JP6233824B1 - Image inspection apparatus, production system, image inspection method, program, and storage medium - Google Patents

Abstract

An object of the present invention is to provide an image inspection apparatus with high detection accuracy for color images, high orthogonality, fast calculation time for determination processing, easy initial setting, and low cost. . An image inspection apparatus 1 according to the present invention includes a photographing means 18 for photographing a test object as a color image 40, a mesh means 31 for dividing the color image 40 into a mesh shape, and the color image 40 having a predetermined gray scale. The imaging device which is divided into a plurality of mesh-like sections by the preprocessing means 32 for converting into the reference image, the reference image storage means, the mesh means 31 and the preprocessing means 32, and converted into the predetermined gray scale A determination unit 34 including template matching for determining the similarity based on a predetermined parameter for each of the plurality of mesh-shaped sections for the inspection image 40T and the reference image 40R photographed by the unit 18, and the interface unit 30 And. [Selection] Figure 4

Description

  The present invention relates to an image inspection apparatus that inspects the state of components mounted on a circuit board, circuit board wiring patterns, solder balls, scratches, dirt, and the like, a production system including the image inspection apparatus, an image inspection apparatus, a program, and The present invention relates to a storage medium storing the program.

  Conventionally, three-dimensional (3D) image inspection has been widely performed for circuit board mounting inspection. As a means for 3D image inspection, for example, there are those that use laser light, those that project moire light, and those that take a picture with a camera from a plurality of directions. He spent effort and time. In addition, it is often mistakenly judged to be a defective product due to slight changes in the printed state of the printed circuit board or differences in printed circuit board lots, and each time readjustment is required, and this readjustment requires a lot of man-hours. Met. Furthermore, the calculation for the image inspection becomes complicated, and there is a problem that it takes a long inspection time. In particular, in the case of multi-product small-volume production, if initial setting or the like takes time, the result of visual inspection is faster, and the advantage of automating the inspection cannot be enjoyed. In small-volume production, if a problem with the mounting machine is found in the inspection after reflow, production may have already been completed. In this case, it is effective to add an inspection device before reflow, which is a previous process, to speed up the feedback to the mounting machine, but it requires a lot of man-hours for initial setting and a high-cost inspection device. It was difficult to add.

  In order to shorten the inspection time, adoption of a two-dimensional (2D) image inspection is also being studied. As disclosed in Patent Literature 1 to Patent Literature 3, in 2D image inspection, the imaging screen is divided into a plurality of sections, and the quality of the circuit board is determined by comparing each section with a reference image. Has been proposed.

  In Patent Literature 1, a captured image of a printed circuit board is divided into a plurality of sections, and compared with reference image data stored in a storage unit independently corresponding to the sections, a printed circuit board image processing is performed. Means for determining pass / fail is disclosed. When the image is divided into sections, the position is set with reference to fiducial marks on the printed circuit board. The quality determination is performed based on whether or not the difference element between the captured image and the reference image data is extracted and the area of the difference element is equal to or less than a preset upper limit difference area. The reference image data initially comprises sample images of one or a plurality of non-defective substrates, but is added as appropriate reflecting the inspection results as the inspection proceeds.

  Japanese Patent Application Laid-Open No. 2004-228688 objectively detects macro-level abnormalities (for example, in a range that can be observed with the naked eye due to sequential changes in film formation results such as contrast based on differences in conditions in the film formation process). In addition, the comparison reference macro image and the inspection target macro image are divided into a plurality of sections in the vertical and horizontal directions, and partial areas are set. Zero average normalized cross-correlation (ZNCC: Zero-mean Normalized) is set for each partial area corresponding to each other. A means for performing image matching by cross-correlation is disclosed. In Patent Document 2, as a result of this image matching, the minimum value of each partial region is calculated as a similarity that is a quantitative evaluation value.

  In Patent Document 3, as an appearance inspection device for a conductive pattern or the like on a printed circuit board, when comparing an image of a substrate to be inspected divided into a standard pattern image divided in a corresponding manner, There has been disclosed a means capable of performing a comparison inspection even when there is a positioning error by performing comparison while relatively moving pattern images. For comparison between the image of the substrate to be inspected and the standard pattern image, a binarized signal is used for each divided section.

JP 2008-051781 A JP 2012-013644 A Japanese Patent Publication No. 06-021769

  In the image inspection apparatus described in Patent Document 1, the position is set with reference to the fiducial mark on the printed circuit board when the image is divided into each section, but the positioning error is completely eliminated. Is difficult, requires precise and expensive equipment. Further, if there is a shift in the mounting position of the component, it is determined as a defective product, and the direct rate decreases. Furthermore, since determination processing for a color image has not been sufficiently studied, it is difficult to optimize determination accuracy.

  In the image inspection apparatus described in Patent Document 2, a countermeasure for the calculation load is not disclosed, and the calculation time may be long. In addition, since determination processing for a color image has not been sufficiently studied, it is difficult to optimize determination accuracy.

  In the image inspection apparatus described in Patent Document 3, a comparative inspection is possible even if there is a positioning error. However, since a binarized signal is used, there is a problem that accurate determination processing cannot be performed on a color image. is there.

  SUMMARY OF THE INVENTION Accordingly, an object of the present invention is to provide an image inspection apparatus that has high detection accuracy for a color image, has a high direct rate, is fast in calculation for determination processing, is easy to perform initial setting, and is low in cost. There is to do.

  Another object of the present invention is to provide a production system using the image inspection apparatus.

  Another object of the present invention is to provide an image inspection method using the image inspection apparatus.

  Another object of the present invention is to provide a program for the image inspection apparatus.

  Furthermore, another object of the present invention is to provide a storage medium storing the above program.

The above object of the present invention can be achieved by the following configurations. That is, the image inspection apparatus according to the first aspect of the present invention includes a photographing unit that photographs a test object as a color image,
Mesh means for dividing the color image photographed by the photographing means into a plurality of mesh-shaped sections based on a designated mesh size;
Pre-processing means for converting the color image photographed by the photographing means into a predetermined gray scale image based on a set value;
Reference image storage means for storing the color image of the inspection object that is a preselected reference imaged by the imaging means as a reference image;
About the inspection image and the reference image that are divided into the mesh-like sections by the mesh means and the pre-processing means and that are captured by the imaging means that have been converted to the predetermined gray scale, the similarity is calculated based on Jo Tokoro parameter for each of a plurality of sections, the similarity in the predetermined search range for relatively moving the test image and the reference image for each of a plurality of sections of the mesh-like is the most A determination means including template matching for searching for a higher position ;
Interface means for setting at least one of the mesh size, the set value, and the parameter, and notifying a determination result;
It is characterized by providing.

The image inspection apparatus according to a second aspect of the present invention is the image inspection apparatus according to the first aspect, wherein the color image is divided into a plurality of mesh-shaped sections by the mesh unit, and then the pre-processing unit performs the process. It is characterized by being converted into a predetermined gray scale image .

An image inspection apparatus according to a third aspect of the present invention includes an imaging unit that captures an inspection object as a color image;
Mesh means for dividing the color image photographed by the photographing means into a plurality of mesh-shaped sections based on a designated mesh size;
Pre-processing means for converting the color image photographed by the photographing means into a predetermined gray scale image based on a set value for each of a plurality of mesh-like sections divided by the mesh means;
Reference image storage means for storing the color image of the inspection object that is a preselected reference imaged by the imaging means as a reference image;
After being divided into a plurality of mesh-like sections by the mesh means, the inspection image and the reference image photographed by the imaging means converted into the predetermined gray scale for each of the plurality of sections by the preprocessing means Determining means including template matching for determining the similarity based on a predetermined parameter for each of the plurality of mesh-shaped sections;
Interface means for setting at least one of the mesh size, the set value, and the parameter, and notifying a determination result;
It is characterized by providing.

An image inspection apparatus according to a fourth aspect of the present invention is the image inspection apparatus according to any one of the first to third aspects, wherein the inspection object is a substrate .

The image inspection apparatus according to a fifth aspect of the present invention is the image inspection apparatus according to any one of the first to fourth aspects, wherein the image inspection apparatus further includes illumination means for illuminating the inspection object .

An image inspection apparatus according to a sixth aspect of the present invention is the image inspection apparatus according to any one of the first to fifth aspects, wherein the reference image can be edited .

An image inspection apparatus according to a seventh aspect of the present invention is the image inspection apparatus according to any one of the first to sixth aspects, wherein a plurality of the reference images are used .

The image inspection apparatus according to an eighth aspect of the present invention is the image inspection apparatus according to any one of the first to seventh aspects, wherein each of the plurality of sections has an overlapping region overlapping with a surrounding section. Features.

An image inspection apparatus according to a ninth aspect of the present invention is the image inspection apparatus according to any one of the first to eighth aspects, wherein the preprocessing means determines the mixing ratio, tone curve, gradation, It is characterized in that at least one of the filters can be set .

The image inspection apparatus according to a tenth aspect of the present invention is the image inspection apparatus according to any one of the first to ninth aspects, wherein the determination means determines the search range, the overlapping amount of the sections, and the template matching according to the parameters. It is characterized in that at least one of the methods including the setting can be set .

The image inspection apparatus according to an eleventh aspect of the present invention is the image inspection apparatus according to any one of the first to tenth aspects, wherein a method including the template matching in the determination unit is set according to a state of an image in the mesh. characterized in that that.

The image inspection device according to a twelfth aspect of the present invention is the image inspection device according to any one of the first to eleventh aspects, wherein the determination means further determines the similarity based on a difference in average luminance and / or standard deviation value after graying. characterized in that it.

The image inspection apparatus according to a thirteenth aspect of the present invention is the image inspection apparatus according to any one of the first to twelfth aspects, wherein at least one of the mesh size, the set value, and the parameter can be set as a preset value. It is characterized by being.

The image inspection apparatus according to a fourteenth aspect of the present invention is the image inspection apparatus according to any one of the first to thirteenth aspects, wherein the image inspection apparatus further includes means for setting an inspection range in the inspection object. .

An image inspection apparatus according to a fifteenth aspect of the present invention is the image inspection apparatus according to the fourteenth aspect, wherein at least a part of the inspection range can be automatically set .

The image inspection apparatus according to a sixteenth aspect of the present invention is the image inspection apparatus according to the fourteenth or fifteenth aspect, wherein at least a part of the inspection range is set based on CAD data of a substrate .

According to a production system of a seventeenth aspect of the present invention, in the image inspection apparatus according to any one of the first to sixteenth aspects , at least a part of the reference image is not an image photographed by the photographing means, but a CAD of a substrate. It is generated based on data .

A production system according to an eighteenth aspect of the present invention includes the image inspection device according to any one of the first to seventeenth aspects .

An image inspection method according to a nineteenth aspect of the present invention includes a step of photographing a test object as a color image by photographing means;
Dividing the color image into a plurality of mesh-shaped sections based on a designated mesh size by mesh means;
Converting the color image photographed by the photographing means by a pre-processing means into a predetermined gray scale image based on a set value;
Storing the color image of the inspection object, which is a preselected reference, photographed by the photographing means, as a reference image;
Wherein the mesh unit and the pre-processing means is divided into a plurality of compartments of the mesh-like, and, for captured test image and the reference image by the image pickup means is converted into the predetermined gray scale, the mesh-like A step of determining similarity by determination means including template matching, searching for a position where the similarity is highest in a predetermined search range in which the reference image and the inspection image are relatively moved for each of a plurality of sections ;
Setting at least one of the mesh size, the set value, and the parameter, and notifying a determination result;
It is characterized by providing.

An image inspection method according to a twentieth aspect of the present invention includes a step of photographing a test object as a color image by photographing means;
Dividing the color image into a plurality of mesh-shaped sections based on a designated mesh size by mesh means;
For each of a plurality of mesh-like sections divided by the mesh means, converting the color image photographed by the photographing means by a preprocessing means into a predetermined gray scale image based on a set value;
Storing the color image of the inspection object, which is a preselected reference, photographed by the photographing means, as a reference image;
After being divided into a plurality of mesh-like sections by the mesh means, the inspection image and the reference image photographed by the imaging means converted into the predetermined gray scale for each of the plurality of sections by the preprocessing means Determining a degree of similarity for each of the plurality of mesh-shaped sections by a determination unit including template matching based on a predetermined parameter;
Interface means for setting at least one of the mesh size, the set value, and the parameter, and notifying a determination result;
It is characterized by providing.

A program according to a twenty-first aspect of the present invention is provided on a computer of a control device that controls the image processing apparatus or the production system so that the image inspection method according to the nineteenth or twentieth aspect is executed by the image processing apparatus or the production system. It is characterized by operating in .

A program according to a twenty-second aspect of the present invention is the program according to the twenty-first aspect, further comprising an additional add-in program .
The program according to the twenty-third aspect of the present invention is characterized in that in the program according to the twenty-first or twenty-second aspect, a modularized inspection function can be further added.
A storage medium according to a twenty-fourth aspect of the present invention stores the program according to any one of the twenty-first to twenty-third aspects.

  According to the brush of the first aspect of the present invention, since the mesh unit, the preprocessing unit, and the determination unit are provided, the detection accuracy for the color image is high, the direct rate is high, and the determination process is performed by parallel processing. Can be made faster. Further, by using a color image to be inspected as a reference selected in advance as a reference image, initial setting is easy. Furthermore, since the software is improved, an expensive apparatus is not required, so that an image inspection apparatus with a low cost can be provided.

According to the image inspection apparatus of the second or third aspect of the present invention , optimal preprocessing can be performed for each section, so that the detection accuracy is higher and the direct rate can be increased .

According to the image inspection apparatus of the fourth aspect of the present invention, it is possible to provide an image inspection apparatus that performs image inspection of a substrate .

According to the image inspection apparatus of the fifth aspect of the present invention, it is possible to improve the determination accuracy by appropriately illuminating the inspection object .

According to the image inspection apparatus of the sixth aspect of the present invention, the reference image can be optimized.

According to the image inspection apparatus of the seventh aspect of the present invention, it is possible to prevent the component misalignment from being determined as a defective product, and to increase the direct rate.

According to the image inspection apparatus of the eighth aspect of the present invention, each of the plurality of sections has an overlapping region overlapping with the surrounding sections, so that the determination accuracy near the boundary of each section can be improved.

According to the image inspection apparatus of the ninth aspect of the present invention, the determination accuracy can be improved by appropriately setting the mixing ratio, tone curve, gradation, and filter of the three primary colors in the preprocessing means.

According to the image inspection apparatus of the tenth aspect of the present invention, the determination means can appropriately improve the determination accuracy by appropriately setting the search range, the overlapping amount of the sections, and the matching method according to the parameters.

According to the image inspection apparatus of the eleventh aspect of the present invention, the determination accuracy can be improved by setting a technique including template matching according to the state of the image in the mesh.

According to the image inspection apparatus of the twelfth aspect of the present invention, the determination accuracy is further improved by further adding a similarity determination using a difference between the average luminance and / or the standard deviation value after graying to the determination means. can do.

According to the image inspection apparatus of the thirteenth aspect of the present invention, the initial setting can be further simplified by enabling the mesh size, the setting value, and the parameter to be set as preset values.

According to the image inspection apparatus of the fourteenth aspect of the present invention, the inspection range in the inspection object can be set appropriately.

According to the image inspection apparatus of the fifteenth aspect of the present invention, since the inspection range in the inspection object can be automated, the initial setting can be further simplified.

According to the image inspection apparatus of the sixteenth aspect of the present invention, since the CAD data of the substrate can be used for setting the inspection range, the initial setting of the inspection range can be performed accurately and easily.

According to the image inspection apparatus of the seventeenth aspect of the present invention, since the image generated based on the CAD data of the substrate can be used as the reference image, the initial setting man-hour can be further reduced.

According to the production system of the eighteenth aspect of the present invention, it is possible to provide a production system that exhibits the effect of the image inspection apparatus.

According to the image inspection method of the nineteenth aspect of the present invention, since the mesh unit, the preprocessing unit, and the determination unit are provided, the detection accuracy for the color image is high, the direct rate is high, and the determination process is performed by parallel processing. The operation for can be made faster. Further, by using a color image to be inspected as a reference selected in advance as a reference image, initial setting is easy. Furthermore, since the software is improved, an expensive apparatus is not required, so that an image inspection apparatus with a low cost can be provided.

According to the image inspection method of the twentieth aspect of the present invention, optimal preprocessing can be performed for each section, so that the detection accuracy is higher and the direct rate can be increased.
According to the program of the 21st aspect of this invention, the program which has the effect of the said image inspection method can be provided.

According to the program of the twenty-second aspect of the present invention, by further providing an additional add-in program, for example, correspondence to hardware communication aspects, correspondence to I / O matching, and a sequencer in a production system It is possible to appropriately deal with the communication.

According to the program of the twenty-third aspect of the present invention, by adding a modularized inspection function, in addition to the appearance inspection, for example, barcode reading, resistance color bar inspection, dimension measurement, character reading, LED lighting inspection Etc. can be easily added.

According to the storage medium of the twenty-fourth aspect of the present invention, it is possible to provide a storage medium that exhibits the effects of the program.

1 is an external perspective view of an image inspection apparatus according to a first embodiment. FIG. 2 is a transparent perspective view of FIG. 1. It is explanatory drawing of the imaging range of a camera. It is a block diagram of an image inspection apparatus. FIG. 5A is a reference image before being divided into a plurality of mesh-like sections, and FIG. 5B is a reference image after the automatic mesh is applied. 6A is a reference image in which the parameters of each mesh are being edited, and FIG. 6B is a reference image after the parameters of each mesh have been edited. It is a determination result display screen. It is a setting screen of a mesh parameter numerical value. It is a block diagram of a production system. It is an external view of the image inspection apparatus of 2nd Embodiment.

  Hereinafter, an image inspection apparatus according to an embodiment of the present invention will be described with reference to the drawings. However, the embodiment described below exemplifies an image inspection apparatus for embodying the technical idea of the present invention, and does not specify the present invention. Other embodiments are included in the scope of the claims. The present invention is equally applicable to the embodiment.

[First Embodiment]
An image inspection apparatus 1 according to a first embodiment of the present invention will be described with reference to FIGS.

  First, a schematic configuration of the image inspection apparatus 1 according to the first embodiment of the present invention will be described with reference to FIGS. 1 is an external perspective view of the image inspection apparatus 1 according to the first embodiment of the present invention, FIG. 2 is a transparent perspective view of FIG. 1, and FIG. 3 is an explanatory diagram of an imaging range of the cameras 18a to 18d. FIG. 4 is a block diagram of the image inspection apparatus 1.

<Schematic Configuration of Image Inspection Apparatus 1>
The image detection apparatus 1 is an apparatus that performs 2D image inspection, and includes an image inspection apparatus main body 10 and a personal computer main body (hereinafter referred to as “PC main body”) 25. The image inspection apparatus main body 10 has a substantially rectangular parallelepiped shape, and a substrate 20 to be inspected is mounted on the lower part of the front, and the substrate 20 take-out position (state in FIG. 2) and accommodation position (state in FIG. 1) A slide table 11 is provided that can slide between them. Further, an indicator lamp 13 is provided on the front upper portion of the image inspection apparatus main body 10 so that the determination result of the quality of the substrate can be displayed. For example, two indicator lamps 13 are provided, one of the indicator lights emits green light to indicate that the substrate is non-defective, and the other indicator light emits red light to indicate that the substrate is defective. Is displayed. On the front side of the slide table 11, a handle 12 is provided for the operator to grip the slide table 11 between the take-out position and the storage position.

  The image inspection apparatus main body 10 is provided with a stopper 23 for fixing the slide table 11 at the storage position. For example, the stopper 23 is an L-shaped member that is provided near the front of the side surface of the slide table housing portion 22, and one side of the L-shape of the stopper is loosened by a screw that can be operated by the operator. Then, it is attached to the side surface of the slide table accommodating part 22 so that it can rotate and can be fixed when the screw is tightened. Thereby, the other side of the L-shape of the stopper can be rotated 180 degrees between the engagement position for engaging the slide table 11 and the release position for releasing the engagement. When the slide table 11 on which the substrate is mounted is fixed at the storage position, the stopper 23 is fixed at the engagement position. On the other hand, when the slide table 11 is pulled out to attach or detach the substrate, the stopper 23 is set to the release position.

  On the upper surface of the slide table 11, in order to position and mount the substrate 20 to be inspected at a predetermined position, the first reference block 14 and the second reference block 15 are provided at positions corresponding to the four sides of the substrate 20, respectively. The 1st movable block 16 and the 2nd movable block 17 are provided. Notches whose cross sections perpendicular to the longitudinal direction of the blocks 14 to 17 are L-shaped are provided up to the same height above the side of the blocks 14 to 17 on which the substrate 20 abuts. The four sides of the substrate 20 are supported at a predetermined height by the notches. By providing backup pins having heights corresponding to the notches as necessary at positions corresponding to the lower surface of the substrate 20, the lower surface of the substrate 20 can be supported at a predetermined height. When the substrate 20 is thin or when there are many slits and deflection occurs, the substrate 20 can be kept horizontal by using backup pins together.

  The first reference block 14 and the second reference block 15 are fixed, and a contact surface of the first reference block 14 with the side surface of the substrate 20 and a contact surface of the second reference block 15 with the side surface of the substrate 20 in plan view. Is the origin 21 of the substrate 20. In FIG. 2, the first reference block 14 and the second reference block 15 are provided at the contact positions, but the present invention is not limited to this. For example, the present invention includes a mode in which the first reference block 14 and the second reference block 15 are separated from the origin 21. In this case, the intersection point between the extension line of the substrate contact surface of the first reference block 14 and the extension line of the substrate contact surface of the second reference block 15 in plan view is set as the origin 21 of the substrate 20.

  When the substrate 20 is mounted on the slide table 11, two sides near the origin 21 among the four sides of the substrate 20 are brought into contact with the first reference block 14 and the second reference block 15 for positioning. In this state, the first movable block 16 and the second movable block 17 are moved until they abut against the other two sides of the substrate 20, and the substrate 20 is supported by the notches provided in the blocks 14 to 17. Since the first movable block 16 and the second movable block 17 are fixed by bolts, the bolts are loosened and the first movable block 16 and the second movable block 17 are moved until they abut against the other two sides of the substrate 20. By tightening the bolts again, the substrate 20 can be fixed in a state of being positioned by the blocks 14-17. For example, a hexagon wrench can be used to tighten the bolt.

  Inside the ceiling surface of the image inspection apparatus main body 10, four fixed cameras 18 a to 18 d are provided as photographing means for photographing the substrate 20 mounted on the slide table 11. The cameras 18a to 18d capture the color image 40 of the substrate 20. In the first embodiment, four cameras 18a to 18d are provided. However, the present invention is not limited to this, and the number of cameras 18a to 18d may be one or plural, for example, Six may be sufficient. As the cameras 18a to 18d, for example, 14 million pixel USB cameras can be used. Further, as the lenses of the cameras 18a to 18d, wide-angle lenses having a deep depth of field can be used. In contrast, conventional image inspection apparatuses often employ a narrow-angle lens, but the narrow-angle lens has a shallow depth of field, so that all parts having different heights can be focused. Have difficulty. For this reason, in the case of the camera 18 using the narrow-angle lens, it is necessary to move the camera 18 in the height direction while moving the camera 18 with a three-axis robot and to shoot while focusing on each component. There are a problem that the moving mechanism becomes complicated, a problem of vibration during movement, a problem of adjusting the magnification of the obtained image, and the like. Since the cameras 18a to 18d of the first embodiment employ wide-angle lenses, a wide range can be photographed at a time, and the optical resolution is, for example, 40 μm / pixel. Further, if the focal point is set with reference to the component with the lowest height, the color image 40 in a state where the focus is achieved with respect to almost all components with the height can be obtained. Therefore, since the cameras 18a to 18d are fixed and do not require a diaphragm, the camera has no moving parts, is low in cost, has a long life, and can be downsized.

  The components mounted on the substrate 20 have some color variations. Further, due to the mounters of the standard machine 93, the variant machine 94, etc. (see FIG. 9), there may be some variation in the mounting position even when components are mounted on the board 20. In the image inspection, if the optical resolution of the cameras 18a to 18d is increased more than necessary, a very slight variation may be determined as defective, and the direct rate may be reduced. Also, for the LED illumination 19 described later, if illumination having a color reproducibility higher than necessary is used, a very slight variation is judged as defective, and the direct rate may be reduced. Therefore, in the image inspection apparatus main body 10 of the first embodiment, a general-purpose USB camera and general illumination are used as the cameras 18a to 18d and the LED illumination 19 instead of a dedicated FA product.

  Since an image photographed with a wide-angle lens may be curved in a spherical shape, it is desirable to correct image distortion by performing image processing such as Hough transform or affine transform. In addition, when the substrate 20 is replaced, it is difficult to maintain the position of the origin 21 strictly. Therefore, by extracting a plurality of feature points on the color image 40, the origin 21 of each color image 40 is determined. Calibrate the alignment.

  The four cameras 18a to 18d are arranged in two vertical rows and two horizontal rows so that, for example, the entire M-size substrate 20 having a size of 330 mm × 250 mm can be photographed without omission. As shown in FIG. 3, the imaging ranges of the cameras 18 a to 18 d are set so as to completely include the substrate 20 and include areas that overlap each other.

  Inside the image inspection apparatus main body 10 between the cameras 18a to 18d and the substrate 20, a total of four plate-type LED lights 19 are provided as illumination means, one on each of four inner walls. . The LED illumination 19 is a white planar light source, the front and back LED illuminations 19 are distributed obliquely toward the substrate 20, and the LED illuminations 19 on both sides are distributed in a direction perpendicular to the side surfaces. Yes. The LED lighting 19 provided on both side surfaces is provided at a position (a lower position) closer to the substrate 20 than the LED lighting 19 provided on the front and back surfaces. As described above, by appropriately setting the light distribution of each LED illumination 19 and controlling the luminance of each LED illumination 19, it is possible to capture color images without shadows or shine by the cameras 18a to 18d.

  The PC main body 25 is provided with a display 26, a keyboard 27, and a mouse 28 as interface means 30. The PC main body 25 is supplied with power from a commercial power supply (not shown) via a power cable, and is connected to the image inspection apparatus main body 10 via, for example, a USB cable. An image inspection program 35 is installed in the PC main body 25, and includes functions of a mesh unit 31, a preprocessing unit 32, a reference image storage unit 33, and a determination unit 34. In addition, the PC main body 25 performs setting and control of the cameras 18a to 18d, and processes captured color images. Furthermore, the PC main body 25 can shoot color images without shadows or shine by the cameras 18 a to 18 d by controlling the LED lights 19. The operator performs various settings and image inspection using the keyboard 27 and the mouse 28 while viewing the image displayed on the display 26. The determination result obtained by performing the image inspection of the substrate 20 is displayed on the display 26 and also on the indicator lamp 13. In FIG. 1, a notebook personal computer is illustrated as the PC main body 25, and in the case of the notebook personal computer, a display 26 and a keyboard 27 are provided integrally with the PC main body 25. The PC main body 25 may be, for example, a tablet terminal. In the case of a tablet terminal, the keyboard 27 and the mouse 28 can be omitted.

  The PC main body 25 desirably employs a multi-core processor having a plurality of processor cores. In a multi-core processor, each processor core is basically independent, so that each processor core can be operated without being influenced by other processor cores. By performing parallel processing with a multi-core processor, the calculation speed can be improved. The number of cores only needs to be two or more, and the larger the number of cores, the more computations can be performed at the same time, so that the computation speed can be further improved. The image inspection program 35 is programmed to cause a multi-core processor to perform parallel processing. Since the image inspection program 35 performs preprocessing and template matching for each mesh, the effect of improving the calculation speed by parallel processing is high. Further, the present invention is not limited to the multi-core processor, and any PC processor that can perform column processing can be used. For example, a multi-processor can be adopted.

  Depending on the specifications of the image inspection apparatus body 10, the specifications, the number, and the arrangement of the cameras 18 a to 18 d are different, and the specifications, the number, and the arrangement of the LED lighting 19 are different. Therefore, by performing the initial setting of the image inspection program 35 according to the specifications of the image inspection apparatus body 10, the image inspection program 35 can be shared with the image inspection apparatus body 10 having various specifications. This initial setting is preferably performed before the shipment of the image inspection apparatus 1.

<Image Inspection Procedure for Substrate 20>
Next, an image inspection procedure for the substrate 20 will be described.

  First, the reference image 40R that is the color image 40 of the non-defective substrate 20R is acquired. When the photographing start button on the non-defective substrate 20R displayed on the display 26 is clicked, the PC main body 25 and the image inspection apparatus main body 10 are in a communication state, and the PC main body 25 controls the cameras 18a to 18d and the LED illumination 19, and the camera The image data from 18a to 18d can be received. The operator prepares a non-defective substrate 20 </ b> R that has been confirmed to be non-defective in advance, and mounts the non-defective substrate 20 </ b> R on the slide table 11. When the operator grasps the handle 12 and pushes the slide table 11 to the back of the slide table housing portion 22, the power of the inspection apparatus body 10 is turned on, and the LED illumination 19 is lit. The four cameras 18a to 18d are photographed in order, that is, starting from the camera 18a, and then a predetermined area of the non-defective substrate 20R is photographed in the order of the camera b, the camera 18c, and the camera 18d. The data is transmitted to the PC main body 25 and stored in the reference image storage means 33. At the time of photographing with each camera 18a to 18d, each LED illumination 19 may be controlled so that the illumination of the portion corresponding to the imaging range of each camera 18a to 18d of the non-defective substrate 20R is optimized. When all of the four cameras 18a to 18d are photographed, the display 26 displays that the acquisition of the reference image 40R of the non-defective substrate 20R is finished. Here, the communication stop button displayed on the display 26 is clicked to stop communication between the PC main body 25 and the image inspection apparatus main body 10. Here, the mode in which the four cameras 18a to 18d photograph the non-defective substrate 20R in order has been described. However, the present invention is not limited to this, and a plurality of cameras in the four cameras 18a to 18d can simultaneously be captured. You may make it image | photograph, and you may make it all the cameras image | photograph simultaneously.

  Next, detailed inspection setting is performed on the reference image 40R of the non-defective substrate 20R using the mesh unit 31 and the preprocessing unit 32. When the reference image 40R to be set is selected from the four reference images 40R displayed on the display 26 with the mouse, the reference image 40R corresponding to the detailed setting screen is displayed.

  In the image inspection of the present invention, meshes of a predetermined pixel size are mapped over the entire inspection range, and after pre-processing each mesh, template matching between the reference image 40R and the test image 40T is performed in units of meshes. To detect defective parts. For matching, select the desired parameter from the three parameters preset for each mesh, ie, “Standard [1]”, “Sweet [2]” and “Strict [3]” In addition, it is possible to select the “Remove from inspection target [0]” setting. In the initial setting state, “automatic mesh” in which meshes with standard [1] parameters are arranged at necessary locations by automatic system judgment is applied, but in addition to this, “all sets” in which all meshes become standard parameter settings. “Clear” for removing all meshes from the inspection target can be selected, and the parameters of each mesh can be adjusted manually. For example, when all sets are selected, manually remove the unnecessary parts from the inspection target and set to [0] to complete the setting. When clear is selected, the parts that require manual inspection Set three parameters. Since the parameters can be adjusted manually in this way, initially all the inspection target locations are set to standard [1], and partially sweetened [2] or strict depending on the inspection results in actual production. Therefore, it can be changed to “3” for optimum setting.

  Regardless of the automatic mesh, all set, or clear method, the manual mesh setting method after that is the same, so here we show how to set the mesh manually after selecting the automatic mesh. 5 and FIG. 5A is a reference image 40R before being divided into a plurality of mesh-like sections, FIG. 5B is a reference image 40R after the automatic mesh is applied, and FIG. 6A is a reference image in which the parameters of each mesh are being edited. 6B is a reference image 40R after editing the parameters of each mesh.

  When the automatic mesh is applied to the reference image 40R before being divided into a plurality of mesh-like sections in FIG. 5A, the standard [1] parameter mesh is arranged at a necessary location by automatic system determination as shown in FIG. 5B. . FIG. 6A shows a state in which the display magnification is increased by operating the mouse 28. When the area to be changed is selected by operating the mouse 28 and the area is fixed, a context menu is displayed. When the standard [1], sweet [2], strict [3], or [0] parameter to be removed from the inspection target is selected by operating the mouse 28 from this menu, all the items in the selected area are selected. The mesh is changed to the selected parameter. When such manual changes are repeated and desired editing is completed, the detailed setting is completed by clicking the OK button displayed on the display 26 (see FIG. 6B). In addition to the setting using the preset value, the detailed setting of the mesh unit 31, the preprocessing unit 32, and the determination unit 34 can be changed as described later (see FIG. 8).

  Next, a substrate inspection is performed. When the photographing start button of the inspection board 20T displayed on the display 26 is clicked, the PC main body 25 and the image inspection apparatus main body 10 are in a communication state, and the PC main body 25 controls the cameras 18a to 18d and the LED illumination 19, and the camera The image data from 18a to 18d can be received. The operator mounts the inspection board 20T on the slide table 11. When the operator grasps the handle 12 and pushes the slide table 11 to the back of the slide table housing portion 22, the power of the inspection apparatus body 10 is turned on, and the LED illumination 19 is lit. The four cameras 18a to 18d are photographed in order, that is, starting from the camera 18a, and then a predetermined area of the inspection board 20T is photographed in the order of the camera b, the camera 18c, and the camera 18d. Data is transmitted to the PC main body 25. When photographing with each camera 18a to 18d, each LED illumination 19 may be controlled so that the illumination of the portion corresponding to the imaging range of each camera 18a to 18d on the inspection board 20T is optimized. In addition, although the aspect which image | photographs the test | inspection board | substrate 20T in order from 4 cameras 18a-18d was demonstrated here, this invention is not limited to this, The several camera in 4 cameras 18a-18d is simultaneous. You may make it image | photograph, and you may make it all the cameras image | photograph simultaneously.

  The four test images 40T are inspected in order. Each test image 40T is subjected to image processing by the mesh unit 31 and the preprocessing unit 32, and then the determination unit 34 performs template matching with the reference image 40R stored in the reference image storage unit 33, thereby determining pass / fail. Is done. The determination of pass / fail corresponds not only to defective parts but also to solder balls, scratches, dirt, etc., which are required for mounting inspection. When the four test images 40T are inspected and all the inspections are judged to be non-defective, the indicator lamp 13 lights in green with a sound of “ping-pong”, and the display 26 indicates that the inspection has been completed. Then, the operator is notified that the inspection board 20T is a non-defective product. The operator pulls out the slide table, replaces it with the next inspection substrate 20T, and continues the inspection. When the inspection of all the inspection substrates 20T is completed, the communication stop button displayed on the display 26 is clicked to stop the communication between the PC main body 25 and the image inspection apparatus main body 10, and the inspection is completed.

  If it is determined that any of the four test images 40T is defective in the inspection, the indicator lamp 13 is lit red together with the sound “Boobu”, and the defective part is displayed on the display 26 ( Thus, the operator is notified that the inspection board 20T may be defective, and the inspection is interrupted. A test image 40T mapped by the mesh 41 is displayed on the display 26, and a red rectangle (see FIG. 7) is displayed in the defect determination portion. Further, the test image 40T can be enlarged to a desired magnification by operating the mouse 28. Furthermore, the enlarged image of the defect determination portion can be displayed in color on the upper left of the screen by arranging the reference image 40R and the test image 40T side by side.

  At this time, when most of the inspection locations are determined to be defective, for example, the slide table 11 is not pushed in all the way, the inspection substrate 20T is mounted on the slide table 11 in the reverse direction, the first movable Since there is a possibility that the block 16 or the second movable block 17 is loose and the inspection board 20T is displaced, or the LED illumination 19 is not lit, the operator holds the inspection board 20T on the slide table 11. Re-install.

  When there are several to several tens of defective portions, the operator performs a visual inspection of the test image 40T. The screen can be enlarged by operating the mouse 28, and the test image 40T and the reference image 40R can be switched and confirmed. As another visual inspection method, it is also possible to manually change the defect determination portion to the non-defective product determination sequentially. When the test image 40T and the reference image 40R of the mesh selected by the mouse 28 are displayed side by side and magnified and displayed as non-defective, it is possible to change the determination of the mesh selected by keyboard operation to non-defective determination. it can. When this operation is performed, the next failure determination location is automatically selected. When such an operation is repeated and finally all meshes are changed to the non-defective product determination, the entire test image 40T is changed to the non-defective product determination, and the process automatically shifts to the next test image 40T inspection.

  Next, a method for optimizing the inspection setting will be described. In order to increase the straightness rate, it is necessary to review the inspection settings for parts and parts that are likely to be erroneously determined as defective in the inspection work. For example, when it is not necessary to inspect the printed portion on the surface of a component such as an aluminum electrolytic capacitor, the parameter is removed from the inspection target [0] or changed to the sweet [2] parameter. Further, for example, when the components of the reference image 40R are shifted, the reference image 40R can be partially replaced. The part to be replaced on the test image 40T is selected with the mouse 28, a context menu is displayed, and by selecting "Replace reference image", the image of the selected part on the test image 40T is associated with the reference image 40R. The part to do is rewritten. In addition, for example, for the reverse mounting of a non-polar component, since the printing direction is different from the reference image 40R, if it is erroneously determined to be defective, the printed part can be mixed with the reverse image in either direction. It can be changed so that the non-defective product is determined. A part to be mixed on the test image 40T is selected with the mouse 28, a context menu is displayed, and by selecting "Mix with reference image", the selected part on the test image 40T is a corresponding part of the reference image 40R. To be mixed. It is also possible to replace the reference image 40R with the test image 40T. In this case, one of the four reference images 40R can be replaced with the corresponding test image 40T, or all of the four reference images 40R can be replaced with the test image 40T.

  The image inspection apparatus 1 according to the first embodiment is an inspection apparatus for a substrate 20, and in addition to the inspection of the state of mounted components, the inspection of the printed wiring pattern of the substrate 20, the inspection of the waterproof and insulating coating range, and the like. Can be used. In the case of the inspection of the range of the waterproof and insulating coating, the property that the coating range of the coating agent emits light by irradiating with black light is used.

<Detailed setting of mesh parameter values>
In addition to the setting using the preset value, it is also possible to change the detailed setting of the mesh parameter numerical values related to the mesh unit 31, the preprocessing unit 32, and the determination unit 34. The mesh parameter value is set for each mesh. The detailed setting of mesh parameter values will be described with reference to FIG. FIG. 8 is a mesh parameter numerical value setting screen.

  The preset value can be selected from three parameters of standard 50, sweet 51 and strict 52, and corresponding to these parameters, pre-processing 60, matching condition 80, judgment standard 85, etc. Many mesh parameter values are set. FIG. 8 shows each mesh parameter numerical value when the standard 50 is selected as the preset value.

  The mesh size 53 represents the size of one mesh in units of pixels. If the mesh size 53 is small, the determination becomes strict, and if the mesh size 53 is large, the determination tends to be sweet. In FIG. 8, it is set to 20 pixels. The mesh size 53 is desirably 3 pixels or more. As described above, the automatic mesh 54, the entire set 55, and the clear 56 are used when the inspection target range is selected. The automatic mesh 54 is selected in the initial setting state, and a mesh at a necessary portion is set to the standard 50 by automatic determination of the system. All sets 55 set all meshes to standard 50. The clear 56 removes all meshes from the inspection target.

  The preprocessing 60 includes setting items for graying 61, contrast 70, and filter 74. In graying 61, one of RGB average 62, RGB blend 63, H67, S68, and V69 can be selected. The RGB average 62 is obtained by averaging the luminances of RGB. The RGB blend 63 is obtained by mixing the luminances of RGB at a numerical value ratio of 64 to 66 of 0 to 1. H67 is used to change the hue value to a luminance value. In step S68, the saturation value is used as a luminance field. V69 changes the intensity value to a luminance value.

  Three items of contrast 70, linear 71, S-shaped curve 72, and subtractive color 73 are defined by integers from 0 to 10, and 0 items are not used. The linear 71 is linear contrast enhancement, the S-curve 72 is contrast enhancement using an S-curve, and the color reduction 73 is contrast enhancement by a gradation subtraction method.

  Three items of the filter 74, the Gaussian 75, the median 76, and the noise reduction 77 are defined by integers from 0 to 10, and the item of 0 is not used. The Gaussian 75 is smoothing by a Gaussian filter, and reduces noise by smoothing edges. The median 76 is smoothed by a median filter and reduces granular noise. The noise reduction 77 is noise removal that maintains the outline, and reduces noise caused by blurring the image while leaving the boundary line.

  The matching condition 80 is a parameter serving as a determination condition, and includes a search range 81, an overlap amount 82, and a matching method 83. The search range 81 is a maximum allowable amount of deviation for each mesh in units of pixels. When performing template matching, the test screen 40T is moved with respect to the reference screen 40R to search for a position with the highest matching rate. The search range 81 is the maximum range for moving the test image 40T. The overlap amount 82 is an overlap amount between meshes in units of pixels. The matching method 83 selects an algorithm to be used for template matching. For example, the matching method 83 is selected from AUTO-1, AUTO-2, ZNCC, NCC, CC, SSD, and SAD. AUTO-1 automatically selects from among a plurality of algorithms used for template matching according to the contrast in the mesh, and has a strict setting. AUTO-2 automatically selects from among a plurality of algorithms used for template matching according to the contrast in the mesh, and is a standard setting. ZNCC (Zero-mean Normalized Cross-Correlation) is a zero-mean normalized cross-correlation technique. NCC is a normalized cross-correlation technique in which matching values are normalized to −1 to 1. CC (Cross Correlation) is a cross-correlation matching method in which the correlation between two images is a similarity, and SSD (Sum of Squared Difference) is the sum of the squares of the differences in luminance values of pixels at the same position. It is a technique. SAD (Sum of Absolute Difference) is a technique in which the sum of absolute values of differences in luminance values of pixels at the same position is used. ZNCC can absorb average brightness fluctuations.

  In the determination standard 85, as a determination method, three methods of matching rate (%) 86, luminance difference (%) 87, and deviation error (%) 88 are adopted (checked) and not adopted (unchecked). And, when adopting, a set value (%) can be designated. The matching rate (%) 86 is a threshold value of the matching rate of template matching, and 1.0 (100%) is a perfect matching rate. The luminance difference (%) 87 is the maximum allowable amount of the average luminance difference in the mesh. The deviation error (%) 88 is the maximum allowable value of the difference between the luminance standard deviation values in the mesh.

  Although omitted in FIG. 8, an initial value parameter setting button may be provided so that the current setting value can be set as the initial value from the next time.

<About production system 90>
A production system 90 using the image inspection apparatus 1 will be described with reference to FIG. FIG. 9 is a configuration diagram of the production system 90.

  The substrate 20 is processed in this order in the solder printer 91, the standard machine 93, the variant machine 94, and the reflow furnace 96, so that the components are soldered to the board 20. In the solder printing machine 91, paste solder is printed at the position where the components on the substrate 20 are soldered. In the standard machine 93, standard parts are mounted at predetermined positions on the substrate 20. In the variant machine 94, a variant part is mounted at a predetermined position on the substrate 20. In the reflow furnace 96, the substrate 20 is passed through a high-temperature furnace and components are soldered to the substrate 20. Examples of the heating method for the reflow furnace 96 include an infrared method, a hot air method, and a vapor phase soldering (VPS; vapor phase soldering) method.

  In the production system 90, a plurality of image inspection devices 92, 95, 97, 99 are provided after the processing devices 91, 93, 94, 96. Early feedback to 91, 93, 94, 96 is possible. A solder printing inspection device 92 is provided at the subsequent stage of the solder printer 91, a pre-reflow component mounting inspection device 95 is provided at the subsequent stage of the variant machine 94, and a 3D solder and component mounting inspection device is provided at the subsequent stage of the reflow furnace 96. 97, and finally, a visual inspection 98 by an inspector and an off-line component mounting inspection device 99 are provided. These image inspection apparatuses 92, 95, 97, and 99 need not all be provided, and the necessary ones can be selected.

  The image inspection apparatus 1 according to the first embodiment is particularly effective when provided as the pre-reflow component mounting inspection apparatus 95 and / or the offline component mounting inspection apparatus 99. In the pre-reflow component mounting inspection apparatus 95, since it is possible to determine a defect such as a component mounting failure before reflow soldering, no repair is necessary, and for the printing machine 91, the standard machine 93, and the profile machine 94, Early feedback is possible. Although it is difficult to arrange an expensive inspection device or an inspection device that requires a long time for setting as the pre-reflow component mounting inspection device 95, the image inspection device 1 of the present invention is simple in setting and low in cost. Since the detection accuracy is high and the direct rate is high, a high effect can be obtained when it is used as the pre-reflow component mounting inspection device 95. Here, when the image inspection apparatus 1 is incorporated in the production system 90 as the pre-reflow component mounting inspection apparatus 95, the substrate 20 is mounted on the slide table 11 in the image inspection apparatus 1 and the substrate 20 is removed from the slide table 11. If a device for automatically taking out is added, the production system 10 can be automated by matching the image inspection program 35 with the sequencer of the production system 10 as described later. It is also possible to arrange an operator for operating the image inspection apparatus 1 so that part of the inspection in the production system 10 is performed manually.

  In addition, if the image inspection apparatus 1 of the present invention is adopted as the off-line component mounting inspection apparatus 99, the setting is simple and the cost is low, and the detection accuracy is high and the direct rate is high. This production system has a high effect and is also effective as a confirmation inspection after substrate repair. Thereby, the burden of the visual inspection 98 by the inspector can be reduced, and the expensive and time-consuming 3D solder and component mounting inspection device 97 can be omitted.

<Operation and Effect of First Embodiment>
The image inspection apparatus main body 10 does not have a complicated configuration, the cameras 18a to 18d are fixed, and the cameras 18a to 18d do not require a diaphragm. The image inspection apparatus main body 10 can be easily manufactured and the cost can be reduced. For example, a USB camera and general illumination can be used instead of the FA camera and FA illumination. Also, a sequencer is not required for I / O, and an IO board can be substituted. With four fixed USB cameras provided with wide-angle lenses, the color image 40 can be obtained by shooting the substrate 20 only once, and the shooting of the substrate 20 can be completed in a short time. By adopting a wide-angle lens, the depth of field can be increased, so that it is not affected by the height of the components, and a wide range of the substrate 20 can be appropriately photographed. In the example of the present embodiment, a chip (0.603 chip) having a size of 0.6 mm × 0.3 mm can be inspected for an M size (330 mm × 250 mm) substrate 20, but the number and specifications of cameras are changed. As a result, it is possible to deal with substrates 20 of various sizes. In addition, by appropriately controlling the brightness of each of the four flat LED illuminations 19 provided on each of the four inner walls of the image inspection apparatus main body 10 in total, a color image free of shadows and shine is obtained. Can be acquired.

  As the PC main body 25, a PC for FA is not necessary, and a general-purpose PC can be used. If a PC equipped with a multi-core processor is employed, it is possible to greatly improve the calculation speed by enabling the image inspection program 35 to perform parallel processing. Since preprocessing and template matching are performed for each mesh, the effect of improving the computation speed by parallel processing is high.

  Since the setting of the reference image 40R is simple, it is particularly suitable for a production system that performs high-mix low-volume production. A color image 40 obtained by photographing the non-defective substrate 20R with the cameras 18a to 18d can be used as the reference image 40R. Although it is necessary to set preprocessing and template matching for each mesh, by selecting the automatic mesh 54, a standard [1] 50 parameter mesh can be arranged at a necessary location by automatic determination of the system. In addition, it is also possible to select all sets 55 in which all meshes are set to standard [1] 50 parameters, and [0] clear 56 to remove all meshes from the inspection target. A desired parameter can be selected from three preset parameters, ie, “Standard [1]”, “Sweet [2]” and “Strict [3]”. You can also select the “Remove from [0]” setting. By using the preset three types of parameters, detailed setting for each mesh can be easily performed. Further, in addition to the setting using the preset value, it is also possible to change the detailed setting of the mesh parameter values related to the mesh unit 31, the preprocessing unit 32, and the determination unit 34.

  Pre-processing is performed for each mesh, an appropriate gray scale is acquired by image processing in the pre-processing, and template matching is performed based on the gray scale. Therefore, the detection accuracy is high, the direct rate is high, and the multi-core processor performs parallel processing. By performing the process, the calculation speed for the determination process can be improved. Further, an algorithm for template matching can be appropriately selected from a plurality of types according to the state of each mesh, so that the detection accuracy is higher and the direct rate can be increased. For example, when AUTO-1 or AUTO-2 is selected, an algorithm used for template matching is automatically selected from a plurality according to the contrast in the mesh.

  It is effective to provide the image inspection apparatus 1 according to the first embodiment as the pre-reflow component mounting inspection apparatus 95 and / or the offline component mounting inspection apparatus 99 in the production system 90. In the pre-reflow component mounting inspection apparatus 95, since it is possible to determine a defect such as a component mounting failure before reflow soldering, no repair is necessary, and for the printing machine 91, the standard machine 93, and the profile machine 94, Early feedback is possible. In addition, if the image inspection apparatus 1 of the present invention is adopted as the off-line component mounting inspection apparatus 99, the setting is simple and the cost is low, and the detection accuracy is high and the direct rate is high. This production system has a high effect and is also effective as a confirmation inspection after substrate repair. Thereby, the burden of the visual inspection 98 by the inspector can be reduced, and the expensive and time-consuming 3D solder and component mounting inspection device 97 can be omitted.

[Second Embodiment]
An image inspection apparatus 1A according to a second embodiment of the present invention will be described with reference to FIG. The components common to those in FIGS. 1 to 9 are denoted by the same reference numerals, and description thereof is omitted. FIG. 10 is an external view of the image inspection apparatus 1A of the second embodiment.

  The image inspection apparatus 1A according to the second embodiment is different from the image inspection apparatus 1 according to the first embodiment in that the PC main body 25 is incorporated into the image inspection apparatus main body 10A to be an all-in-one type. A PC main body 25 is incorporated in the image inspection apparatus main body 10A, and an LCD monitor 26A is attached to the upper part of the image inspection apparatus main body 10A via a monitor arm 57. In addition, an indicator lamp 13A is provided on the front upper portion of the image inspection apparatus main body 10 so that the determination result of the quality of the substrate can be displayed. For example, three indicator lamps 13A are provided, indicating that the preparation for inspection is completed by emitting blue light, indicating that the substrate is non-defective by emitting green light, and red The light is emitted to indicate that the substrate is defective.

  As the interface means 30, in addition to the LCD monitor 26A, a wireless keyboard 27A and a wireless mouse 28A are provided. Since the configuration of the slide table 11, the stopper 23, the cameras 18a to 18d, the LED illumination, and the like is the same as that of the first embodiment, the description thereof is omitted.

  According to the image inspection apparatus 1A of the second embodiment, since it is an all-in type, the PC main body 25 and the LCD monitor 26A are incorporated in the image inspection apparatus main body 10A, and there is no need to provide a separate PC main body 25. The operator performs various settings and image inspections using the attached wireless keyboard 27A and wireless mouse 28A while viewing the image displayed on the LCD monitor 26A. Since there is no need to provide a separate PC main body 25, there is no need to connect the PC main body 25 and the image inspection apparatus main body 10A with a USB cable, and there is no need to connect a power cable for the PC main body 25 to a commercial power source. Therefore, wiring can be omitted.

[Third Embodiment]
An image inspection apparatus 1B according to a third embodiment of the present invention will be described. Components common to those in FIGS. 1 to 10 are denoted by the same reference numerals, and description thereof is omitted. The image inspection apparatus 1B according to the third embodiment uses the CAD data of the substrate 20 at the time of the automatic mesh 54 and / or the acquisition of the reference image 40R. Different from device 1.

  First, a method of using CAD data of the substrate 20 in the automatic mesh 54 will be described. In the first embodiment, when the automatic mesh 54 is selected, the mesh of the standard [1] parameter is arranged at a necessary location by automatic system determination. In the third embodiment, the CAD data of the substrate 20 is stored in the PC body 25. In addition, the CAD data of the substrate 20 can be used by the system when the system is automatically determined. In order to store the CAD data of the substrate 20 in the PC main body 25, for example, the CAD data can be read via a communication network, or the CAD data can be stored in a removable memory. Since the CAD data is used, it is possible to accurately determine a place where inspection is necessary and a place where inspection is not necessary, so that the mesh of the standard [1] parameter can be arranged more accurately. Furthermore, not only the setting of the standard [1] parameter but also the presence or absence of a specific part, the state of the wiring pattern, etc. are judged using CAD data, and the standard [1], the sweet [2] or the stricter The system may automatically determine each parameter of “3” and set each parameter to each mesh.

  Although the mesh can be manually set after the automatic mesh 54 is selected, the CAD data can be used when setting the manual mesh. For example, by replacing the reference image 40R and the image of the substrate 20 drawn by CAD data and displaying them, the standard [1], the sweet [2], or the tightening is manually performed while comparing and confirming both screens. Each parameter of “3” can be set more appropriately.

  Next, a method of using the CAD data of the substrate 20 when acquiring the reference image 40R will be described. In the first embodiment, the color image 40 obtained by photographing the non-defective substrate 20R with the cameras 18a to 18d is acquired as the reference image 40R. However, a part or all of the reference image 40R is drawn by CAD data. It is also possible to replace the image of the substrate 20 that has been changed. The part to be replaced on the image of the board 20 drawn by CAD data is selected with the mouse 28, a context menu is displayed, and the selected part on the CAD data image is selected by selecting “Replace reference image”. The corresponding portion of the reference image 40R is rewritten with the image.

  Further, when the image of the substrate 20 drawn by CAD data and the reference image 40R are to be mixed, a portion to be mixed on the CAD data image is selected with the mouse 28, and a context menu is displayed. By selecting “Mix with”, the selected portion of the CAD data image is mixed with the corresponding portion of the reference image 40R. Furthermore, the reference image 40R can be replaced with an image of CAD data. In this case, one of the four reference images 40R can be replaced with the corresponding CAD data image, or all of the four reference images 40R can be replaced with the CAD data image. . Since the reference image 40R can be made closer to an ideal image by using CAD data, it is possible to improve the detection accuracy and increase the orthogonality rate. For example, it is effective to use CAD data as the reference image 40R for the inspection of the wiring pattern of the substrate 20 or the like.

  In addition, although the difference with 1st Embodiment was demonstrated about 3rd Embodiment, the aspect of 2nd Embodiment can also be included in the image inspection apparatus 1B of 3rd Embodiment.

[Fourth Embodiment]
An image inspection apparatus 1C according to a fourth embodiment of the present invention will be described. Components common to those in FIGS. 1 to 10 are denoted by the same reference numerals, and description thereof is omitted. An image inspection apparatus 1C according to the fourth embodiment is different from the image inspection apparatus 1 according to the first embodiment in that a plurality of reference images 40R are used.

  In the image inspection apparatus 1 according to the first embodiment, since there is one reference image 40R, when there are a plurality of types of non-defective substrates 20R, the direct rate may be reduced. In the first embodiment, when a part of the reference image 40R is mixed, for example, for a reverse mount of a non-polar part, since the printing direction is different from the reference image 40R, a defect is erroneously determined. However, depending on the form of a plurality of non-defective images, it may not be possible to deal with only by mixing reference images. Therefore, the image inspection apparatus 1C according to the present embodiment includes a plurality of reference image storage means 33, and stores a plurality of reference images 40R, or at least a predetermined part of the reference image 40R. By memorizing it, the direct rate is improved. The inspection is performed based on the first reference image 40R, and the inspection is terminated when it is determined to be a non-defective product, but when it is determined to be defective, the inspection is again performed based on the second reference image 40R. Perform an inspection. As described above, by repeatedly performing the inspection based on the plurality of reference images 40R, if it is determined that any one of the reference images 40R is a non-defective product, the substrate 20 is determined to be a non-defective product, and all the reference images are displayed. By determining that the substrate 20 is defective only when it is determined that the image 40R is defective, the direct rate can be improved.

  In the case where a plurality of predetermined parts of the reference image 40R are stored, the entire image is first inspected based on the first reference image 40R, and if it is determined to be non-defective, the inspection is terminated. If it is determined to be defective, only a predetermined part of the image is inspected again based on the second reference image 40R. As described above, when a predetermined part of the image is repeatedly inspected based on the plurality of reference images 40R, and it is determined that any one of the reference images 40R is non-defective, the substrate 20 is non-defective. The straight line rate can be improved by determining that the substrate 20 is defective only when it is determined that the substrate 20 is defective for all the reference images 40R.

  Although the difference between the fourth embodiment and the first embodiment has been described, the image inspection apparatus 1C of the fourth embodiment can include the aspects of the second to third embodiments.

[Fifth Embodiment]
An image inspection apparatus 1D according to a fifth embodiment of the present invention will be described. Components common to those in FIGS. 1 to 10 are denoted by the same reference numerals, and description thereof is omitted. An image inspection apparatus 1D according to the fifth embodiment is different from the image inspection apparatus 1 according to the first embodiment in terms of the number and arrangement of cameras.

  First, a mode in which a plurality of cameras 18 arranged in a row is moved by a single-axis robot will be described. In the first embodiment, four fixed cameras 18a to 18d are used. However, when shooting a larger substrate 20, the number of cameras needs to be increased. Therefore, in order to reduce the number of cameras 18, a plurality of cameras, for example, three cameras arranged in a row at equal intervals are moved by a single-axis robot in a direction orthogonal to the arrangement direction of the cameras 18. 18 shooting ranges can be expanded. That is, after shooting with the camera 18 in the initial position, the camera 18 is moved to a predetermined position by the single-axis robot, and the shooting range by the camera 18 can be expanded by repeating this operation. Compared to the fixed cameras 18a to 18d, it is necessary to provide a movable mechanism for the camera 18 by a single-axis robot. However, since the single-axis robot has a simple structure, the number of cameras can be reduced after suppressing the increase in size and complexity of the apparatus. The cost can be reduced by reducing.

  Here, an example in which a single-axis robot capable of moving the camera 18 in the X direction (horizontal direction) has been described, but instead of the single-axis robot, the camera 18 is moved in the X direction (horizontal direction) and the Y direction (vertical direction). It is also possible to use a possible two-axis robot or a three-axis robot that can move the camera 18 in the X direction (lateral direction), the Y direction (vertical direction), and the Z direction (height direction). When a two-axis robot or a three-axis robot is employed, the number of cameras 18 is not limited to a plurality, and may be one. In addition, when a three-axis robot is employed, the camera 18 can be moved in the height direction according to the height of the component, so that an optimum image can be obtained using the fixed focus camera 18. is there.

  Next, a mode in which one camera 18 is moved by an articulated arm robot will be described. Unlike the case of the fixed camera 18 or the case where the camera 18 is moved by a one-axis, two-axis or three-axis robot, when the camera 18 is moved by an articulated arm robot, the shooting position, posture and angle by the camera 18 are changed. Since it can be set freely, it is possible to realize an image inspection similar to that of a three-dimensional image inspection apparatus in spite of being a two-dimensional image inspection apparatus 1D. Since images of desired postures and angles can be acquired at various positions, it is possible to perform an inspection with higher detection accuracy and a higher direct rate.

  Although the difference between the fifth embodiment and the first embodiment has been described, the image inspection apparatus 1D according to the fifth embodiment can include the aspects of the second to fourth embodiments.

[Sixth Embodiment]
An image inspection apparatus 1E according to a sixth embodiment of the present invention will be described. Components common to those in FIGS. 1 to 10 are denoted by the same reference numerals, and description thereof is omitted. The image inspection program 35 of the image inspection apparatus 1E according to the sixth embodiment is different from the image inspection apparatus 1 according to the first embodiment in that it further includes an additional add-in program.

  In the image inspection program 35, initial settings, mesh processing, preprocessing, template matching, and the like are all incorporated as a completed application. For this reason, the image inspection program 35 can cope with the image inspection apparatus main body 10 having different specifications by adjusting the initial setting values according to the number of cameras, the type of camera, the size of the apparatus, and the like. . When the image inspection apparatus 1E is provided with the image inspection apparatus main body 10 and the image inspection program 35 as a set, the initial setting of the image inspection program 35 is completed at the time of shipment.

  When incorporating the image inspection apparatus 1E into an existing production system, it can be dealt with by adding an add-in program to the image inspection program 35, which is a completed application. For example, an additional add-in program when matching with the communication form of the hardware of the production system, when matching with I / O of the production system, or when matching with the sequencer of the production system It is possible to cope with this.

  Although the difference between the sixth embodiment and the first embodiment has been described, the image inspection apparatus 1E according to the sixth embodiment can include the aspects of the second to fifth embodiments.

[Seventh Embodiment]
An image inspection apparatus 1F according to a seventh embodiment of the present invention will be described. Components common to those in FIGS. 1 to 10 are denoted by the same reference numerals, and description thereof is omitted. The image inspection program 35 of the image inspection apparatus 1F according to the seventh embodiment has a modular inspection function, and can select and employ only a necessary inspection function. Different from the inspection apparatus 1.

  The image inspection program 35 has a function of inspecting the appearance of the substrate 20, and this appearance inspection function is modularized. As other functions of appearance inspection, for example, barcode reading function, resistance color bar inspection function, dimension measurement function, character reading function, LED lighting inspection function, etc. are modularized. The image inspection program 35 can be incorporated. Since each inspection function is modularized, in addition to being able to select only a desired inspection function, it becomes possible to add a function that has become necessary after the introduction of the image inspection apparatus 1F.

  Although the difference between the seventh embodiment and the first embodiment has been described, the image inspection apparatus 1F of the seventh embodiment can include the aspects of the second to sixth embodiments.

[Eighth Embodiment]
An image inspection apparatus 1G according to an eighth embodiment of the present invention will be described. Components common to those in FIGS. 1 to 10 are denoted by the same reference numerals, and description thereof is omitted. The image inspection apparatus 1G according to the eighth embodiment is different from the image inspection apparatus 1 according to the first embodiment in that the image inspection apparatus 1G is applied to uses other than the inspection of the substrate 20.

  First, an aspect in which the image inspection apparatus 1G is applied to a medical site or a nursing care site will be described. In order to prevent the care recipient from falling off the bed or taking an inappropriate posture, the care recipient may be restrained on the bed with a restraint, but the posture of the care recipient is monitored at any time. If notification can be made when the posture becomes inappropriate, an effect of reducing the frequency of restraining the care recipient with the restraint can be expected. Therefore, by applying the image inspection apparatus 1G of the present invention, the camera 18 is provided so that the entire bed can be photographed, the appropriate posture of the care recipient is stored as the reference image 40R, and the care recipient's appropriate posture is stored at predetermined time intervals. The state is photographed by the camera 18 to obtain a test image 40T. Then, by determining the test image 40T by template matching with respect to the reference image 40R for each mesh, it is determined at every predetermined time interval whether the posture of the care recipient is appropriate. If it is defective, the caregiver can be notified. If the camera 18 is an infrared camera, the posture of the care recipient can be determined even when the bed is in a dark environment.

  Next, a mode in which the image inspection apparatus 1G is applied to a surveillance camera for crime prevention will be described. The present invention is applied to a security camera, provided with a camera 18 that captures the entire surveillance area, stores an image of no intruder among images captured by the camera 18 as a reference image 40R, and has a predetermined time interval. Then, the monitoring area is photographed by the camera 18 to obtain the test image 40T. Then, by determining the test image 40T with respect to the reference image 40R for each mesh by template matching, it is determined at every predetermined time interval whether there is an intruder in the monitoring area, and there is an intruder in the monitoring area. In this case, it is possible to notify the observer of the presence of the intruder. If the camera 18 is an infrared camera, the presence of an intruder can be appropriately determined even when the monitoring area is in a dark environment.

  Furthermore, an aspect in which the image inspection apparatus 1G is applied to a surveillance camera for preventing beast harm will be described. The present invention is applied to a surveillance camera for prevention of beast harm, and a camera 18 for photographing the entire surveillance area is provided, and an image in a state where no harmful beast exists from images taken by the camera 18 is used as a reference image 40R. The test area 40 is acquired by photographing the monitoring area with the camera 18 at predetermined time intervals. Then, by determining by template matching the test image 40T with respect to the reference image 40R for each mesh, it is determined at every predetermined time interval whether or not there is a pest in the monitoring area, and there is a pest in the monitoring area. In this case, it is possible to notify the observer of the presence of a pest. In addition, when it is determined that there is a pest, it is possible to drive away the pest by automatically driving the alarm means. Further, if the camera 18 is an infrared camera, the presence of a pest can be appropriately determined even when the monitoring area is dark.

  Although the difference between the eighth embodiment and the first embodiment has been described, the image inspection apparatus 1G according to the eighth embodiment can include the aspects of the second to seventh embodiments.

DESCRIPTION OF SYMBOLS 1 ... Image inspection apparatus 10 ... Image inspection apparatus main body 11 ... Slide table 12 ... Handle 13 ... Indicator lamp 14 ... 1st reference | standard block
DESCRIPTION OF SYMBOLS 15 ... 2nd reference | standard block 16 ... 1st movable block 17 ... 2nd movable block 18 ... Camera 19 ... LED illumination 20 ... Board | substrate 20R ... Non-defective board | substrate 20T ... Test board | substrate 21 ... Origin 22 ... Slide table accommodating part 23 ... Stopper 25 ... PC main body 26 ... display 27 ... keyboard 28 ... mouse 31 ... mesh means 32 ... pre-processing means 33 ... reference image storage means 34 ... determination means 35 ... image inspection program 40 ... color image 40R ... reference image 40T ... test image 41 ... mesh 50 ... Standard 51 ... Sweet 52 ... Strict 53 ... Mesh size 54 ... Automatic mesh 55 ... All set 56 ... Clear 57 ... Monitor arm 60 ... Pre-processing 61 ... Graying 62 ... RGB average 63 ... RGB blend 64-66 ... Ratio 67 ... H 68 ... S
69 ... V70 ... Contrast 71 ... Linear 72 ... S-curve 73 ... Color reduction 74 ... Filter 75 ... Gaussian 76 ... Median 77 ... Noise reduction 80 ... Matching condition 81 ... Search range 82 ... Overlap amount 83 ... Matching method 85 ... Judgment criteria 86 ... Match rate (%)
87 ... Brightness difference (%) 88 ... Deviation error (%) 90 ... Production system 91 ... Printing machine 92 ... Printing inspection device 93 ... Standard machine 94 ... Atypical machine 95 ... Pre-reflow component mounting inspection device 96 ... Reflow furnace 97 ... Parts Mounting inspection device 98 ... Visual inspection 99 ... Offline component mounting inspection device

Claims (24)

Photographing means for photographing the inspection object as a color image;
Mesh means for dividing the color image photographed by the photographing means into a plurality of mesh-shaped sections based on a designated mesh size;
Pre-processing means for converting the color image photographed by the photographing means into a predetermined gray scale image based on a set value;
Reference image storage means for storing the color image of the inspection object that is a preselected reference imaged by the imaging means as a reference image;
About the inspection image and the reference image that are divided into the mesh-like sections by the mesh means and the pre-processing means and that are captured by the imaging means that have been converted to the predetermined gray scale, the similarity is calculated based on Jo Tokoro parameter for each of a plurality of sections, the similarity in the predetermined search range for relatively moving the test image and the reference image for each of a plurality of sections of the mesh-like is the most A determination means including template matching for searching for a higher position ;
Interface means for setting at least one of the mesh size, the set value, and the parameter, and notifying a determination result;
An image inspection apparatus comprising:   2. The image inspection apparatus according to claim 1, wherein the color image is divided into a plurality of mesh-shaped sections by the mesh unit, and then converted into the predetermined gray scale image by the pre-processing unit. . Photographing means for photographing the inspection object as a color image;
Mesh means for dividing the color image photographed by the photographing means into a plurality of mesh-shaped sections based on a designated mesh size;
Pre-processing means for converting the color image photographed by the photographing means into a predetermined gray scale image based on a set value for each of a plurality of mesh-like sections divided by the mesh means;
Reference image storage means for storing the color image of the inspection object that is a preselected reference imaged by the imaging means as a reference image;
After being divided into a plurality of mesh-like sections by the mesh means, the inspection image and the reference image photographed by the imaging means converted into the predetermined gray scale for each of the plurality of sections by the preprocessing means Determining means including template matching for determining the similarity based on a predetermined parameter for each of the plurality of mesh-shaped sections;
Interface means for setting at least one of the mesh size, the set value, and the parameter, and notifying a determination result;
An image inspection apparatus comprising: It said object image inspection apparatus according to any one of claims 1-3, characterized in that a substrate. Image inspection apparatus according to any one of claims 1-4, characterized in that it comprises an illumination means and the image inspection apparatus for further illuminate the test object. Image inspection apparatus according to any one of claims 1 to 5, characterized in that editable said reference image. Image inspection apparatus according to any one of claims 1 to 6, characterized by using a plurality of the reference images. Image inspection apparatus according to any one of claims 1 to 7, characterized in that it has an overlap region overlapping with the periphery of the compartment and each of said plurality of compartments. By the setting value in the pre-processing means, the mixing ratio of the three primary colors, tone curve, images according to any one of claims 1 to 8, characterized in that at least one of the gradation and the filter can be set Inspection device. By the parameter in the determination unit, the search range, according to any one of claims 1 to 9, characterized in that it is possible to set at least one of techniques including overlap amount and the template matching of the partition Image inspection device. Said method comprising the template matching in the determination means, the image inspection apparatus according to any one of claims 1 to 10, characterized in that is set in accordance with the state of the image in the mesh. It said determining means further image inspection apparatus according to any one of claims 1 to 11, characterized in that to determine the similarity by the difference in the average luminance and / or standard deviation value after gray reduction. The mesh size, at least one, image inspection apparatus according to any one of claims 1 to 12, characterized in that it is set as a preset value of the set value and the parameters. Image inspection apparatus according to any one of claims 1 to 13, characterized in that it comprises means for setting the inspection range in the image inspecting apparatus further inspected.   The image inspection apparatus according to claim 14, wherein at least a part of the inspection range can be automatically set. The image inspection apparatus according to claim 14, wherein at least a part of the inspection range is set based on CAD data of a substrate. The at least part of the reference image is not an image photographed by the photographing unit, but is generated based on CAD data of a substrate. 17. Image inspection equipment. Production system characterized in that it comprises an image inspection apparatus according to any one of claims 1 to 17. A step of photographing a test object as a color image by photographing means;
Dividing the color image into a plurality of mesh-shaped sections based on a designated mesh size by mesh means;
Converting the color image photographed by the photographing means by a pre-processing means into a predetermined gray scale image based on a set value;
Storing the color image of the inspection object, which is a preselected reference, photographed by the photographing means, as a reference image;
Wherein the mesh unit and the pre-processing means is divided into a plurality of compartments of the mesh-like, and, for captured test image and the reference image by the image pickup means is converted into the predetermined gray scale, the mesh-like A step of determining similarity by determination means including template matching, searching for a position where the similarity is highest in a predetermined search range in which the reference image and the inspection image are relatively moved for each of a plurality of sections ;
Setting at least one of the mesh size, the set value, and the parameter, and notifying a determination result;
An image inspection method comprising: A step of photographing a test object as a color image by photographing means;
Dividing the color image into a plurality of mesh-shaped sections based on a designated mesh size by mesh means;
For each of a plurality of mesh-like sections divided by the mesh means, converting the color image photographed by the photographing means by a preprocessing means into a predetermined gray scale image based on a set value;
Storing the color image of the inspection object, which is a preselected reference, photographed by the photographing means, as a reference image;
After being divided into a plurality of mesh-like sections by the mesh means, the inspection image and the reference image photographed by the imaging means converted into the predetermined gray scale for each of the plurality of sections by the preprocessing means Determining a degree of similarity for each of the plurality of mesh-shaped sections by a determination unit including template matching based on a predetermined parameter;
Interface means for setting at least one of the mesh size, the set value, and the parameter, and notifying a determination result;
An image inspection method comprising: 21. A program that operates on a computer of a control device that controls the image processing apparatus or the production system so that the image inspection method according to claim 19 or 20 is executed by the image processing apparatus or the production system. The program according to claim 21 , further comprising an additional add-in program. The program according to claim 21 or 22 , wherein a modularized inspection function can be further added. A storage medium storing the program according to any one of claims 21 to 23 .