JP2013160531A - Inspection apparatus, robot apparatus, inspection method, and inspection program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To inspect a defect of appearance of an inspection object with high accuracy.SOLUTION: An inspection apparatus comprises: an imaging unit taking images of an inspection object and generating image data including an inspection area that is an image area corresponding to an inspection region of the inspection object; a smoothing unit smoothing pixel values of respective pixels included in the image data and generating smoothed image data; a difference unit generating difference image data by obtaining, for each of the pixels, a difference of pixel values between pixels of the image data and pixels of the smoothed image data corresponding to the pixels; and a determination unit determining whether the inspection object is good or not based on similarity between reference image data as an inspection reference of the inspection region and the difference image data.

Description

  The present invention relates to an inspection apparatus, a robot apparatus, an inspection method, and an inspection program.

  A technique for performing inspection of an object to be inspected based on image recognition processing is known (see, for example, Patent Document 1). The inspection apparatus disclosed in this document is an apparatus that detects a defect of an inspection target object based on the similarity between an image obtained by imaging the inspection target object and a non-defective image that has been captured in advance.

JP 2009-258968 A

  However, in the above inspection apparatus, if there is a portion where the individual difference of the surface of the inspection target object is remarkable, such as a minute unevenness or pattern on the surface of the inspection target object, the inspection target object is similar even if it is a non-defective product. The degree became smaller and the individual difference was sometimes detected as a defect. In this case, the above-described inspection apparatus has a problem that it is not possible to inspect a defect in the outer shape of the inspection target object with high accuracy.

  The present invention has been made to solve the above problem, and an inspection apparatus and robot capable of inspecting a defect in the outer shape of an inspection target object with high accuracy even if there are individual differences on the surface of the inspection target object An object is to provide an apparatus, an inspection method, and an inspection program.

[1] In order to solve the above-described problem, an inspection apparatus according to an aspect of the present invention images an inspection target object and includes an inspection region that is an image region corresponding to an inspection region of the inspection target object. An imaging unit that generates data, a smoothing unit that generates smoothed image data by smoothing pixel values of each pixel included in the image data, a pixel of the image data, and the smoothed image data corresponding to the pixel The difference between the pixel values of each pixel is obtained for each pixel, and the examination is performed based on the difference between the difference unit that generates difference image data, and the reference image data as the examination reference of the examination region and the difference image data. And a determination unit that determines whether or not the target object is a non-defective product.
With this configuration, the inspection apparatus can inspect defects on the outer shape of the inspection target object with high accuracy even if there are individual differences on the surface of the inspection target object.

[2] The inspection apparatus according to [1], further including a setting unit that sets a cut-off frequency of a low-pass filter based on a spatial frequency of outer shape image data indicating an outer shape portion of an inspection site included in the image data. The smoothing unit generates the smoothed image data by performing an operation on the image data using a low-pass filter having the cutoff frequency.
With this configuration, the inspection apparatus can set the characteristics of the low-pass filter for generating smooth image data according to the characteristics of the image data. As a result, the inspection apparatus can inspect the external defect of the inspection target object with high accuracy even if there is an individual difference on the surface of the inspection target object. Moreover, since the inspection apparatus performs smoothing using a known low-pass filter, it can generate smoothed image data with a simple configuration.

[3] In the inspection apparatus according to [2], the setting unit sets the spatial frequency intensity of the outline image data to a cutoff frequency lower than the strongest frequency except for a constant component, and A cut-off frequency is set.
With this configuration, the inspection apparatus generates smooth image data in which the shade of each pixel indicating the outer portion of the inspection site is removed and the shade of each pixel indicating the surface portion of the inspection site is not removed. can do. Therefore, the inspection apparatus retains the shading of each pixel indicating the outer portion of the inspection region as difference image data obtained by calculating the difference between the pixel values of the smoothed image data and the image data before smoothing. The difference image data from which the shading of each pixel indicating the surface portion is removed can be generated. Thereby, the inspection device can reduce the influence of the surface portion of the inspection region and inspect the outer shape portion of the inspection region with high accuracy.

[4] In order to solve the above-described problem, the robot apparatus according to one aspect of the present invention images an inspection target object and includes an inspection region that is an image region corresponding to the inspection region of the inspection target object. An imaging unit that generates data, a smoothing unit that generates smoothed image data by smoothing pixel values of each pixel included in the image data, a pixel of the image data, and the smoothed image data corresponding to the pixel The difference between the pixel values of each pixel is obtained for each pixel, and the examination is performed based on the difference between the difference unit that generates difference image data, and the reference image data as the examination reference of the examination region and the difference image data. A determination unit that determines whether the target object is a non-defective product and a robot main body that movably supports at least the imaging unit are provided.
With this configuration, the robot apparatus can inspect defects in the outer shape of the inspection target object with high accuracy even if there are individual differences on the surface of the inspection target object.

  [5] The robot apparatus according to [4], further including a setting unit that sets a cutoff frequency of a low-pass filter based on a spatial frequency of outer shape image data indicating an outer shape portion of an examination site included in the image data, The smoothing unit generates the smoothed image data by performing an operation on the image data using a low-pass filter having the cutoff frequency.

  [6] In the robot device according to the above [4] or [5], the setting unit may set a cutoff frequency at which the spatial frequency intensity of the outline image data is lower than the strongest frequency excluding a constant component, A cutoff frequency of the low-pass filter is set.

[7] The robot apparatus according to any one of [4] to [6], further including a pedestal portion that supports the robot body and a portable portion that makes the pedestal portion portable. .
With this configuration, the robot apparatus can be miniaturized because it incorporates the inspection apparatus and the control apparatus necessary for the inspection. Further, the robot apparatus can perform inspection by changing the place by the portable unit. For this reason, for example, the robot apparatus can inspect even a large inspection object that is difficult to move.

  [8] In order to solve the above-described problem, an inspection method according to an aspect of the present invention includes an inspection region that includes an inspection region that is an image region corresponding to an inspection region of the inspection target object by imaging the inspection target object. An imaging step for generating image data of a target object image, a smoothing step for smoothing pixel values of each pixel included in the image data to generate smooth image data, a pixel of the image data, and the corresponding pixel The difference between the pixel values of the smoothed image data is calculated for each pixel and the difference step for generating the difference image data, and the similarity between the difference image data and the reference image data as the examination reference of the examination region And a determination step of determining whether or not the object to be inspected is a non-defective product.

  [9] In order to solve the above-described problem, an inspection program according to an aspect of the present invention includes an inspection area that is an image area corresponding to an inspection region of an inspection target object by using a computer to image the inspection target object. An imaging unit that generates image data included, a smoothing unit that generates smoothed image data by smoothing pixel values of each pixel included in the image data, the pixels of the image data, and the pixels corresponding to the pixels The difference between the pixel values of the pixels of the smooth image data is obtained for each pixel, and based on the similarity between the difference unit that generates the difference image data and the reference image data as the inspection reference of the inspection region and the difference image data And a determination unit that determines whether or not the object to be inspected is a non-defective product.

  Therefore, according to each aspect of the present invention, an inspection apparatus, a robot apparatus, an inspection method, and an inspection capable of inspecting a defect in the outer shape of the inspection target object with high accuracy even if there are individual differences on the surface of the inspection target object A program can be provided.

FIG. 2 is a schematic external view of a robot body and an inspection target object in the robot apparatus according to the first embodiment of the present invention. 2 is a block diagram illustrating a schematic functional configuration of a robot apparatus according to the present embodiment. FIG. It is a block diagram showing the functional structure of the inspection apparatus in this embodiment. It is a schematic diagram which shows an example of the image data which the imaging part in this embodiment produced | generated. It is a schematic diagram which shows an example of the external shape image data which shows the external shape part of the test | inspection site | part in this embodiment. It is a graph which shows an example of the spatial frequency of the external shape image data in this embodiment. It is a schematic diagram which shows an example of the smooth image data which the smoothing part in this embodiment produced | generated. It is a schematic diagram which shows an example of the difference image data which the difference part in this embodiment produced | generated. It is a flowchart which shows an example of operation | movement of the test | inspection apparatus in this embodiment. It is a general | schematic external view of the robot apparatus which is 2nd Embodiment of this invention.

Hereinafter, embodiments for carrying out the present invention will be described in detail with reference to the drawings.
[First embodiment]
FIG. 1 is a schematic external view of a robot body and an inspection object in the robot apparatus according to the first embodiment of the present invention. As shown in the figure, the robot apparatus 1 is configured by providing an imaging unit 21 in a robot body 10.

The robot body 10 supports the imaging unit 21 movably. Specifically, the robot body 10 includes a support base 11 fixed to the ground, link parts 12-1 to 12-5, joint parts 16-1 to 16-6, and a link part 13. Is done. Here, the link part 12-1 is connected with the support stand 11 through the joint part 16-1. The link part 12-2 is connected to the link part 12-1 via the joint part 16-2. The link part 12-3 is connected to the link part 12-2 via the joint part 16-3. The link part 12-4 is connected to the link part 12-3 via the joint part 16-4. The link part 12-5 is connected to the link part 12-4 via the joint part 16-5. In other words, the link portion 12-5 is connected to the support base 11 through the link portions 12-1 to 12-4 and the joint portions 16-1 to 16-5 so as to be capable of turning and bending. Moreover, the link part 13 is connected with the link part 12-5 via the joint part 16-6 so that rotation and a swing are possible. Moreover, the imaging part 21 is being fixed to the link part 13 via the connection part. Hereinafter, since each link part and each joint part have the same configuration, the link parts 12-1 to 12-5 are the link part 12 and the joint parts 16-1 to 16-6 unless otherwise distinguished. It is described as a joint part 16.
As described above, the robot body 10 is, for example, a 6-axis vertical articulated robot, and has 6-axis degrees of freedom by the coordinated operation of the support base 11, the link unit 12, and the link unit 13. The position and orientation can be freely changed in the three-dimensional space.

Note that the robot body 10 may change the imaging unit 21, tools, parts, and the like according to the work purpose. Further, the degree of freedom of the robot body 10 is not limited to that with six axes.
Moreover, you may install the support stand 11 in the place fixed with respect to the grounds, such as a wall and a ceiling. In addition to the link unit 12 and the link unit 13 that support the imaging unit 21, the robot body 10 includes a link unit (not shown) that supports tools and parts, and these multiple link units are independently or interlocked. It is good also as a structure to move. The imaging unit 21 may be configured to be fixed to the link unit 13 as described above, and may be configured to be gripped by a hand unit (not shown) provided at the tip of the link unit 13. May be.

As shown in FIG. 1, for example, in the movable range of the tip of the link portion 13 of the robot body 10, an inspection target object 5 that is a visual inspection target is placed on a table (not shown). This inspection target object 5 has an inspection region.
That is, the robot apparatus according to the present embodiment is an apparatus that inspects the appearance of the inspection target object 5 to inspect the state of the inspection part, for example, whether or not there is an inspection object in the inspection part.

FIG. 2 is a block diagram illustrating a schematic functional configuration of the robot apparatus according to the present embodiment.
As shown in the figure, the robot apparatus 1 includes a robot body 10, an inspection apparatus 20, and a control apparatus 30. As shown in FIG. 1, the robot body 10 includes an imaging unit 21 and a link unit 12.

  The control device 30 transmits control signals such as an imaging start request signal and an imaging stop request signal to the imaging unit 21. In addition, the control device 30 drives the link unit 12 to control the posture of the robot body 10 in order to change the shooting direction of the imaging unit 21 in the three-dimensional space.

  The imaging unit 21 images the inspection target object 5 and generates image data including an inspection region that is an image region corresponding to the inspection region of the inspection target object 5. The imaging unit 21 of the present embodiment is, for example, a still image camera device capable of monochrome photography or color photography that images the inspection target object 5 and outputs image data. For example, the imaging unit 21 captures an image with a predetermined exposure time (for example, 1/30 second). Note that the imaging unit 21 may be a video camera device. In accordance with the imaging start request signal supplied from the control device 30, the imaging unit 21 images the inspection target object 5 shown in FIG. In addition, the imaging unit 21 stops the imaging operation according to the imaging stop request signal supplied from the control device 30.

The inspection device 20 includes a determination device 22 and the imaging unit 21 described above.
The determination device 22 takes in the image data output from the imaging unit 21 sequentially or every plurality of frames. Then, the determination device 22 outputs inspection result data that determines the state of the inspection region based on the reference image as the inspection reference of the inspection region that has been imaged in advance and the captured image data.

FIG. 3 is a block diagram showing a functional configuration of the inspection apparatus 20 in the present embodiment.
As described above, the inspection device 20 includes the imaging unit 21 and the determination device 22.
The determination device 22 includes an image data storage unit 221, a smoothing unit 222, a smoothed image data storage unit 223, a difference unit 224, a determination unit 225, a reference image storage unit 226, and a setting unit 227. .

  Image data is stored in the image data storage unit 221 by the imaging unit 21. In the image data storage unit 221 of the present embodiment, image data output from the imaging unit 21 is stored.

  The setting unit 227 sets the cutoff frequency of the low-pass filter based on the spatial frequency of the outline image data indicating the outline portion of the examination site included in the image data. Here, the contour image data indicating the contour portion of the examination site is a portion indicating the contour portion of the examination site (for example, the ridge line portion of the examination target object 5) in the image data generated by the imaging unit 21. For example, in each pixel included in the image data generated by the imaging unit 21, a portion where the pixel value greatly changes with respect to a neighboring pixel is an outer portion of the examination region (for example, a ridge line portion of the examination target object 5). . For example, the setting unit 227 of the present embodiment reads the image data generated by the imaging unit 21 from the image data storage unit 221 and performs a known edge extraction process on the read image data to indicate the outer portion of the examination site. Outline image data is generated. Further, the setting unit 227 calculates a spatial frequency of the outline image data by applying a known Fourier transform to the generated outline image data. The setting unit 227 sets the cutoff frequency of the low-pass filter based on the spatial frequency calculated in this way.

  More specifically, the setting unit 227 sets the cut-off frequency of the low-pass filter to a cut-off frequency lower than the frequency having the highest spatial frequency intensity of the outline image data. Here, the frequency having the highest spatial frequency intensity is the pixel value when the known Fourier transform is applied to the outline image data and the relationship between the frequency of the outline image data and the energy intensity of the pixel value is calculated. Is the frequency with the strongest energy intensity. For example, the setting unit 227 of the present embodiment obtains the energy intensity of the pixel value for each predetermined frequency band with respect to the frequency of the spatial frequency of the calculated outline image data. Then, the setting unit 227 sets a frequency lower than the frequency band where the energy intensity of the pixel value is the strongest as the cutoff frequency. The setting unit 227 outputs the set cutoff frequency to the smoothing unit 222. In the above description, the configuration in which the setting unit 227 generates the outline image data has been described. However, the setting unit 227 may have a configuration in which the outline image data is supplied from the outside. In this case, the configuration of the setting unit 227 can be simplified.

The smoothing unit 222 smoothes the pixel value of each pixel included in the image data to generate smoothed image data. For example, the smoothing unit 222 according to the present embodiment calculates the pixel value of the smoothing target pixel included in the image data as an average value of the pixel value of the smoothing target pixel and the pixel value of the pixel adjacent to the smoothing target pixel. Is applied to each pixel to generate smoothed image data.
More specifically, smooth image data is generated by performing an operation on the image data using a low-pass filter having a cutoff frequency. Here, the smoothing unit 222 acquires the cutoff frequency set by the setting unit 227 from the setting unit 227, and generates smoothed image data using the acquired cutoff frequency as the cutoff frequency of the low-pass filter. Further, the smoothing unit 222 stores the generated smoothed image data in the smoothed image data storage unit 223 in association with the image data before smoothing. Details of the configuration of the smoothing unit 222 will be described later.
The smoothed image data storage unit 223 stores the smoothed image data generated by the smoothing unit 222 in association with the image data by the imaging unit 21.

  The difference unit 224 obtains a difference between the pixel value of the pixel of the image data and the pixel of the smoothed image data corresponding to the pixel, and generates difference image data. The difference unit 224 of the present embodiment reads, for example, image data generated by the imaging unit 21 stored in the image data storage unit 221 and smoothed image data generated by the smoothing unit 222 associated with the read image data. Are read from the smoothed image data storage unit 223. And the difference part 224 calculates | requires the difference of the pixel value of each corresponding pixel between the read image data and the read smooth image data, and produces | generates difference image data. Further, the difference unit 224 outputs the generated difference image data to the determination unit 225 described later.

  In the reference image storage unit 226, reference image data as an inspection reference for the inspection region is stored in advance. In the reference image storage unit 226 of this embodiment, for example, non-defective image data that is a reference for inspection of the inspection object 5 is stored in advance as reference image data. Similar to the above-described difference image data, the reference image data is image data generated in advance by calculating image data obtained by capturing a non-defective product by the smoothing unit 222 and the difference unit 224.

  The determination unit 225 determines the state of the inspection region based on the reference image data as the inspection reference of the inspection region and the difference image data. The determination unit 225 according to the present embodiment reads the reference image data stored in the reference image storage unit 226 and obtains the similarity between the difference image data generated by the difference unit 224 and the read reference image data. Then, when the obtained similarity is larger than a predetermined value (that is, when the similarity between the difference image data and the reference image data is large), the determination unit 225 determines that the inspection target object 5 is a non-defective product. On the other hand, when the obtained similarity is smaller than a predetermined value (that is, when the similarity between the difference image data and the reference image data is small), the determination unit 225 determines that the inspection target object 5 is a defective product. .

Next, a configuration for generating difference image data from image data will be described with reference to FIGS.
First, an example of image data generated by the imaging unit 21 will be described with reference to FIG.
FIG. 4 is a schematic diagram illustrating an example of image data generated by the imaging unit 21 in the present embodiment.
As described above, the imaging unit 21 of the present embodiment images the inspection target object 5 with an exposure time of 1/30 seconds, for example, and generates image data illustrated in FIG. 4, for example.

Next, an example of the cutoff frequency set by the setting unit 227 will be described with reference to FIGS. 5 and 6.
FIG. 5 is a schematic diagram illustrating an example of outer shape image data indicating the outer shape portion of the examination site in the present embodiment.
As described above, the setting unit 227 of the present embodiment generates, for example, outer shape image data as illustrated in FIG. 5 that indicates the outer shape portion of the examination site included in the image data.

FIG. 6 is a graph showing an example of the spatial frequency of the outline image data in the present embodiment.
As described above, the setting unit 227 calculates a spatial frequency of the outline image data as illustrated in FIG. 6 by applying a known Fourier transform to the generated outline image data. Here, it is assumed that the spatial frequency of the contour image data has a distribution indicated by the distribution waveform W in FIG. The spatial frequency distribution waveform W of the outline image data is divided into a frequency domain A and a frequency domain B shown in FIG. Based on the spatial frequency property of the image data obtained by the Fourier transform described above, the frequency region B is a frequency region including the frequency PF having the strongest spatial frequency intensity of the outline image data. Further, the frequency region B represents an outer shape portion (for example, the contour portion shown in FIG. 5) of the inspection site of the inspection target object 5. Similarly, the frequency region A is a frequency region having a frequency lower than the frequency PF in which the intensity of the spatial frequency of the outline image data is the strongest, and is a surface portion (for example, in FIG. This represents a portion other than the contour portion shown).
For example, the setting unit 227 of the present embodiment sets a frequency (that is, a frequency included in the frequency region A) lower than the frequency PF having the highest spatial frequency intensity of the outline image data as the cutoff frequency CF. At this time, the characteristic of the low-pass filter that cuts off the frequency higher than the cutoff frequency CF is, for example, a characteristic waveform LPF shown in FIG. Therefore, this low-pass filter has a characteristic of blocking the frequency region B (for example, the contour portion shown in FIG. 5).

Next, a configuration in which the smoothing unit 222 of the present embodiment generates smoothed image data will be described.
FIG. 7 is a schematic diagram illustrating an example of smoothed image data generated by the smoothing unit 222 according to the present embodiment.
The smoothing unit 222 according to the present embodiment generates smooth image data as shown in FIG. 7, for example, using a low-pass filter that cuts off a frequency higher than the cutoff frequency CF set in this way. Therefore, the smoothed image data generated by the smoothing unit 222 is data that does not include the frequency component of the outer portion (for example, the contour portion shown in FIG. 5) of the examination site in the image data. As described above, the smooth portion 222 has the characteristics of the surface portion (for example, the portion other than the contour portion shown in FIG. 5) in the inspection region, and the outer shape portion (for example, in FIG. 5). Smooth image data from which the feature of the contour portion shown) is removed is generated.

Next, a configuration in which the difference unit 224 of the present embodiment generates difference image data will be described.
FIG. 8 is a schematic diagram illustrating an example of difference image data generated by the difference unit 224 in the present embodiment.
As described above, the difference unit 224 obtains the difference between the pixel value of the pixel of the image data and the pixel of the smoothed image data corresponding to the pixel, and generates difference image data as shown in FIG. At this time, the difference unit 224 has removed the features of the surface portion (for example, the portion other than the contour portion shown in FIG. 5) of the examination site in the image data, and the outer portion ( For example, the difference image data in which the feature of the contour portion shown in FIG. 5 is left is generated.

  In this way, the determination unit 225 according to the present embodiment determines the state of the examination part based on the difference image data and the reference image data as the examination reference of the examination part. Here, the reference image data is difference image data in which image data in which a good product is captured in advance is generated with the above-described configuration. As described above, the determination unit 225 according to the present embodiment calculates the similarity between the difference image data based on the image data obtained by imaging the non-defective product and the difference image data based on the image data obtained by imaging the inspection target object 5. By doing so, the state of the examination site is determined.

Next, operation | movement of the test | inspection apparatus 20 in this embodiment is demonstrated.
FIG. 9 is a flowchart showing an example of the operation of the inspection apparatus 20 in the present embodiment.
First, the control device 30 controls the link unit 12 of the robot body 10 to move the position of the imaging unit 21 to a position where the imaging unit 21 can capture the inspection area of the inspection target object 5 (step S201). In the present embodiment, the inspection target object 5 is arranged at a predetermined position, for example, and the control device 30 sets the imaging unit 21 to the imaging position based on the movement amount of the link unit 12 stored in advance. Move.

Next, the imaging unit 21 images the inspection target object 5 and generates image data of an inspection target object image including an inspection region that is an image region corresponding to the inspection region of the inspection target object 5 (step S203, imaging). Step).
Next, the imaging unit 21 stores the generated image data in the image data storage unit 221 (step S204).

  Next, the setting unit 227 sets the cut-off frequency of the low-pass filter based on the spatial frequency of the outline image data indicating the outline portion of the examination site included in the image data (Step S205). As described above, the setting unit 227 of the present embodiment reads, for example, the image data generated by the imaging unit 21 from the image data storage unit 221 and performs a known edge extraction process on the read image data to perform an examination site. The external shape image data indicating the external shape portion of the image is generated. Further, the setting unit 227 calculates a spatial frequency of the outline image data by applying a known Fourier transform to the generated outline image data. More specifically, the setting unit 227 sets the cut-off frequency of the low-pass filter to a cut-off frequency lower than the frequency having the highest spatial frequency intensity of the outline image data. The cut-off frequency may be set once according to the inspection site of the inspection target object 5. That is, the cutoff frequency does not need to be changed as long as the inspection site of the inspection object 5 is the same.

  Next, the smoothing unit 222 smoothes the pixel value of each pixel included in the image data to generate smoothed image data (step S206, smoothing step). As described above, the smoothing unit 222 of the present embodiment, for example, calculates the pixel value of the smoothing target pixel included in the image data between the pixel value of the smoothing target pixel and the pixel value of the pixel adjacent to the smoothing target pixel. Smoothing image data is generated by applying an arithmetic operation for averaging to each pixel. More specifically, smooth image data is generated by performing an operation on the image data using a low-pass filter having a cutoff frequency. Then, the smoothing unit 222 stores the generated smoothed image data in the smoothed image data storage unit 223 in association with the image data before smoothing.

  Next, the difference unit 224 obtains the difference between the pixel value of the pixel of the image data and the pixel of the smoothed image data corresponding to the pixel, and generates difference image data (step S207, difference step). As described above, the difference unit 224 of the present embodiment reads, for example, the image data generated by the imaging unit 21 stored in the image data storage unit 221, and the smoothing unit 222 associated with the read image data. The generated smooth image data is read from the smooth image data storage unit 223. And the difference part 224 calculates | requires the difference of the pixel value of each corresponding pixel between the read image data and the read smooth image data, and produces | generates difference image data.

Next, the determination unit calculates the similarity between the reference image as the inspection reference of the inspection site and the difference image data (step S208). The determination unit 225 of this embodiment calculates the similarity between the reference image read from the reference image storage unit 226 and the difference image data, for example, by a known pattern matching method.
Next, the determination unit 225 determines whether or not the similarity obtained in step S208 is greater than a predetermined value (step S209, determination step). If the determination unit 225 determines that the obtained similarity is greater than a predetermined value (step S209: YES), the determination unit 225 determines that the inspection target object 5 is a non-defective product (step S211) and ends the process. If the determination unit 225 determines that the obtained similarity is not greater than the predetermined value (step S209: NO), the determination unit 225 determines that the inspection target object 5 is a defective product (step S210) and ends the process. To do.

  As described above, the inspection apparatus 20 included in the robot apparatus 1 according to the first embodiment of the present invention includes the imaging unit 21, the smoothing unit 222, the difference unit 224, and the determination unit 225. The imaging unit 21 images the inspection target object 5 and generates image data including an inspection region that is an image region corresponding to the inspection region of the inspection target object 5. In addition, the smoothing unit 222 smoothes the pixel value of each pixel included in the image data to generate smoothed image data. The difference unit 224 obtains a difference between the pixel values of the image data pixel and the smooth image data pixel corresponding to the pixel, and generates difference image data. The determination unit 225 determines the state of the inspection region based on the reference image data and the difference image data as the inspection reference of the inspection region.

  Here, in general, the inspection apparatus obtains the external defect of the inspection target object 5 by obtaining the similarity between the image data in which the non-defective product is captured in advance and the image data in which the inspection target object 5 is captured. Judgment may be made. In this case, when there is an individual difference in the inspection target object 5 (for example, on the surface of the inspection target object 5), the similarity with the non-defective product may be low. For example, when the inspection target object 5 is cut and has a cut mark, the reflection direction of light on the surface of the inspection target object 5 may change depending on the angle at which the inspection target object 5 is imaged. In this case, the similarity between the image data obtained by imaging the inspection target object 5 and the non-defective image data may be low. As a result, even if the inspection object 5 is a non-defective product, it may not be determined that the inspection target object 5 has a defect in its outer shape.

  On the other hand, the robot apparatus 1 including the inspection apparatus 20 and the inspection apparatus 20 according to the present embodiment smoothes the pixel values of each pixel included in the image data to generate smooth image data, and the imaging unit 21 uses the pixels of the image data. And the difference of the pixel value of the pixel of the smooth image data corresponding to the said pixel is calculated | required, and difference image data is produced | generated. Thereby, the robot apparatus 1 provided with the inspection apparatus 20 of this embodiment and the inspection apparatus 20 can reduce the influence of the cutting trace of the inspection target object 5 mentioned above, for example. That is, the inspection apparatus 20 according to the present embodiment and the robot apparatus 1 including the inspection apparatus 20 can inspect for defects in the outer shape of the inspection target object with high accuracy even if there are individual differences on the surface of the inspection target object.

  In addition, the inspection apparatus 20 of the present embodiment includes a setting unit 227 that sets the cutoff frequency of the low-pass filter based on the spatial frequency of the outline image data indicating the outline portion of the examination site included in the image data. In addition, the smoothing unit 222 included in the inspection device 20 of the present embodiment generates smoothed image data by performing an operation on the image data using a low-pass filter having a cutoff frequency. Thereby, the inspection apparatus 20 can set the characteristics of the low-pass filter for generating smooth image data according to the characteristics of the image data. As a result, the inspection apparatus 20 can inspect defects on the outer shape of the inspection target object 5 with high accuracy even if there are individual differences on the surface of the inspection target object 5. In addition, since the inspection apparatus 20 performs smoothing using a known low-pass filter, it can generate smoothed image data with a simple configuration.

  In addition, the setting unit 227 included in the inspection apparatus 20 according to the present embodiment sets the cutoff frequency of the low-pass filter to a cutoff frequency lower than the frequency having the highest spatial frequency intensity of the outline image data. Thereby, the inspection apparatus 20 can generate smooth image data in which the shade of each pixel indicating the outer shape portion of the inspection site is removed and the shade of each pixel indicating the surface portion of the inspection site is not removed. . Therefore, the inspection apparatus 20 retains the shading of each pixel indicating the outer shape portion of the inspection region as difference image data obtained by calculating the difference between the smoothed image data and the image data before smoothing. It is possible to generate difference image data in which the shading of each pixel indicating the surface portion of the part is removed. Thereby, the inspection apparatus 20 can reduce the influence of the surface portion of the inspection region and inspect the outer shape portion of the inspection region with high accuracy.

[Second Embodiment]
The robot apparatus 1 which is 1st Embodiment mentioned above was the structure by which the support stand 11 was fixed and installed in the floor. The robot apparatus 1a which is 2nd Embodiment of this invention is a structure provided with the portable part 15 while the base part 14 supports the support stand 11a.
Hereinafter, the robot apparatus 1a of the present embodiment will be described with reference to FIG. In the present embodiment, the same components as those in the first embodiment are denoted by the same reference numerals, and the description thereof is omitted.

FIG. 10 is a schematic external view of a robot apparatus 1a according to the second embodiment of the present invention.
As shown in the figure, the robot apparatus 1a includes a robot main body 10a, a pedestal portion 14, and a portable portion 15.
The robot body 10 a includes a support base 11 a, a first link part 12 a and a second link part 12 b as the link part 12, and a first link part 13 a and a second link part 13 b as the link part 13.
The first link unit 13 a includes a first imaging unit 21 a as the imaging unit 21. Similarly, the second link unit 13 b includes a second imaging unit 21 b as the imaging unit 21.
The support base 11a is connected to the link portion 12 (for example, the first link portion 12a and the second link portion 12b), and supports the link portion 12.
The pedestal portion 14 supports the robot body 10a via the support base 11a, and stores the determination device 22 and the control device 30.
The portable part 15 is provided with a caster, for example, and makes the base part 14 portable. In addition, although the portable part 15 has shown the structure which moves the robot apparatus 1a on a floor surface in FIG. 10, the structure which moves a wall surface, a ceiling, etc. may be sufficient, for example.

  As described above, the robot apparatus 1a according to the second embodiment of the present invention includes the pedestal portion 14 that supports the robot body 10a and the portable portion 15 that makes the pedestal portion 14 portable. Moreover, the base part 14 of this embodiment is provided with the determination apparatus 22 and the control apparatus 30 among the inspection apparatuses 20 mentioned above inside. Thus, since the robot apparatus 1a of this embodiment incorporates the inspection apparatus 20 and the control apparatus 30 necessary for inspection, it can be reduced in size. Moreover, the robot apparatus 1a of this embodiment can perform an inspection by changing the location by the portable unit 15. For this reason, for example, the robot apparatus 1a can inspect a large inspection object 5 that is difficult to move.

  Note that the inspection apparatus 20 in the above embodiment or each unit included in the inspection apparatus 20 may be realized by dedicated hardware, or may be realized by a memory and a microprocessor. .

  Each unit included in the inspection device 20 is configured by a memory and a CPU (central processing unit), and a function for realizing the function of each unit included in the inspection device 20 is loaded into the memory and executed. It may be realized.

  In addition, by recording a program for realizing the function of each unit included in the inspection apparatus 20 on a computer-readable recording medium, causing the computer system to read and execute the program recorded on the recording medium, the inspection apparatus You may perform the process by each part with which 20 is provided. Here, the “computer system” includes an OS and hardware such as peripheral devices.

Further, the “computer system” includes a homepage providing environment (or display environment) if the WWW system is used.
The “computer-readable recording medium” refers to a storage device such as a flexible disk, a magneto-optical disk, a portable medium such as a ROM and a CD-ROM, and a hard disk incorporated in a computer system. Furthermore, a “computer-readable recording medium” dynamically holds a program for a short time, like a communication line when transmitting a program via a network such as the Internet or a communication line such as a telephone line. In this case, a volatile memory in a computer system serving as a server or a client in that case, and a program that holds a program for a certain period of time are also included. The program may be a program for realizing a part of the functions described above, and may be a program that can be realized by a combination with a program already recorded in the computer system.

  As mentioned above, although embodiment of this invention was explained in full detail with reference to drawings, the specific structure is not restricted to that embodiment, The design of the range which does not deviate from the summary of this invention, etc. are included.

DESCRIPTION OF SYMBOLS 1,1a Robot apparatus 5 Inspection object 10 Robot main body 11, 11a Support stand 12, 12-1 to 12-5 Link part 13 Link part 14 Base part 15 Transportable part 16, 16-1 to 16-6 Joint part 20 Inspection device 21 Imaging unit 30 Control device 221 Image data storage unit 222 Smoothing unit 223 Smoothing image data storage unit 224 Difference unit 225 Determination unit 226 Reference image storage unit 227 Setting unit

Claims (9)

An imaging unit that images an inspection target object and generates image data including an inspection region that is an image region corresponding to an inspection region of the inspection target object;
A smoothing unit that smoothes the pixel value of each pixel included in the image data and generates smooth image data;
A difference unit for obtaining a difference between the pixel of the image data and a pixel value of the pixel of the smoothed image data corresponding to the pixel for each pixel, and generating difference image data;
An inspection apparatus comprising: a determination unit that determines whether or not the object to be inspected is a non-defective product based on a similarity between reference image data as an inspection reference of the inspection region and the difference image data. Based on the spatial frequency of the contour image data indicating the contour portion of the examination site included in the image data, comprising a setting unit for setting the cutoff frequency of the low-pass filter,
The smoothing part is
The inspection apparatus according to claim 1, wherein the smoothed image data is generated by performing an operation on the image data using a low-pass filter having the cutoff frequency. The setting unit
The inspection according to claim 2, wherein the cut-off frequency of the low-pass filter is set by setting the spatial frequency intensity of the outline image data to a cut-off frequency lower than the strongest frequency excluding a constant component. apparatus. An imaging unit that images an inspection target object and generates image data including an inspection region that is an image region corresponding to an inspection region of the inspection target object;
A smoothing unit that smoothes the pixel value of each pixel included in the image data and generates smooth image data;
A difference unit for obtaining a difference between the pixel of the image data and a pixel value of the pixel of the smoothed image data corresponding to the pixel for each pixel, and generating difference image data;
A determination unit that determines whether or not the inspection target object is a non-defective product based on the similarity between the reference image data as the inspection reference of the inspection region and the difference image data;
A robot apparatus comprising: at least a robot body that movably supports the imaging unit. Based on the spatial frequency of the contour image data indicating the contour portion of the examination site included in the image data, comprising a setting unit for setting the cutoff frequency of the low-pass filter,
The smoothing part is
The robot apparatus according to claim 4, wherein the smoothed image data is generated by performing an operation on the image data using a low-pass filter having the cutoff frequency. The setting unit
The robot according to claim 5, wherein the cut-off frequency of the low-pass filter is set by setting the spatial frequency intensity of the outline image data to a cut-off frequency lower than the strongest frequency excluding a constant component. apparatus. A pedestal for supporting the robot body;
The robot apparatus according to claim 4, further comprising a portable unit that makes the pedestal unit portable. An imaging step of imaging an inspection target object and generating image data of an inspection target object image including an inspection area corresponding to an inspection region of the inspection target object;
A smoothing step of smoothing the pixel value of each pixel included in the image data to generate smoothed image data;
A difference step of obtaining a difference image data by obtaining a difference between a pixel value of the pixel of the image data and a pixel value of the pixel of the smoothed image data corresponding to the pixel, and
And a determination step of determining whether or not the object to be inspected is a non-defective product based on the similarity between the reference image data as the inspection reference of the inspection region and the difference image data. Computer
An imaging unit that images an inspection target object and generates image data including an inspection region that is an image region corresponding to an inspection region of the inspection target object;
A smoothing unit that smoothes the pixel value of each pixel included in the image data and generates smooth image data;
A difference unit for obtaining a difference between the pixel of the image data and a pixel value of the pixel of the smoothed image data corresponding to the pixel for each pixel, and generating difference image data;
An inspection program for functioning as a determination unit that determines whether or not the object to be inspected is a non-defective product based on a similarity between reference image data as an inspection reference of the inspection region and the difference image data.