number of samplings for determining the accuracy error in various dimensions can be increased, and the reliability is improved.
In addition, since the uneven part of the reference jig imaged at the target position in the field of view is displayed on the monitor as an image cross-sectional figure, the accuracy error of the measurement can be obtained intuitively, making it easy to determine the accuracy error. become.

As a third configuration of the present invention, a fixed field of view is divided into a fixed number, and a reference jig is positioned by jig driving means for each small region of the partitioned field of view.
According to the present invention, since the accuracy error of the entire field of view can be examined in detail by determining the accuracy error for each partitioned field of view, when actually measuring the object to be measured, in which partition of the field of view It is possible to know how much accuracy error is present.
In addition, since the reference jig is positioned in the field section by the jig driving means, when checking the accuracy error for each section again, it can be easily positioned at the same position in each section to determine the accuracy error. Therefore, the reproducibility of the accuracy error determination is obtained, and the reliability is improved.

As a fourth configuration of the present invention, an image cross-sectional figure can be displayed on a two-dimensional plane of a monitor display, or an image cross-sectional figure can be displayed on a three-dimensional screen of a monitor display.
According to the present invention, since the image cross-sectional figure is displayed on the two-dimensional plane on the monitor display as the camera field of view, it is possible to easily determine which position in the camera field of view has the accuracy error, or overlay the two-dimensional plane. Since it is displayed as a three-dimensional screen so as to correspond to different focal lengths of the camera field of view, the accuracy error in the depth direction within the camera field of view can be easily determined.

As a fifth configuration of the present invention, the concavo-convex portion of the reference jig is set in a sawtooth shape.
According to the present invention, since the linear light is directly irradiated onto the concave and convex portions formed by the inclined surfaces of the concave and convex portions without being interrupted, the shape of the concave and convex portions can be clearly imaged, so that the determination with high accuracy is possible. Is possible.

As a sixth configuration of the present invention, the holding means holds and drives the linear light generating means so that the linear light is irradiated onto the irradiation surface of the reference jig positioned at a plurality of target positions in the visual field. It was set as the structure to do.
According to the present invention, even when the reference jig is driven to the target position in the field of view by the jig driving means, the irradiation surface of the reference jig can always be irradiated with the linear light. Even if the reference jig is positioned at the target position, the accuracy measurement determination result can be easily obtained.

1 is a schematic configuration diagram of an accuracy determination device according to a first embodiment of the present invention. The enlarged view of the image section figure of Embodiment 1 concerning the present invention. FIG. 5 is a correspondence diagram of the camera field of view and the camera image field of embodiment 1 according to the present invention. FIG. 6 is a correspondence diagram of a camera field of view and a camera image field of embodiment 2 according to the present invention. FIG. 10 is a correspondence diagram of a camera field of view and a camera image field of view according to Embodiment 3 of the present invention. FIG. 6 is a layout diagram of a CCD camera according to a fourth embodiment of the present invention. The figure which shows the reference | standard jig | tool of Embodiment 5 which concerns on this invention. The schematic block diagram of the precision discrimination | determination apparatus of Embodiment 6 which concerns on this invention. The enlarged view of the camera visual field of Embodiment 6 which concerns on this invention. The display figure of the camera image visual field of Embodiment 6 concerning the present invention. The enlarged view of the small area | region of the camera image visual field of Embodiment 6 which concerns on this invention. The display figure of the camera image visual field of Embodiment 6 concerning the present invention. The display figure of the camera image visual field of Embodiment 6 concerning the present invention. FIG. 10 is a correspondence diagram of a camera field of view and a camera image field of embodiment 7 according to the present invention. The schematic block diagram of the actual shape measuring apparatus using the conventional structure. The figure which shows the reference | standard jig | tool for determining the precision of the conventional shape measuring apparatus. The accuracy discrimination | determination apparatus figure for determining the precision of the shape measuring apparatus using the conventional reference | standard jig | tool. The arrangement | positioning figure when using the standard jig | tool of a different magnitude | size conventionally.

Embodiment 1
FIG. 1 is a schematic configuration diagram illustrating an embodiment of an accuracy determination method, an accuracy measurement device, and an accuracy determination reference jig according to the present invention. In the figure, the same reference numerals are used for the same parts as in FIGS.
A shape measuring device including a laser light generator 3 and a CCD camera 5, a computer 6, and a monitor 7 positioned on the left and right of the front surface of the partition 1m is an object of this embodiment.
The CCD camera 5 performs imaging within the camera field of view 1.

In the present embodiment, for the purpose of determining the accuracy of the shape measuring apparatus, it is elongated in the Y direction (vertical direction) covering the entire upper position 1a, center position 1b, and lower position 1c of the virtual camera field 1 in front of the partition 1m. The reference jig 10 is positioned by holding means not shown. The front surface 11 of the reference jig 10 is irradiated from the laser light generator 3 with slit light 4 as linear light extending in the longitudinal direction at the center in the width direction. On the irradiation surface of the front surface 11 of the reference jig 10, an uneven portion 10 j made of continuous unevenness is formed along the extending direction of the slit light 4.
Specifically, the concavo-convex portion 10j is formed in a sawtooth shape composed of inclined surfaces 10f and 10g having known dimensions. The reflected light of the slit light 4 is reflected by the surface of the concavo-convex portion 10j and is received by the CCD camera 5 as an imaging means.

  The concavo-convex cross-section real figure 13 of the concavo-convex portion 10j irradiated with the slit light 4 is displayed as an image cross-section figure 13 'on a two-dimensional plane in the camera image field 1' of the screen of the monitor 7 as display means. The monitor 7 displays the entire camera image field 1 'corresponding to the virtual camera field 1, and the image sectional figure 13' extends in the vertical direction together with the displayed camera image field 1 '. Is displayed. An enlarged display unit 14 is provided on the side (margin portion) of the camera image visual field 1 'in the display screen of the monitor 7 as a display unit, and a cursor 15 controlled by the input unit 6m is provided in the enlarged display unit 14. A part of the image sectional figure 13 'designated by the above is displayed in an enlarged manner.

  A portion adjacent to the enlarged display portion 14 of the monitor 7 is provided with a concave / convex cross-section actual graphic display portion 18 of the reference jig 10, and the actual concave / convex cross-section real graphic 13 of the reference jig 10 is displayed on the display portion 18. . The display of the concavo-convex cross-section actual figure 13 is created by the operator based on the operation of the input means 6m and directly input (not via the CCD camera 5) and displayed under the control of the computer 6.

In the reference jig 10, only the uneven portion 10j is displayed in the camera image field of view 1 ′, and the rectangular base F is unnecessary. Shall be processed. Further, the CCD camera 5 images the partition 1m from the tilt direction for convenience of layout, but an image cross-sectional figure which is a display image according to the position (distance), direction (tilt angle), etc. of the reference jig 10 with respect to the camera. A fixed correction is applied to 13 'by the conversion table 6f. The target position signal 17m of the CCD camera 5 is input to the conversion table 6f. Specifically, in the image cross-sectional figure 13 ′ displayed on the monitor 7, the position and direction of the reference jig 10 differ depending on the position where the image is taken in the camera visual field 1. That is, when the reference jig 10 is close to the CCD camera 5, the image cross-sectional pattern 13 ′ is large, and when the reference jig 10 is far from the CCD camera 5, the image cross-section graphic 13 ′ is small. Therefore, when comparing the image cross-section figure 13 ′ with the concavo-convex cross-section real figure 13 in the conversion table 6 f, the image cross-section figure is based on the imaged position and direction (target position signal 17 m) of the reference jig 10 in the camera visual field 1. The size, shape, etc. of 13 'are converted to match the actual dimensions. Thus, the conversion table 6f automatically converts the size and direction of the captured image cross-sectional figure 13 'so that it matches the actual size, and is corrected so that it can be compared with the concave-convex cross-section real figure 13.
In addition, the slit light 4 of the laser light generator 3 of this example is always irradiated from a fixed position regardless of the position of the uneven portion 10j of the reference jig 10.

The operation of the above configuration is as follows.
As shown in FIG. 1, the reference jig 10 is arranged in, for example, the central portion in the left-right direction (X direction) in the camera visual field 1 in front of the partition 1 m, and the slit light 4 is irradiated from the laser light generator 3, The linear reflected light from the concavo-convex portion 10 j is received by the CCD camera 5, processed by the computer 6, and displayed on the monitor 7 as an image sectional figure 13 ′. This image sectional figure 13 ′ is a figure corresponding to the concave-convex sectional real figure 13.
Next, the cursor 15 is operated up and down to specify a necessary part in the image cross-sectional figure 13 ′ with the cursor 15 and displayed on the enlarged display unit 14 in an enlarged manner. Note that the concave / convex cross-section actual graphic display portion 18 displays the concave / convex cross-section real graphic 13 of the concave / convex portion 10j of the actual reference jig 10 together with the dimensions of the inclined surfaces 10f and 10g. The operator checks the image cross-sectional figure 13 ′ in the enlarged display portion 14 at the positions 1 a, 1 b, 1 c against the concave-convex cross-section real figure 13 in the concavo-convex cross-section real figure display section 18, so that the CCD camera 5 It is possible to determine which of the positions 1a, 1b, and 1c has a camera error. The comparison between the image cross-section figure 13 'and the concavo-convex cross-section real figure 13 is performed by comparing the length dimension or angle of the slope of the serrated concavo-convex part 10j, the distance between the vertices of the concavo-convex part 10j, and the like. In addition, regarding the dimensional determination of the concavo-convex section actual figure 13, software as dimensional determination means for calculating the actual dimension of the sawtooth reference jig 10 in the computer 6 is used, and each side of the sawtooth concavo-convex portion 10 j is calculated. Dimensions are required.

  Specifically, it is obtained by the following calculation procedure. First, pixels constituting each side of the image cross-section figure 13 'of the uneven portion 10j of the reference jig 10 are treated as a data point group, and the point group of each side is linearly approximated by the least square method, and the approximated straight line By calculating the intersection of each other as a serrated bending point and combining the bending points, the length dimension or angle of the inclined surface and the interval dimension between the apexes of the concavo-convex portion 10j are determined.

According to the above configuration, as shown in FIG. 2, for example, the image cross-sectional figure 13 ′ at the center position 1 b ′ in the vertical direction of the camera image visual field 1 ′ is compared with the actual concave-convex cross-sectional figure 13 of the camera visual field 1. If an error is recognized (in this example, the dimensions of the inclined surface 10f 'and the inclined surface 10g' are longer than the actual dimensions and an error is recognized), the CCD camera 5 is assumed to have a measurement error at the center position 1b. Therefore, the accuracy can be corrected by correcting the lens system or the image processing system at this position.
In this example, the size of the image cross-section figure 13 'is corrected by conversion by the conversion table 6f according to the position and direction of the reference jig 10, so that the position and direction of the reference jig 10 are changed. Accordingly, since the size or shape does not change, it is possible to accurately compare with the uneven cross-section actual figure 13.
In this embodiment, by changing the position of the reference jig 10 in the X direction (lateral direction) in the camera visual field 1 and positioning the reference jig 10 in the left part and the right part, the same operation is performed, so that the entire area of the camera visual field 1 is obtained. The camera accuracy can be determined at

  In this embodiment, as shown in FIG. 3, the camera field of view 1 is divided into three in the vertical direction and divided into small regions 1p, 1q, and 1r, and the vertical length of the reference jig 10 is set short. You may make it change the position of the jig | tool 10 to a Y direction for every imaging of the uneven | corrugated | grooved part 10j. Reference numeral 13a denotes an actual concave-convex cross-section figure of the reference jig 10, which is displayed on the monitor 7 as an image cross-section figure 13a '. Since the reference jig 10 is picked up and displayed in the order of the center position 1b, the upper position 1a, and the lower position 1c, the image cross-sectional figure 13a 'at each position 1a', 1b ', 1c' By observing in comparison with the actual figure 13a, the camera accuracy at each position 1a, 1b, 1c can be determined. In this case, as shown in FIG. 3, the reference jig 10 is positioned at a position straddling the boundary 1t, and the uneven cross-section real figure 13b is imaged by the camera, and image determination is performed. Can be checked.

Embodiment 2
Further, in FIG. 3, the camera field of view 1 displayed as a virtual image is vertically divided into three parts. However, as shown in FIG. -R9, the reference jig 10 is positioned for each of the small areas R1 to R9, and the uneven cross-section actual figure 13c is an image cross section for each of the small areas R1 'to R9' of the camera image visual field 1 'in the monitor 7. The figure 13c 'may be configured to be displayed on a two-dimensional plane, and the reference jig 10 may be moved to cover the entire small areas R1 to R9.
With such a configuration, the image cross-sectional figure 13c ′ can be displayed in the small areas R1 ′ to R9 ′ for each of the small areas R1 to R9 in the camera field of view 1. By comparing with the above, the camera accuracy may be determined for each of the small regions R1 to R9.

Embodiment 3
Further, as shown in FIG. 5, the reference jig 10 is formed in a longitudinal shape in the X direction (lateral direction) so that the concavo-convex portion 10 j extends in the horizontal direction, and is continuously formed in the same direction. The reference jig uneven portion 10j is irradiated with the horizontal slit light 4 and imaged by the camera, and an image cross-sectional figure 13d 'formed of an image extending in the horizontal direction (X direction) is displayed on the monitor 7 for determination. It may be.

Embodiment 4
In addition, as shown in FIG. 6, the CCD camera 5 may display an image on the monitor 7 using a dual type camera installed at the left and right.
According to this, when the object to be actually measured becomes a blind spot in the shape measurement by imaging only from the CCD camera 5A, the shape measurement can be performed without the blind spot by providing the CCD camera 5B and imaging. It becomes like this. However, when the CCD camera 5A and the CCD camera 5B are used in this way, it is necessary to know the relationship between the measurement accuracy of each other.
Therefore, in the accuracy discriminating apparatus, the difference in accuracy between each other, that is, the relative accuracy between the CCD camera 5A and the CCD camera 5B can be measured by imaging the concavo-convex portion 10j with the CCD camera 5A and the CCD camera 5B. It becomes like this. In addition, absolute accuracy can be obtained by comparing the image cross-sectional images obtained by the respective CCD cameras 5A and 5B with the actual dimensions of the cross-sectional real graphics of the reference jig 10.
Specifically, the CCD camera 5A and the CCD camera 5B are compared by comparing the distance between the vertices of the image cross-section figure captured by the CCD camera 5A and the distance between the vertices of the image cross-section figure captured by the CCD camera 5B. Relative accuracy can be measured.

Embodiment 5
Moreover, as shown in FIG. 7 (a), the cross-sectional shape of the concavo-convex portion 10j formed on the reference jig 10 may be trapezoidal, and the concave portion is serrated and convex as shown in FIG. 7 (b). As shown in FIG. 7 (c), the convex portion may be serrated as shown in FIG. 7 (c). In short, the concave and convex portions are formed so as to be continuous along the slit light 4. That's fine.

Embodiment 6
FIG. 8 is an apparatus configuration diagram showing another embodiment of the accuracy determining apparatus of the shape measuring apparatus. FIG. 9 shows an enlarged view of the display means of the accuracy discriminating apparatus. In the sixth embodiment, three camera fields 1 (fields 20, 21, 22) are arranged on the front and rear sides of the nine-divided small regions R1 to R9 shown in FIG. 4, and the reference jig 10 is automatically positioned. Therefore, the image cross-sectional figure is displayed as a three-dimensional screen.
As shown in FIG. 8, the accuracy discriminating apparatus for confirming the measurement accuracy of the shape measuring apparatus that performs three-dimensional shape measurement includes a reference jig 10 for confirming the measurement accuracy, and linear light in the reference jig 10. A laser light generator 3 as a linear light generating means for irradiating the slit light 4 as a CCD camera 5 as an imaging means for imaging the concavo-convex portion 10j of the reference jig 10 irradiated by the laser light generator 3. The XYZ table 30 as jig driving means for moving the reference jig 10 back and forth, left and right, and up and down (three-dimensional direction) with respect to the CCD camera 5, and the linear light to the reference jig 10 moved by the XYZ table 30 It is comprised by the XY table 40 as a holding means which moves the laser beam generator 3 so that it may irradiate. The length of the reference jig 10 in the vertical direction is set to a length that falls within the vertical range of each of the small regions R1 to R9.

  In the reference jig 10, an uneven portion 10 j is formed on one end surface of a rectangular body so that serrated unevenness is continuous. The concavo-convex portion 10j is precisely formed such that concave and convex inclined surfaces are alternately triangular, and the concavo-convex cross-section solid figure 13f has known dimensions at each location. The serrated uneven portion 10j is formed in a sawtooth triangular shape in which concave and convex are alternately arranged, and as an example, it is formed to be an isosceles triangle or an equilateral triangle. In addition, the surface of the concavo-convex portion 10j is subjected to a surface treatment such as “no-finish” finish so that the irradiated linear light is not irregularly reflected.

The XYZ table 30 includes three slide stages that move in the X, Y, and Z directions (three-dimensional directions) in the drawing. The three slide stages are an XZ base 31 that is the base of the XYZ table 30, an X direction stage 32 that slides in the X direction on the XZ base 31, a Z direction stage 33 that slides in the Z direction on the X direction stage 32, and a Z direction. A Y base 35 is fixed on the stage 33, and is constituted by a Y direction stage 34 that slides in the Y direction along the Y base 35. The reference jig 10 is attached to the Y direction stage 34. Therefore, the reference jig 10 can be positioned by moving in the three-dimensional direction.
The driving of the X-direction stage 32, the Y-direction stage 34, and the Z-direction stage 33 is controlled by positioning means 17 described later.

The laser light generator 3 includes an infrared line laser that irradiates the irradiation surface (uneven portion 10 j) of the reference jig 10 with infrared slit light 4, and is attached to the XY table 40.
The XY table 40 includes moving means for moving the laser light generator 3 in the X direction and the Y direction shown in FIG. 8, and irradiates the slit light 4 so as to cut the concavo-convex portion 10j of the reference jig 10 in the Y direction. Therefore, driving is controlled by positioning means 17 described later. In the present embodiment, when determining the accuracy of each field of view (fields 20, 21, 22), the XY table holding the laser light generator 3 is slit with respect to the reference jig 10 in a fixed state. Irradiate light 4.
The CCD camera 5 is a monochrome CCD area camera that records captured images as digital data, and includes an infrared transmission filter. That is, by imaging the reflected light from the concavo-convex portion 10j of the reference jig 10 irradiated with light through the infrared transmission filter, only the portion irradiated with the light of the concavo-convex portion 10j is picked up, and this captured image data Is output to the processing means comprising the computer 6.

At least a monitor 7 as a display means and a keyboard and a mouse as an input means 6m are connected to the control means comprising the computer 6, and the image data taken and output by the CCD camera 5 is analyzed and the image cross-section figure is displayed on the monitor 7. 13f ′ is displayed, predetermined contents input from the input unit 6m are displayed, and software is executed to perform measurement based on the command input from the input unit 10.
The positioning means 17 drives the XYZ table 30 based on a signal output from the computer 6 to move the reference jig 10 to a predetermined position in the camera visual field 1 and drives the XY table 40 to drive the laser beam. The XY table 40 is driven so that the slit light 4 irradiated from the generator 3 is irradiated to the uneven portion 10j of the reference jig 10. Further, the positioning unit 17 outputs a signal to the focus control unit 16 so that the focus control unit 16 controls the CCD camera 5 to focus on the uneven portion 10 j of the reference jig 10.

  The CCD camera 5 is focused on a near field 20 at a position closest to the camera, a middle field 21 at a position farther from the near field 20, and a far field 22 at a position farther from the middle field 21. Thus, the near field 20, the middle field 21, and the far field 22 are set by the focus control means 16. These visual fields 20, 21, and 22 are virtual diagrams, and are not actually displayed. A partition 1m (not shown) is positioned behind the far field 22.

  Corresponding to each of these three visual fields 20, 21, and 22, the monitor 7 has a target position input screen 23 comprising three near-field input screens 20a, medium-field input screens 21a, and far-field input screens 22a. And an output screen 24 to be described later are displayed side by side. The target position input screen 23 is a screen in which a near-field input screen 20a, a middle-field input screen 21a, and a far-field input screen 22a are arranged in the depth direction from the front. When the target position input is completed, it disappears from the monitor 7 and the subsequent middle-field input screen 21a projects to the front. When the target position input to the middle-field input screen 21a is completed, the subsequent far-field input screen 22a projects to the front. It is configured as follows.

  In this case, the near field 20, the middle field 21, and the far field 22 are divided into nine regions as described with reference to FIG. Correspondingly, the near-field input screen 20a, the middle-field input screen 21a, and the far-field input screen 22a are similarly set as small areas R1 ′ to R9 ′ divided into nine.

  The target position input screen 23 enables the reference jig 10 to be set at a three-dimensional position. The details will be described. For example, the small area R2 ′ of the near-field input screen 20a is input means 6m or a cursor, or the like. When the touch panel method is selected and designated, the XYZ table 30 is driven, and as shown in FIG. 9, the center portion of the small region R2 of the near field 20 remains with the slit light 4 being irradiated on the reference jig 10. To be aligned. At this time, the focus position of the camera by the focus control means 16 is adjusted to the near field 20.

  In this way, when all of the target position input is completed using the near-field input screen 20a of the target position input screen 23, the near-field input screen 20a is deleted, and the subsequent middle-field input screen 21a is replaced with the front. Any one of the small regions R1 ′ to R9 ′ of the intermediate visual field input screen 21a can be selected. For example, when the small area R1 ′ of the intermediate visual field input screen 21a is selected, the XYZ table 30 is driven and the reference jig 10 is automatically set at the center of the small area R1 of the intermediate visual field 21 as shown in FIG. Aligned.

In this way, the target position is input to a necessary portion of the middle visual field input screen 21a of the target position input screen 23, and the reference jig 10 is aligned corresponding to the target position.
Similarly, when any or all of the small areas R1 ′ to R9 ′ of the far field input screen 22a of the target position input screen 23 are selected, the XYZ table 30 is driven and the reference jig 10 is moved to the far field. The target position input screen 22a is positioned at a position corresponding to any one of the selected small regions R1 ′ to R9 ′.

  On the right side of the screen of the monitor 7, an output screen 24 of the image cross-sectional figure 13 f ′ is displayed side by side on the target position input screen 23. As shown in FIG. 10, the output screen 24 includes a near-field output screen 20b, a middle-field output screen 21b, and a far-field output screen 22b, which are arranged in the depth direction of the screen. That is, the image cross-sectional figure 13f ′ is displayed on a three-dimensional monitor display composed of the near-field output screen 20b, the middle-field output screen 21b, and the far-field output screen 22b. However, the image cross-sectional figure 13f ′ displayed three-dimensionally on the near-field output screen 20b, the middle-field output screen 21b, and the far-field output screen 22b may be displayed on a two-dimensional plane as described later (shown in FIG. 14). Is possible.

As shown in FIG. 10, the near-field output screen 20b is divided into nine sub-regions R1 ′ to R9 ′ corresponding to the near-field input screen 20a, and each of the sub-regions R1 ′ to R9 ′ includes A captured image of the concave-convex cross-section real figure 13f positioned in any or all of the small areas R1 to R9 of the near field 20 is displayed as an image cross-section figure 13f ', and is stored and held by storage means outside the figure. .
As shown in FIG. 12, the middle visual field output screen 21b is divided into nine divided small regions R1 ′ to R9 ′ corresponding to the middle visual field input screen 21a, and each small region R1 ′ to R9 ′ includes A captured image of the concave-convex cross-section actual figure 13f positioned in any or all of the small areas R1 to R9 of the middle visual field 21 is displayed as an image cross-section figure 13f ', and is stored and held by a storage means outside the figure. .
As shown in FIG. 13, the far-field output screen 22b is divided into nine divided small regions R1 ′ to R9 ′ corresponding to the far-field input screen 22a, and each of the small regions R1 ′ to R9 ′ includes A captured image of the concave-convex cross-section actual figure 13f positioned in any or all of the small areas R1 to R9 of the far field 22 is displayed as an image cross-section figure 13f ', and is stored and held by storage means outside the figure. .
Also in this case, the image cross-sectional figure 13f 'is always constant because the image cross-sectional figure 13f' is converted by the conversion table 6f so that the size does not change according to the imaging position and direction of the reference jig 10. Displayed in the size of.
When the accuracy in each field of view is determined, the laser light generator 3 and the CCD camera 5 are not moved and the reference treatment is performed so that the relationship among the laser light generator 3, the reference jig 10, and the CCD camera 5 does not change. The reference jig 10 is moved only in the irradiation direction of the slit light 4 or in the width direction of the slit light 4 so that the position of the tool 10 irradiated with the slit light 4 is not changed.

In the above configuration, the operation will be described below.
First, when the operator designates, for example, the small area R1 ′ of the near-field input screen 20a of the target position input screen 23 from the input means 6m or the like, the XYZ table 30 is driven by the positioning means 17 so that the reference jig 10 The slit light 4 from the laser light generator 3 is positioned in a state where the central position of the reference jig 10 is irradiated on the small region R1 of the near field 20 corresponding to the small region R1 ′.

  The focus of the CCD camera 5 is focused on the near field 20 by the focus control means 16. An image of the concavo-convex cross-section actual figure 13f is captured by the CCD camera 5 to obtain an image cross-section figure 13f '. The image cross-section figure 13f' at this time is a small region R1 of the near-field output screen 20b as shown in FIG. 'Is displayed. In the same way, when another small area of the required near-field input screen 20a is specified, the reference jig 10 is positioned at the target position, and the image of the reference jig 10 is performed by the CCD camera 5, The image cross-sectional figure 13f ′ is displayed in another small area of the near-field output screen 20b.

  When the capturing of the image cross-sectional graphic 13f ′ to the necessary portion of the near-field output screen 20b is completed, the operator checks the image cross-sectional graphic 13f ′. As shown in FIG. 11, this check is performed by, for example, enlarging and displaying the image cross-sectional figure 13f ′ of the small region R1 ′ in the near-field output screen 20b on the monitor 7 along with the dimensions A, B, and C. By an operation such as data display, the contents of the image cross-sectional figure 13f 'are compared with the actual dimensions of the concave-convex cross-sectional real figure 13f of the reference jig 10, and an error is calculated. By this check, the accuracy error of the CCD camera 5 with respect to the small region R1 ′ corresponding to the near-field output screen 20b, that is, the small region R1 of the near-field 20 can be evaluated.

  When the determination of the accuracy error of the CCD camera 5 with respect to the near-field 20 is completed, the medium-field input screen 21a and the medium-field output screen 21b are read out to the front of the screen of the monitor 7 and the target position is input to the necessary part of the middle-field input screen 21a. Then, the concave-convex cross-section actual figure 13f is positioned at a required portion of the middle visual field 21, and then imaged by the CCD camera 5, and as shown in FIG. Is displayed, and the image cross-section figure 13f 'is checked.

When the determination of the accuracy error of the CCD camera 5 with respect to the middle visual field 21 is completed, the far-field input screen 22a and the far-field output screen 22b are read out to the front surface of the monitor 7, and the target position is set at a necessary position on the far-field input screen 22a. Is input to the required portion of the far field 22 to position the concave-convex sectional real image, and then imaged by the CCD camera 5, and as shown in FIG. 13, the image sectional graphic 13f is displayed at the required portion of the far field output screen 22b. 'Is displayed, and the image cross-section figure 13f' is checked.
The reference jig 10 is positioned at a predetermined position corresponding to the near-field input screen 20a, the middle-field input screen 21a, and the far-field input screen 22a, but the position and direction with respect to the CCD camera 5 change. When the image cross-section figure 13f 'is displayed, conversion and correction are performed by the conversion table 6f and displayed in the same size and can be easily compared with the uneven cross-section real figure 13f.

As described above, according to the present embodiment, the camera field of view 1 is divided into small regions, and the concave-convex cross-section real figures can be positioned in the respective small regions. Thus, it is possible to determine the accuracy error of the camera, so that it is possible to determine the accuracy of a high range.
In addition, since the camera visual field 1 is set to near, middle, and far, it is possible to determine the accuracy corresponding to each focal position of the camera. In this case, even if the reference jig 10 is set at a position in the near, middle, far, left and right, up and down directions of the camera field of view 1, the size of the image cross-section figure 13f 'is the size of the uneven cross-section real figure 13f by the conversion table 6f. Since it is converted so as to match, the comparison between the two can be performed accurately.

Embodiment 7
In the apparatus configuration shown in FIG. 8 in the sixth embodiment, when the reference jig 10 is set to face the laser light generator 3 as shown in FIG. 3 shows a mode in which the XY table 40 as a holding means for moving the head 3 is fixed and only the reference jig 10 is moved in the Y direction or the Z direction to check the measurement accuracy of the shape measuring apparatus. In this case, the measurement accuracy can be confirmed by operating the accuracy discrimination device as follows.
First, in a small area R1 to R9 of an intermediate position of a 9-divided type in which the fixed camera field of view 1 is divided into 9, the reference jig 10 is temporarily positioned with the small area R5 at the center position as a reference position, and the reference is made at this position. The slit light 4 is irradiated from the front directly onto the concave and convex portion 10j of the jig 10, and the laser light generator 3 is fixed so as to cover the entire Y direction of the camera visual field 1, and the reference jig 10 is moved to the camera. The field of view 1 is moved in the Y and Z directions in the figure. Note that the small regions R1 to R9 in FIG. 14 are regions at intermediate positions, and the small regions R10 to R12 are those of the camera field of view 1 when the reference jig 10 is brought close to the facing direction of the laser light generator 3. This is a near position area corresponding to the small areas R7 to R9. Further, the small regions R13 to R15 are regions at far positions corresponding to the small regions R1 to R3 of the camera visual field 1 when the reference jig 10 is moved away from the laser light generator 3 in the facing direction.
In this case, the slit light 4 is set long in the vertical direction so that the slit light 4 is irradiated at any position of the reference jig 10. The CCD camera 5 is set to wide-angle imaging so as to cover the reference jig 10 at all positions.

Specifically, as shown in FIG. 14, the reference jig 10 is moved to the small region R14 from the position of the small region R5 of the camera visual field 1 so that the distance from the CCD camera 5 is increased. By capturing the real figure 13f, the image cross-sectional figure 13f 'converted to the actual size is output via the conversion table 6f at the position of the small region R2' of the camera image visual field 1 '. From this position, the reference jig 10 is moved upward so as to be positioned in the small region R13 and imaged, so that the actual size is obtained via the conversion table 6f to the position of the small region R1 ′ of the camera image field of view 1 ′. The converted image cross-sectional figure 13f 'is output. Further, by moving the reference jig 10 downward from the small region R14 so as to be positioned in the small region R15 and taking an image, the actual size is obtained via the conversion table 6f at the position of the small region R3 ′ of the camera image field of view 1 ′. The image cross-section figure 13f ′ converted into is output.
Similarly, the reference jig 10 is moved to the small area R11 so that the distance from the position of the small area R5 in the camera field of view 1 to the facing direction of the laser light generator 3 in the Z direction is moved to the uneven cross section. By imaging the figure 13f, an image sectional figure 13f ′ converted to an actual size is output via the conversion table 6f at the position of the small region R8 ′ of the camera image field of view 1 ′. From this position, the reference jig 10 is moved upward so as to be positioned in the small region R10 and imaged, so that the actual size is obtained via the conversion table 6f to the position of the small region R7 ′ in the camera image field 1 ′. The converted image cross-sectional figure 13f 'is output. Further, by moving the reference jig 10 downward from the small region R11 so as to be positioned in the small region R12 and taking an image, the actual size is obtained via the conversion table 6f at the position of the small region R9 'of the camera image visual field 1'. The image cross-section figure 13f ′ converted into is output.
That is, the small areas R1 ′ to R3 ′ of the camera image field 1 ′ as a two-dimensional screen correspond to the small areas R13 to R15, and the small areas R4 ′ to R6 ′ of the camera image field 1 ′ and the small area of the camera field 1 The areas R4 to R6 correspond to each other, and the small areas R7 'to R9' of the camera image field of view 1 'correspond to the small areas R10 to R12.
That is, the image cross-sectional figure 13f ′ is configured to be displayed on a monitor on a two-dimensional plane with respect to each position of the reference jig 10 in the three-dimensional direction.

According to such a configuration, the reference jig 10 is placed within the irradiation range of the slit light 4 in a plurality of small areas in the camera visual field 1 while the laser light generator 3 and the CCD camera 5 are fixed at specific positions. A slit irradiated to the concavo-convex portion 10j of the reference jig 10 by moving the distance in the vertical direction (Y direction) and the facing direction between the reference jig 10 and the laser light generator 3 closer to or away from each other. Since the accuracy error of the camera can be determined over the entire area of the camera visual field 1 without changing the position of the light 4, it is possible to determine the accuracy in the camera visual field 1 with respect to the focus of one CCD camera 5 with higher accuracy. Become.

Also in this example, the size of the image cross-section figure 13 'is converted by the conversion table 6f in accordance with the position where the reference jig 10 is imaged, so that the size of the image cross-sectional figure 13' is determined according to the position where the reference jig 10 is imaged Since the size does not change, it is possible to accurately compare with the uneven cross-section actual figure 13.

According to the first to seventh embodiments, for example, it is possible to check the accuracy of the dual head camera for the mold measuring device and the accuracy of the dual head camera for the first pot inspection device.
In the above description, the imaging means has been described as a CCD camera, but another camera may be used as long as the camera outputs image data that can be digitally processed by a computer or the like. Moreover, although the linear light generating means has been described as a laser light generator, it may be one that can irradiate the measured object with slit light having the same width. Also, in the accuracy determination in the field of view, it has been described that the reference jig is moved while the slit light is fixed at a predetermined position of the reference jig. However, if accuracy is not required, it may be moved together with the reference jig. Good.

1 camera field of view (virtual), 1 'camera image field of view, 1a upper position, 1b center position,
1c Lower position, 1m partition, 1p, 1q, 1r small area, 1t boundary, 2 measured object,
3 Laser light generator, 4 slit light, 5 CCD camera, 6 computer,
6m input means, 7 monitor, 8 and 9 cross section actual figure, 8 'and 9' image cross section figure,
9a Reference jig, 10 Reference jig, 10j Concavity and convexity,
10f, 10g inclined surface, 10f ', 10g' inclined surface, 11 front surface,
13 concavo-convex cross section real figure, 13 'image cross-section figure, 14 enlarged display part, 15 cursor,
16 focus control means, 17 positioning means, 18 concavo-convex section real figure display section,
20 near field, 21 middle field, 22 far field,
20a Near-field input screen, 21a Middle-field input screen, 22a Far-field input screen,
20b near field output screen, 21b medium field output screen, 22b far field output screen,
23 Target position input screen, 24 output screen, 30 XYZ table,
40 XY table, F base, R1-R9 small area, R1'-R9 'small area,
a Vertical dimension, b Depth dimension.

Claims (6)

By irradiating linear light from the linear light generating means to the measurement object is positioned within the field of view of a certain measuring the reflected light of the linear light is received by the image pickup means, the object to be measured In the accuracy determination device of the shape measuring apparatus that enables the monitor to display the captured image as a cross-sectional shape on the display means,
A reference jig that is positioned at a predetermined position in the field of view and that is irradiated with the linear light is provided, and an irradiation surface that is irradiated with the linear light of the reference jig is continuous along the extending direction of the linear light. Formed as a concavo-convex part, configured so that the concavo-convex part is imaged by the imaging means and displayed on the monitor by the display means, and the displayed image cross-sectional figure is compared with the actual concavo-convex figure, thereby measuring the shape thereby enabling determination of accurate error,
An accuracy discriminating apparatus for a shape measuring apparatus, wherein an input screen for displaying a small area divided for each field of view is displayed on the display means, and a target position can be input by designating the small area . Jig driving means for holding the reference jig and driving it to a target position in the field of view;
Holding means for holding the linear light generating means for irradiating the irradiation surface of the reference jig with linear light, positioning the reference jig at a plurality of target positions in the field of view, and a plurality of target positions 2. The accuracy determining apparatus for a shape measuring apparatus according to claim 1 , wherein an image cross-sectional figure of the reference jig can be displayed on a monitor. 3. The field of view according to claim 2 , wherein the field of view having a certain width is partitioned into a certain number, and the reference jig is positioned by jig driving means for each small region of the sectioned field of view. Accuracy determination device for shape measuring device. The image cross-section shape of the displayable on a two-dimensional plane of the monitor display, or the image cross-section shape or any claims 1 to 3, characterized in that to be displayed on the monitor display of a three-dimensional screen the An accuracy determination apparatus for the shape measuring apparatus described. Uneven portion of the reference jig, the accuracy determination device in the form measuring apparatus according to any one of claims 1 to 4, characterized in that set in the serrated. The holding means holds and drives the linear light generating means so that linear light is irradiated onto an irradiation surface of the reference jig positioned at a plurality of target positions in the visual field. precision determination device of a shape measuring device according to any 請 Motomeko 2 to claim 5.

Images