TW201818049A - Inspection device and inspection method of sphere surface capable of accurately performing surface inspection of a spherical object at a high speed - Google Patents

Abstract

The present invention provides an inspection device that can accurately perform surface inspection of a spherical object at a high speed. The inspection device 10 of this invention includes a rotating unit 11 for rotating the spherical object B, an illuminating unit 12 for irradiating the surface of the spherical object B with a sheet-shaped white light, an imaging unit 13 for photographing the white light irradiated on the surface of the rotating spherical object, a direction adjustment unit 14 that changes the direction of the spherical object B, a supply unit 15 that supplies the spherical object B, and a control unit 16 that controls the above-described operations and inspect the surface of the spherical object B. The control unit 16 includes an image processing unit 17 for processing a two-dimensional image obtained by the imaging unit 13, and a defect detection unit 18 for detecting defects on the surface of the spherical object B according to information obtained from the image processing unit 17.

Description

Device and method for inspecting spherical object surface

The invention relates to an inspection device and a method for inspecting the surface of a spherical object.

A golf ball is composed of an inner core (core) and a cover made of synthetic resin covering the outer periphery thereof. The outer mold is formed using a molding die that can be divided into two. For example, the following methods may be mentioned: a compression molding method in which a core is wrapped with two pre-shaped hemisphere shells and compression molding is performed using a molding mold; or a support pin is provided at a center in a cavity of the molding mold. The injection molding method in which the supporting core is supported and the molding material is injected into the gap between the inner core and the inner surface of the mold, and at the same time the support pin is drawn out and cooled. In these manufactures, defects such as flaws, burrs (bulges around dimples), and weld holes (holes caused by gas generated by the filled resin) are sometimes generated on the golf ball surface. . In addition, after forming, the burrs formed on the parting lines of the two molds are ground, but sometimes bulging (friction marks) caused by lightening of the grinding may occur, or excessive grinding may occur. Indentation (excessive friction).

The industry is seeking to automate the detection and measurement of such defects on the golf ball surface. For example, Patent Document 1 discloses an inspection device that irradiates a golf ball rotating at a constant speed with a line camera and shoots it along a line orthogonal to its direction of rotation. Based on the image data of the line camera, A two-dimensional image is acquired, and a defective portion is discriminated based on a change in brightness of the two-dimensional image. In addition, Patent Document 2 discloses an inspection device that irradiates the surface of a golf ball with laser light, and detects the positional relationship between the laser light irradiation position and the light receiving unit, thereby determining a relative change in the height of the golf ball surface. [Prior Art Literature] [Patent Literature]

[Patent Document 1] Japanese Patent Laid-Open No. 9-292349 [Patent Document 2] Japanese Patent Laid-Open No. 11-30508

[Problems to be Solved by the Invention] However, in Patent Document 1, a defective portion is detected while comparing a change in brightness in a rotation direction of a ball appearing on a two-dimensional image in a predetermined unit and converting the change. Therefore, it is difficult to detect a defect when a brightness change occurs in the same manner as a recess in a defect such as a flaw. In addition, when the sphere is irradiated with light, the irradiation angle of the light on the surface of the sphere gradually increases toward both ends of the light, and a change in brightness occurs. Furthermore, when a printed matter such as a mark is provided, the brightness changes in accordance with the printed matter. Therefore, it is difficult to accurately grasp the shape of the cavity using only the change in brightness. On the other hand, although Patent Document 2 performs a three-dimensional inspection with high measurement accuracy, it is a time-consuming inspection of the entire golf ball because it is a spot measurement. In addition, both of Patent Document 1 and Patent Document 2 detect defects in the shape of a cavity, but do not perform a color inspection of a golf ball mark or the like, and such inspection must be performed by other steps. In particular, the final inspection before packing of stains and the like is performed by an operator's visual inspection. An object of the present invention is to provide an inspection device and an inspection method that can perform a surface inspection of a spherical object at high speed and accuracy. [Means to solve the problem]

The apparatus for inspecting the surface of a spherical object according to the present invention includes: a rotating part that rotates the spherical object; an illumination part that irradiates the surface of the spherical object with a sheet-shaped white light parallel to the rotation axis; A two-dimensional image obtained by photographing the white light on the surface of the sphere object; the image processing section includes a first processing section and a second processing section; The cutting line is processed to obtain a surface height image of the spherical object, and the second processing unit is configured to process the light cutting line of the two-dimensional image to obtain a surface chromaticity image of the spherical object; And a defect detection unit having a first detection unit and a second detection unit, wherein the first detection unit detects a defect on the surface of the spherical object based on the surface height image, and the second detection unit detects a defect based on the surface color Degree image to detect defects on the surface of the sphere.

The inspection device of the present invention can simultaneously perform defect detection on the surface of the spherical object by the first detection unit and defect detection on the surface of the spherical object by the second detection unit. The first detection unit may detect defects such as a three-dimensional shape or a flaw on the surface of the spherical object using the surface height image. The second detection unit may use a surface chromaticity image to detect defects such as shapes, patterns, and colors of graphics such as marks on the surface of a sphere. In particular, since inspection data (surface height image and surface chromaticity image) suitable for each inspection are obtained for each defect detection, the defect detection can be performed accurately and quickly. Furthermore, since the two inspection data (surface height image and surface chromaticity image) for different inspections are obtained based on the same two-dimensional image, the illumination unit and the imaging unit can be shared, and the necessary inspection equipment can be used. The parts are set to a minimum. Thus, the inspection apparatus of this invention is suitable as a final inspection apparatus in the manufacturing process of a spherical object.

In the inspection device of the present invention, it is preferable that the inspection device includes a pair of rotation support portions that sandwich the spherical object from both sides. In this case, since the sphere is firmly held, the sphere can be stably rotated at a constant speed.

In the inspection device of the present invention, it is preferable that a light amount at both end portions of the sheet-like white light is larger than a light amount at a central portion of the sheet-like white light. When a spherical object is irradiated with sheet-like light, the intensity of the reflected light of the light gradually decreases as it goes toward both ends of the sheet-like white light, but by increasing the amount of light at both ends of the sheet-like white light, Reduce its brightness change to obtain a uniform level of reflected light.

The inspection device of the present invention is preferably an inspection device for inspecting a spherical object having a concave portion and / or a convex portion on the surface. In particular, the inspection device of the present invention is suitable as an inspection device for inspecting the surface of a golf ball having a plurality of recesses on the surface.

In the inspection device of the present invention, it is preferable that the surface height image is an image showing the distance between the surface of the spherical object and the reference surface in a direction perpendicular to the reference surface of the spherical object in shades. The "reference surface of a sphere" in the present invention refers to the surface of a sphere which is supposed to have no unevenness. In an image showing the distance between the spherical object surface and the reference surface in a direction perpendicular to the reference surface of the spherical object in shades, the brightness of the reflected light caused by the curved shape of the spherical object and the uneven shape of the surface is changed. Does not appear on the image. In addition, a change in the brightness of the reflected light caused by a mark or the like provided on the surface of the sphere is not displayed on the image. Therefore, the uneven shape (including defects) on the surface of the sphere can be accurately discriminated.

In the inspection device of the present invention, it is preferable that the surface chromaticity image is a plane-corrected image in which the surface of the spherical object is developed into a plane, and the chromaticity of the surface of the spherical object is displayed. By setting the surface chromaticity image as a plane-corrected image, when the marks and the like on the curved surface of the spherical object are observed from a predetermined position, the deformation of the marks and the like generated as the distance from the center is minimized. Therefore, the shape, pattern, and color of marks and the like on the surface of the spherical object can be accurately discriminated. In particular, the plane-corrected image is preferably an image that is viewed substantially vertically with respect to a reference surface of a spherical object. In this case, not only the deformation of the marks and the like due to the curved shape of the spherical object can be suppressed to a minimum, but also the deformation of the marks and the like due to the unevenness of the surface can be minimized. For example, when a printed matter such as a mark is provided in the concave portion, the influence of the concave portion can be minimized and the shape of the printed matter such as a mark can be clearly displayed on the image, so that defect detection can be performed more accurately.

In the inspection device of the present invention, it is preferable that the surface chromaticity image is an orthographic projection image of a spherical object centered on a portion to be inspected. The "orthographic image of a sphere" as used herein refers to an image obtained by projecting a sphere on a plane using a light source at infinity. By setting the surface chromaticity image as an orthographic projection image of a spherical object centered on the portion to be inspected, in the portion to be inspected, marks caused by the curved shape of the spherical object and the unevenness of the surface, etc. Small deformation. Therefore, the marks and the like on the surface of the sphere can be accurately discriminated.

The method for inspecting the surface of a spherical object according to the present invention includes the following steps: a step of rotating the spherical object; a step of irradiating the surface of the spherical object with a sheet-shaped white light parallel to a rotation axis; A step of performing a two-dimensional image obtained by photographing the white light on the surface of the object; a step of processing a light cut-off line of the two-dimensional image to obtain a surface height image of the spherical object; A step of obtaining a surface chromaticity image of the spherical object by processing the light cut line of the two-dimensional image; a step of detecting a defect on the surface of the spherical object based on the surface height image; and according to the surface color Step of detecting the defects on the surface of the sphere by taking an image. The inspection method of the present invention processes the same two-dimensional image and obtains two inspection data (surface height image and surface chromaticity image) for different inspections, so the steps can be reduced. In addition, it is possible to reduce the size and cost of the apparatus for performing the inspection.

In the inspection method of the present invention, it is preferable that the spherical object is supported from both sides by a pair of rotation support portions and the spherical object is rotated. In this case, the sphere can be stably rotated. The inspection method of the present invention that supports a spherical object with a rotation support portion and rotates the spherical object preferably includes a step of changing the direction of the spherical object after the step of obtaining a two-dimensional image. When the sphere is supported by the rotation support portion and the sphere is rotated, the supported area of the sphere cannot be inspected, and the entire surface of the sphere can be inspected smoothly by changing the direction.

In the inspection method of the present invention, preferably, the step of processing the light cut-off line of the two-dimensional image to obtain a surface height image of the spherical object is the following step: The cutting line is bent and corrected to create a light-cut straight line. Based on the light-cut straight line, height information of the surface of the sphere is obtained. Based on the height information, the vertical and horizontal directions with respect to the reference surface of the sphere are obtained. A surface height image of the distance between the surface of the sphere object and the reference surface in the direction. The “bend correction” in the present invention refers to a correction in which a curved light-cut line is adjusted to a straight line while maintaining a length. A straight line obtained by correcting a curved light cut line is referred to as a light cut straight line. That is, the light-cut line before correction and the light-cut line after correction have the same length. As for the curvature correction, in addition to the polar coordinate conversion, a known image conversion method that corrects a non-linear deformation can be used. The "height information" in the present invention refers to information indicating the position of the reference surface of the sphere and the difference in height between the reference surface at that position and the surface of the sphere (distance in the radial direction of the sphere).

In the inspection method of the present invention, preferably, the step of processing the light cut-off line of the two-dimensional image to obtain the surface chromaticity image of the spherical object is the following step: The light cut line is bent to correct the light cut line, and the chromaticity information of the surface is obtained based on the light cut line. Based on the chromaticity information, the surface of the spherical object is developed into a flat surface and the spherical object is obtained. Chromaticity of the surface corrects the image.

In the inspection method of the present invention, preferably, the step of processing the light cut-off line of the two-dimensional image to obtain a surface chromaticity image of the spherical object is the following step: A light-cut line is used to create a three-dimensional model with color information of a sphere. Based on the three-dimensional model with color information, an orthographic projection image centered on the portion to be inspected is obtained.

Next, an embodiment of the inspection device for inspecting the surface of a spherical object according to the present invention will be described with reference to the drawings. The inspection device of this embodiment is a golf ball having a recess (hereinafter referred to as a recessed portion) on the surface as an inspection object. However, the present invention is not limited to the following embodiments. Therefore, the spherical object to be inspected may be a bearing ball, an industrial plastic ball, a billiard ball, or the like, and may also have a convex shape or an uneven shape on the surface. In addition to golf, for example, soft baseball, hard baseball, volleyball, and basketball can be cited.

The inspection device 10 of FIG. 1 includes a rotating part 11 that rotates the allocated spherical object B, an illumination part 12 that irradiates the surface of the spherical object B with sheet-like white light, and an imaging unit 13 that irradiates the rotated spherical object B. The white light on the surface is used for shooting; the direction adjustment unit 14 changes the direction of the sphere B; the supply unit 15 supplies the sphere B; and the control unit 16 controls the operations of these parts and checks the surface of the sphere B. As shown in FIG. 2, the control unit 16 includes an image processing unit 17 that processes a two-dimensional image obtained by the imaging unit 13, and a detection unit that detects defects on the surface of the spherical object B based on information obtained from the image processing unit 17. Defect detection section 18.

As shown in FIGS. 3 (a) and 3 (b), the rotating part 11 rotates (rotates) the sphere B at a constant speed in a manner to ensure a shooting range A, which includes irradiation by sheet-like white light. Face L. The rotation section 11 includes a pair of rotation support sections 21a and 21b facing each other with a gap on the rotation axis Z1, a drive section 22 for rotating one of the rotation support sections 21a, and an actuator 23a and an actuator 23b. , Respectively, to move the rotation support portion 21a and the rotation support portion 21b forward and backward in the direction of the rotation axis Z1.

The rotation support portion 21a and the rotation support portion 21b are each capable of freely rotating around the rotation axis Z1 and moving freely back and forth in the direction of the rotation axis Z1 (the left and right directions in FIG. 3 (a) and FIG. 3 (b)). Be supported. In this embodiment, they are supported by a pair of pillars 25a and 25b of the frame 25, respectively. The pillars 25a and 25b are moved back and forth in the rotation axis Z1 direction by the actuators 23a and 23b, whereby the rotation support portion 21a and the rotation support portion 21b are moved back and forth in the rotation axis Z1 direction. The rotation support portion 21a and the rotation support portion 21b are cylindrical bodies having the rotation axis Z1 as a central axis, and a shaft core 21a1 and a shaft core 21b1 are provided on the rear surface. In addition, the front end inner surface 21a2 and the front end inner surface 21b2 are spherical surfaces having an inner diameter that is substantially the same as the outer diameter of the held spherical object B. Furthermore, in order not to damage the surface of the spherical object B, a soft protective surface may be provided on the front end inner surface 21a2 and the front end inner surface 21b2.

The driving portion 22 is fixed in the support column 25a and is connected to the shaft core 21a1 of one of the rotation support portions 21a. The rotation speed of the driving unit 22 is controlled by the control unit 16. The rotation speed can be appropriately selected according to the shutter speed of the imaging unit 13. The number of revolutions in one inspection is preferably set to 1 or more revolutions, more preferably 1.2 or more revolutions, and even more preferably 1.4 or more revolutions. Since the number of repetitions increases, it is preferable to set the number of turns to 2 or less, more preferably set to 1.8 times or less, and even more preferred to set it to 1.6 or less. Further, the two rotation support portions 21a and 21b may be connected to the driving portion 22, respectively. In this case, the two driving units are synchronized by the control unit 16.

The actuators 23a and 23b are fixed in the pillars 25a and 25b, respectively, and move the pillars 25a and 25b back and forth in the direction of the rotation axis Z1. This drive is synchronized by the control unit 16 and moves the pillar 25a and the pillar 25b in opposite directions, respectively. In other words, if the actuators 23a and 23b are driven in one direction, the pillars 25a and 25b are moved toward each other at the same time. If they are driven in the other direction, the pillars 25a and 25b are moved away from each other. Move at the same time. In addition, the pillar 25a and the pillar 25b may be fixed, and the rotation support portion 21a and the rotation support portion 21b may be directly moved forward and backward by the actuators 23a and 23b. The rotation support portion 21a and the rotation support portion 21b may also be moved. The shaft core 21a1 and the shaft core 21b1 extend and contract. In either case, only one of them can be moved in the direction of the rotation axis Z1. Furthermore, an elastic body can also be provided in two or one of the rotation support parts 21a and 21b so that the sphere B may be pressed against each other by the two rotation support parts 21a and 21b.

The rotating portion 11 configured as described above operates as follows. As shown in FIG. 3 (a), the spherical object B is supplied to the center of the rotating part 11 in a state where the rotating supporting part 21 a and the rotating supporting part 21 b of the rotating part 11 are separated from each other. As shown in FIG. 3 (b), when the spherical object B is supplied, the rotation support portion 21a and the rotation support portion 21b move on the rotation axis Z1 so as to approach each other, and the front end inner surface 21a2, the front end inner surface 21b2 and the spherical object B abut. Then, it is clamped by pressing the sphere B. Then, the driving portion 22 is driven in this clamped state, and the rotation support portion 21a and the rotation support portion 21b are rotated around the rotation axis Z1. That is, the spherical object B rotates. When the inspection of the spheroid B is completed and the rotation of the spheroid B is stopped, the spheroid B is supported by the direction adjustment section 14 or the supply section 15 described later, and the rotation support section 21a and the rotation support section 21b are separated from each other and returns to FIG. 3 (a )status.

The rotating part 11 holds and rotates the sphere B with the rotation support 21a and the rotation support 21b, so that the sphere B can be wound without slipping with the inner surfaces of the front ends of the rotation support 21a and the rotation support 21b. The rotation axis Z1 rotates. That is, the spherical object B can be accurately rotated at an arbitrary speed. In addition, the surface of the spherical object B is kept open except for the rotation support portion (except the region Z in FIG. 5 (b)), and the surface can be inspected in a wide area. Therefore, the degree of freedom in the arrangement of the illumination section 12 or the imaging section 13 is high. For example, in this embodiment, the lighting unit 12 and the imaging unit 13 are disposed diagonally above the device, and the diagonally upper portion of the spherical object B is set as the imaging range A (see FIGS. 3 (b) and 6 (b)). However, the illumination section 12 and the imaging section 13 may be disposed diagonally below the device, and the diagonally lower portion of the spherical object B may be set as the imaging range.

The illuminating part 12 is parallel to the rotation axis Z1 of a spherical object, and irradiates a sheet-like white light to a rotation axis. In terms of inspection accuracy, the sheet-like white light is irradiated such that the width of the irradiation surface is preferably 750 μm or less, more preferably 600 μm or less, and even more preferably 550 μm or less. On the other hand, when the imaging unit 13 photographs the irradiated surface of white light at an angle θ (see FIG. 4 (a)), and when irregularities are provided on the surface of the spherical object B, the irradiated surface cannot be seen from the imaging unit. If the width of the irradiated surface is too small, it will become a light-cut line partially cut off. Therefore, it is preferable to irradiate a sheet-like white light so that it may become 1 micrometer or more. On the other hand, for thorough inspection, the sheet-like white light is irradiated so that the length of the irradiation surface becomes the diameter of the spherical object B, preferably 60% or more, more preferably 70% or more, and even more preferably 75% or more. Furthermore, if the width of the white light is too long, it will be irradiated to the outside of the inspection section, which will cause measurement errors (noise) caused by stray light. Therefore, it is preferable that the length of the irradiation surface be the diameter of the spherical object B. The sheet-like white light is irradiated in a manner of 99% or less, more preferably 95% or less, and even more preferably 90% or less.

The sheet-like white light is preferably set so that the light amount at both ends is higher than that at the center. When white light is irradiated on the surface of the spherical object B, the irradiation angle between the surface of the spherical object B and the light gradually increases toward the two ends of the light, and the brightness of the reflected light received by the imaging unit 13 gradually increases. Decrease. That is, the brightness of both end portions of the white light entering the imaging section 13 is smaller than the brightness of the central portion of the white light. Therefore, by increasing the amount of light at both end portions of the sheet-like white light, it is possible to reduce a change in brightness caused by a change in the irradiation angle of the light due to the curvature of the surface of the spherical object B. The amount of white light may be increased continuously toward both ends, or may be increased intermittently. A method of adjusting the light amount of such white light includes, for example, a method of providing a filter for reducing the amount of light in the central portion of the lighting portion 12 and the like.

The imaging unit 13 is an area camera that photographs a part of the surface (the imaging range A, see FIG. 3 (b)) of the surface of the sphere B including the irradiation surface L of the sheet-like white light from a different angle from the illumination unit 12. The imaging unit 13 acquires a plurality of two-dimensional images that are continuously captured during a period in which the spherical object B makes one revolution. When the number of two-dimensional images is large, accurate surface height images and surface chromaticity images can be obtained. Therefore, the number of the two-dimensional images continuously captured during a period in which the sphere rotates is preferably 100 or more, more preferably 500 or more, and even more preferably 750 or more. If the number of sheets is too large, the number of processed sheets will be too large and the inspection speed will be slow. Therefore, the number of 2D images that are continuously captured during a sphere rotation is preferably 2000 or less, and more preferably 1500 or less. Especially preferred is 1250 or less. The shutter speed of the imaging unit 13 can be appropriately selected depending on the rotation speed and the number of two-dimensional images. For example, since a large amount of light is required for shooting, the shutter speed of the imaging unit 13 is preferably 100 μs or more, more preferably 500 μs or more, and even more preferably 700 μs or more. On the other hand, since image blurring becomes large, the shutter speed of the imaging unit 13 is preferably 1500 μs or less, more preferably 1000 μs or less, and even more preferably 500 μs or less.

As shown in FIG. 4 (a), the angle θ between the irradiation direction of the white light of the illumination section 12 and the imaging direction of the imaging section 13 is preferably centered on the surface (shooting range) of the spherical object B, and is preferably 30. It is more preferably 35 degrees or more, and even more preferably 40 degrees or more. When the angle θ is large, the displacement Δ of the concave portion B2 on the two-dimensional image with respect to the depth of the concave portion B2 can be increased (see FIG. 4 (b)), and the accuracy of the height information can be improved. On the other hand, if the angle θ becomes too large, the white light reflected by the unevenness on the surface of the spherical object B enters a dead angle and the light cut-off line is likely to be interrupted. In addition, the intensity of the reflected light entering the imaging unit 13 is reduced, which is preferable. In order to be 85 degrees or less, it is more preferably 80 degrees or less, and even more preferably 75 degrees or less.

Since the illuminating unit 12 and the imaging unit 13 are configured in this manner, as shown in FIG. 4 (b), if the reference surface of the sphere B is irradiated from the illuminating unit 12 with a sheet-like white light at an irradiation angle of 90 degrees, the light is used and illuminated The two-dimensional image is captured by the imaging unit 13 at an angle θ from the portion 12, and the depth of the recessed portion B2 appears as a displacement Δ on the light-cutting line of the two-dimensional image. The displacement Δ at this time and the depth H of the recessed portion B2 (the depth of the recessed portion B2 with respect to the reference surface B1 (a surface of a sphere having no recessed portion)) can be expressed by the following relational expression. Equation 1: Δ = H · sin θ Therefore, as described above, the illuminating section 12 is arranged such that the sheet-shaped white light passes through the rotation axis Z1 (the irradiation angle of the surface of the sphere B is 90 degrees), thereby increasing The displacement Δ on the light cut-off line of the two-dimensional image. In addition, the reflection intensity of the white light on the surface of the spherical object B can be increased, and the installation of the illumination section 12 is also simple.

The direction adjustment unit 14 is a portion that changes the direction of the spherical object B. Since the rotating part 11 of the inspection device 10 holds the spherical object B and rotates the spherical object B by the rotating supporting part 21a and the rotating supporting part 21b, it is not possible to contact the area in contact with the rotating supporting part 21a and the rotating supporting part 21b ( Areas of both ends of the spherical object B) Z (refer to FIG. 5 (b)) are subjected to surface inspection. Therefore, by changing the direction of the spherical object B by the direction adjustment unit 14 and performing multiple inspections, the entire surface of the spherical object B can be inspected. As shown in FIGS. 5 (a) to 5 (b), the direction adjustment section 14 includes a pair of support sections 31 a and 31 b, which are opposed to each other on the Z2 line crossing the rotation axis Z1, and the drive section 32, One of the support portions 31a is rotated; and the actuators 33a and 33b move each of the support portions 31a and 31b in the Z2 line direction. That is, the support part 31a and the support part 31b are provided in the same height as the rotation support part 21a and the rotation support part 21b (refer FIG. 1). The Z2 line is preferably orthogonal to the rotation axis Z1.

The support portion 31a and the support portion 31b are supported so as to rotate freely around the Z2 line and move back and forth freely in the direction of the Z2 line. In this embodiment, it is supported by a pair of pillars 25c and 25d of the frame 25. The pillars 25c and 25d are moved by the actuators 33a and 33b, and the support portions 31a and 31b are moved in the Z2 line direction. The support portion 31a and the support portion 31b are cylindrical bodies having the Z2 line as a central axis, and a shaft core 31a1 and a shaft core 31b1 are provided on the rear surface. In addition, the front end inner surface 31a2 and the front end inner surface 31b2 are spherical surfaces having an inner diameter substantially the same as the outer diameter of the spherical object B held. Furthermore, in order not to damage the surface of the spherical object B, a soft protective surface may be provided on the inner surface of the front end. Furthermore, in order not to damage the surface of the spherical object B, a soft protective surface may be provided on the inner surface of the front end.

The driving portion 32 is fixed in the pillar 25c and is connected to the shaft core 31a1 of one of the supporting portions 31a. The rotation angle of the driving section 32 is selected based on a range that can be inspected by rotating the spherical object B once. In terms of an excessive increase in the number of times of the direction adjustment by the direction adjustment unit 14, the rotation angle is preferably 20 degrees or more, and more preferably 45 degrees or more. On the other hand, if it is too large, the range of repetition becomes large, so it is preferably 90 degrees or less. Furthermore, when it is desired to rotate 90 degrees or more, the time can be shortened by reverse rotation. In addition, the two support portions 31a and 31b may be connected to the driving portion 32, respectively. In this case, the two driving units are synchronized by the control unit 16.

The actuators 33a and 33b are fixed in the pillars 25c and 25d, respectively, and move the pillars 25c and 25d back and forth in the Z2 line direction. As with the actuators 23a and 23b, this drive is synchronized by the control unit 16 to move the pillars 25c and 25d in opposite directions, respectively. In addition, the pillars 25c and 25d may be fixed and the support portions 31a and 31b may be directly moved, and the shaft cores 31a1 and 31b1 of the support portions 31a and 31b may be extended and contracted. Furthermore, in either case, only one of them can be moved in the direction of the Z2 line. Furthermore, an elastic body can also be provided in two or one of the support portions 31a and 31b so that the sphere B is pressed against each other by the two support portions 31a and 31b.

The direction adjustment unit 14 configured as described above operates as follows. As shown in FIG. 5 (a), in a state where the support portion 31 a and the support portion 31 b of the direction adjustment portion 14 are separated from each other, a surface inspection is performed while the spherical object B is rotated while facing the spherical object B. As shown in FIG. 5 (b), when the inspection is completed and the sphere B is stationary, the support portion 31 a and the support portion 31 b move on the Z2 axis in a manner approaching each other to clamp the sphere B. Then, after the rotation support part 21a and the rotation support part 21b of the rotation part 11 leave the sphere B, the drive part 32 drives, and the support part 31a and the support part 31b rotate around a Z2 line. That is, the spherical object B is rotated around the Z2 line, and the direction of the spherical object B is changed. After changing the direction of the sphere B, the sphere B is held by the rotation support portion 21a and the rotation support portion 21b of the rotation portion 11, and at the same time, the support portion 31a and the support portion 31b are separated from each other and leave the sphere B. Then, the inspection of the spherical object B is performed again in this direction. For a spheroid B, it is preferred to change the direction 2 to 5 times, and even more preferred to change the direction 2 to 3 times.

Since the rotation axis (line Z2) of this direction adjustment unit 14 is orthogonal to the rotation axis Z1 of the rotation unit 11, the spheres B can be transferred without mutual interference. Furthermore, the lines of the rotation axis Z1 and the rotation axis Z2 may have an angle of less than 90 degrees within a range in which their actions do not interfere with each other. In addition, since the spherical object B is held by the supporting portion 31 a and the supporting portion 31 b in the same manner as the rotating portion 11, the spherical object B can be accurately rotated.

The supply unit 15 transports the spherical object B from below the housing 25 to the inspection position. As shown in FIG. 6 (a), the supply unit 15 includes a hole 15 a provided in the center of the frame body 25, and a support portion 15 b that moves up and down in the hole 15 a. The upper surface 15b1 of the support part 15b becomes a spherical surface so that the spherical object B can be supported (refer FIG. 6 (b)).

The support portion 15b receives the spheroid B under the frame 25, and then lifts the spheroid B to the inspection position. While the sphere B is held by the rotation support portion 21a and the rotation support portion 21b, the support portion 15b moves to the standby position (see FIGS. 6 (b) and 6 (c)). When the inspection is completed, the support portion 15b rises to receive the spheroid B, descends and discharges the spheroid B out of the frame 25, and receives the next spheroid B. Furthermore, in a state where the support part 15b of the supply part 15 supports the spherical object B, the cylindrical support part 15b that moves up and down may be rotated around the shaft core. In this case, the support part 15b of the supply part 15 functions as a direction adjustment part. In this case, since the direction adjusting section 14 in FIG. 1 is not required, the side of the inspection device 10 is opened, and the degree of freedom of other components is further improved.

As shown in FIG. 2, the control unit 16 includes an image processing unit 17 and a defect detection unit 18. The image processing unit 17 includes a first processing unit 17 a and a second processing unit 17 b. The light cut-off line of the two-dimensional image is processed to obtain a surface height image of the spherical object B, and the second processing unit 17b is processed to obtain the surface chromaticity diagram of the spherical object B. For example, the defect detection unit 18 includes a first detection unit 18a and a second detection unit 18b. The first detection unit 18a detects a three-dimensional defect on the surface of the spherical object B based on a surface height image. The part 18b detects a defect of the color of the surface of the spherical object B based on the surface chromaticity image. The control unit 16 controls operations of the rotation unit 11, the illumination unit 12, the imaging unit 13, the direction adjustment unit 14, and the supply unit 15.

Next, an example of the operation of the control unit 16 is shown in FIG. 7 (a) while referring to FIGS. 6 (a) to 6 (c). After the start, the supply unit 15 is caused to supply the spherical object B to the inspection position (S1, see FIG. 6 (a)). The rotating part 11 holds the spherical object B, and the rotating part 11 rotates the spherical object B at a constant speed (S2, see FIG. 6 (b)). The illuminating part 12 is made to irradiate the surface of the rotating spherical object B with a sheet-like white light (S3, refer FIG. 6 (b)). The imaging unit 13 is configured to continuously acquire a plurality of two-dimensional images obtained by photographing the imaging range A of the surface of the spherical object B including the irradiation surface L of white light (S4, see FIG. 6 (b)). While the spherical object is being rotated, the imaging unit 13 acquires 1,000 two-dimensional images, for example. Next, the image processing unit 17 reads the two-dimensional image, and the image processing unit 17 processes the light cut line of the two-dimensional image to calculate the surface height image of the sphere B and the surface of the sphere B Chroma image (S5, see FIG. 6 (b)). Here, the two-dimensional image is preferably read in by the image processing unit 17 whenever it is acquired by the imaging unit 13. That is, it is preferable to simultaneously perform the processing of the two-dimensional image captured before in the state where the imaging unit 13 is capturing. The operations of the image processing unit 17 (step S5A using the first processing unit 17a and step S5B using the second processing unit 17b) will be described later. The defect detection unit 18 is caused to perform a surface inspection (detection of a surface defect) of the spherical object B based on the surface height image and the surface chromaticity image (S6, see FIG. 6 (b)). That is, two operations of the first detection section 18 a and the second detection section 18 b of the defect detection section 18 are performed. Preferably, they are performed simultaneously. Confirm that the check is completed and add 1 to the number of the counter, and check whether the value of the counter is N (S7). When the number of the counter is N, the supply unit 15 is caused to discharge the spherical object B from the inspection device 10 and the process ends. When the next spheroid B exists, the process returns to the beginning. On the other hand, when the number of the counter is smaller than N, the direction adjustment unit 14 is caused to change the direction of the spherical object B (S8, see FIG. 6 (c)). Then, it returns to step S2. The number N of the counter is selected based on the number of rotations of the rotating portion 11, the size (length) of the irradiation surface of the white light, the angle of the direction adjusting portion 14, and the like. It is preferably 2 to 5 times, and particularly preferably 2 to 3 times. The angle of the direction adjustment unit 14 may be changed according to the number of the counter. For example, when the rotating part 11 rotates the sphere B by 1.5 turns during an inspection, the counter rotates 60 degrees when the counter is 2, and the counter rotates 60 degrees when the counter is 3. To different angles.

Next, an operation of the image processing unit 17 (step S5) will be described. The image processing unit 17 includes operations of a first processing unit 17 a that acquires a surface height image (step S5A) and operations of a second processing unit 17 b that acquires a surface chromaticity image (step S5B).

First, step S5A will be described. The first processing unit 17a of the image processing unit 17 obtains a surface height image of the spherical object. The surface height image of the spherical object represents the surface of the spherical object in a direction perpendicular to the reference surface of the spherical object in shades. The distance from the reference surface. Specifically, the light cut line of the two-dimensional image is subjected to bending correction to create a light cut straight line, and height information of the surface of the sphere is obtained based on the light cut straight line, and a surface height image is created based on the height information.

An example of the operation sequence of step S5A is shown in FIG. 7 (b). The light cut line of the two-dimensional image is subjected to polar coordinate transformation (bend correction), and the light cut line of a substantially elliptical arc shape is converted into a light cut line of a substantially straight line (step S5A-1). The center of gravity point in the width direction of the light-cut straight line obtained by performing polar coordinate conversion is obtained (step S5A-2). Set the gray value (brightness value) according to the coordinates of the center of gravity point, and find the height-dark straight line (the height information of the irradiated surface of the sphere B) of the irradiated surface that represents the height of each X-coordinate of the light-cut line with the gray value. (Step S5A-3). Steps S5A-1 to S5A-3 are the number of two-dimensional images n (for example, 1000 times) when the spherical object is rotated at least once. Combine all this information into the height information of the surface of the sphere. Steps S5A-1 to S5A-3 are preferably performed at any time in accordance with the order of the obtained two-dimensional images. That is, it is preferable to perform the processing of the previous two-dimensional image at the same time as the operation of the previous step. However, the steps can be concentrated. The height-darkness straight lines of the respective irradiated surfaces obtained from the light cut-off lines of the two-dimensional images are connected in the shooting order to obtain a surface height image of a spherical object (step S5A-4).

Next, each operation will be described in detail. First, polar coordinate conversion is performed on the light cut-off line of the two-dimensional image, and the light cut-off line having a substantially elliptical arc shape is converted into a light-cut straight line having a substantially linear shape (step S5A-1). The light cut-off line L1 of the two-dimensional image showing the cross-sectional shape of the irradiation surface of the spheroidal object B has an elliptical arc shape. For example, FIG. 8A is a two-dimensional image when a golf ball is captured, and it can be seen that the light cut line on the image is curved into an elliptical arc shape. To further elaborate, as shown in FIG. 8 (b), the light-cutting line L1 has a shape in which a part of an elliptical circle (elliptical arc) fluctuates. In a two-dimensional image obtained by irradiating a sheet-like white light on the surface of a spherical object (hereinafter referred to as a "reference surface") which is supposed to have no unevenness, and photographing the reflected light of the illuminated surface of the spherical object, A light cutting line (hereinafter referred to as a “reference light cutting line”) of an elliptical arc having a diameter on the rotation axis Z1 of the spherical object as a long axis (see FIG. 8 (d)). Then, the coordinates of the point P of the irradiation surface in FIG. 8 (d) are transformed into a distance R from the center of the sphere B and the point P (the vertical axis (Y axis) in FIG. 8 (c)) and the connection The coordinates of the point P 'in FIG. 8 (c), which is represented by the straight line of the center of the sphere B and the point P and the angle α of the rotation axis Z1 (the horizontal axis (X axis) in FIG. 8 (c)). The reference light cut line of the arc is converted into a horizontal (parallel to the X axis) reference light cut line. The light-cutting line L1 is subjected to polar coordinate transformation substantially the same as the coordinate transformation to obtain the light-cutting straight line cL1. As shown in FIG. 9 (b), the light-cutting straight line cL1 has a shape in which a straight line fluctuates horizontally (parallel to the X axis). The image in FIG. 9 (a) is obtained by performing polar coordinate transformation on the two-dimensional image in FIG. 8 (a).

Then, the point of gravity of the light cut line cL1 in the width direction, that is, the longitudinal direction (Y-axis direction) is obtained (step S5A-2). Since the light-cut straight line fluctuates in the Y-axis direction, it can be seen that the surface of the sphere has a concave portion. However, as shown in FIG. 9 (b), the thickness of the sheet of white light appears as a band having a certain width, so the height is not clear. By finding the center of gravity point of the irradiation surface (spherical object surface), the height of the recessed portion can be clearly known. When determining the point of gravity, for a point whose X coordinate is x, as shown in FIG. 9 (c), the brightness value is plotted on the Y coordinate (the width direction of the light-cut straight line) to create a brightness value above a predetermined value. The curve S formed by connecting the points, for the area between the curve S and the predetermined brightness value cd ₁ , the Y coordinate y _g at half the area is obtained from a line segment orthogonal to the Y axis. Furthermore, it is also possible to find the center point instead of the center point. When the center point is obtained, as shown in Fig. 9 (d), for the point whose X coordinate is x, the luminance value is plotted on the Y coordinate (the width direction of the light-cut straight line), and the luminance value above the predetermined value is obtained. The Y-coordinate y _m of the average value of the Y-coordinates (the maximum value y _H and the minimum value y _L ) at both ends of the light-cut straight line. By obtaining the center point of the light cut-off line in this way, it is possible to suppress the influence of the printed mark or the like on the change in brightness. In addition, the peak of the luminance value can also be easily obtained. As shown by an imaginary line in FIG. 9 (b), a line cL _g formed by connecting the centers of gravity of the light-cutting straight lines is undulating. This line cL _g and the reference light-cut straight line are repeated except for the undulating portion of the light-cut straight line. That is, the line cL _g (height of the surface of the sphere) of the undulating portion and the displacement (Δ in FIG. 4 (b)) from the reference light cut straight line (reference surface) become the height of the recess of the sphere B. By finding the center of gravity or center point in this way, the height of each X coordinate becomes clear, which can simplify the calculation of subsequent steps.

Next, a gray value (brightness value) is set based on the coordinates of the center of gravity point, and a height-dark straight line (height information of the irradiated surface of the sphere B) of the irradiated surface representing the height of each X coordinate is obtained by the gray value (step S5A -3). The X coordinate representing the horizontal position of the light cut straight line corresponds to the X coordinate of the reference light cut straight line, and corresponds to the position of the reference surface. That is, as shown in FIG. 10 (a), the height shading straight line of the irradiation surface represents the position of the reference surface of the spherical object on the irradiation surface, and the difference between the reference surface at that position and the height of the surface of the spherical object (radius of the spherical object). Direction difference). The gray value g _x of the center of gravity point when the X coordinate is x can be expressed as follows by the Y coordinate (y _m ) of the center of gravity point. Formula 2: g _x = k ₁ · y _m + k ₂ Here, k ₁ and k _{2 are} preferably set as shown in FIG. 10 (a) in the following manner, that is, when the Y coordinate of y _m is the maximum value It becomes the maximum value (white) of the gray value, and becomes the minimum value (black) of the gray value when the Y coordinate of y _m is the minimum value. However, k ₁ and k ₂ can also be arbitrarily set. In this way, through steps S5A-1 to S5A-3, a height shading straight line of the irradiation surface (height information of the irradiation surface of the spherical object B) is obtained from a light cut line of a two-dimensional image. The high-density straight line is a straight-line image in which the irradiation surface is viewed substantially perpendicularly with respect to the reference surface, and is a straight-line image showing the distance between the surface of the irradiation surface and the reference surface in shades.

Then, the height-dark straight lines obtained from the light-cut lines of the two-dimensional images are connected in the shooting order to obtain a surface height image of the spherical object (step S5A-4). At this time, the height-shade straight line is used as the luminance value of the pixel row in the X-axis direction of the predetermined Y-coordinate, and the height-shade straight line obtained continuously is shifted little by little in the Y-axis direction and synthesized. The amount of movement in the Y-axis direction is preferably equal to the distance that the surface of the sphere moves in the rotation direction during continuous shooting. Furthermore, it is preferable to adjust the rotation amount and the shooting interval so that the shift amount becomes one pixel in the Y-axis direction. A plurality of height-dark straight lines are all connected to obtain a surface height image of the synthesized sphere B. That is, an image is obtained in which the surface of the spherical object is observed substantially perpendicular to the reference surface, and the distance between the surface of the spherical object and the reference surface is shown in shades. Fig. 10 (b) is a surface height image of a golf ball. The X-axis of the surface height image is the weft of a spherical object with the rotation axis as the ground axis, and is the irradiation surface of each light cut line. The surface height image is produced by sequentially connecting the obtained height-darkness straight lines in order, which can be calculated quickly and is preferable. However, it is also possible to combine all the height-dark straight lines.

In the surface height image of the spherical object B obtained by the first processing unit 17a (step S5A), when the Y coordinate is set to white and the Y coordinate is set to black, there is no concave portion corresponding to the spherical object B. The part of the surface (the highest part) is the brightest, and the part of the recess (the cavity of the golf ball in FIG. 10 (b)) is darkened in proportion to the depth. That is, in this surface height image, the change in the brightness of the reflected light caused by a printed object such as a curved shape of a spherical object, a shape of a concave portion on the surface, and a mark provided on the surface is not apparent. Therefore, the uneven shape (including defects) on the surface of the sphere can be accurately discriminated. As such, the surface height image is suitable for detecting defects in the three-dimensional shape of the surface of the spherical object B.

In this embodiment, a light-cut straight line obtained by performing bending correction (polar coordinate transformation) is prepared to obtain a surface height image, but the center of gravity (or center point) may be obtained without performing coordinate transformation while maintaining an elliptical arc shape. However, in this case, calculations become cumbersome. In this embodiment, as the height information of the irradiation surface, the height shading straight line of the irradiation surface is listed. For example, the position and height of the reference surface correspond to a table. As long as the relationship between the position and height of the reference surface is shown, There is no particular limitation. In this embodiment, as the surface height image of the sphere B, a three-dimensional (Three Dimensions, 3D) distance image showing the height in shades is listed, but for example, it may be obtained by drawing a 3D three-dimensional structure from a specific angle. The image of is not particularly limited as long as it can confirm the position and height of the surface of the sphere.

The second processing unit 17b of the image processing unit 17 obtains a plane-corrected image. The plane-corrected image is an image in which the surface of the sphere is expanded into a plane, and the chromaticity of the surface of the sphere is indicated. Specifically, the light cut line of the two-dimensional image is subjected to bending correction to create a light cut straight line. Based on the light cut straight line, the chromaticity information of the surface of the sphere is obtained, and a plane corrected image is obtained based on the chromaticity information. . In this embodiment, the plane-corrected image becomes a "surface chromaticity image".

An example of the operation sequence of step S5B is shown in FIG. 7 (c). The light cut line of the two-dimensional image is subjected to polar coordinate conversion (bending correction), and the light cut line of the elliptical arc shape is converted into a substantially straight light cut line (step S5B-1). A chromaticity straight line (chromaticity information of the irradiated surface of the sphere) is obtained from the light-cut straight line. The chromaticity straight line indicates the position of the reference surface of the sphere and a line connecting the position to the center of the sphere (radial direction). Line) on the surface of the sphere (step S5B-2). Steps S5B-1 to S5B-2 are the number of two-dimensional images n (for example, 1000 times) when the spherical object is rotated at least once. Moreover, steps S5B-1 to S5B-2 are preferably performed at any time in the order of the obtained two-dimensional images. That is, it is preferable to perform the processing of the previous two-dimensional image at the same time as the operation of the previous step. However, the steps can be concentrated. The chromaticity straight lines of the respective irradiated surfaces obtained from the light cut-off lines of the two-dimensional images are connected in the shooting order to obtain a plane-corrected image representing the surface of the spherical object (step S5B-3).

Next, each operation will be described in detail. First, a polar coordinate transformation is performed on the light cut-off line of the two-dimensional image, and the elliptic arc shape is transformed into a substantially linear shape (step S5B-1). The polar coordinate conversion of the light cut line is the same as that of step S5A-1.

Then, a chromaticity straight line (chromaticity information of the irradiation surface of the spherical object) is obtained from the light-cut straight line obtained by performing polar coordinate conversion (step S5B-2). As described above, the X coordinate representing the position in the horizontal direction of the light cut straight line is the same as the X coordinate of the reference light cut straight line, and corresponds to the position of the reference surface. On the other hand, for the point where the X-coordinate of the light-cutting straight line is x, a point having a predetermined brightness value or more is selected, and an average brightness value of the points is obtained. The average luminance value is obtained for each of a plurality of monochrome images constituting a two-dimensional image. The combination of single colors is preferably a combination of white colors. In particular, a combination of red, blue, and green is preferred in terms of easy availability. By obtaining the average value of the brightness of the color, the plane-corrected image obtained in the subsequent steps can be made clear. Then, the average luminance value is drawn on the reference light cut-off straight line to form a chromaticity straight line CL (chromaticity information of the irradiation surface of the spherical object). As shown in FIGS. 11 (a) to 11 (b), it can be considered that the chromaticity straight line CL is a straight line obtained by projecting the color information of the light-cut straight line cL1 onto a straight line of the same length. That is, the chromaticity straight line is an image in which the irradiation surface is viewed substantially perpendicularly with respect to the reference surface. In this way, through steps S5B-1 and S5B-2, a chromaticity straight line is obtained from a light-cut line of a two-dimensional image.

Then, the chromaticity straight lines are connected in the shooting order to obtain a plane-corrected image (step S5B-3). The continuously obtained chromaticity straight lines are shifted little by little in the Y-axis direction and synthesized. The amount of movement in the Y-axis direction is preferably equal to the distance that the surface of the sphere moves in the rotation direction during continuous shooting. Furthermore, it is preferable to adjust the rotation amount and the shooting interval so that the shift amount becomes one pixel in the Y-axis direction. By connecting all of the plurality of chromaticity straight lines, a plane-corrected image representing the surface of the spherical object B can be synthesized. Fig. 11 (b) is a plane correction image of a golf ball. The plane-corrected image is produced by connecting the obtained chromaticity straight lines gradually in order, and thus it can be calculated quickly and is preferable. However, all chromaticity straight lines can also be assembled.

Since the plane correction image of the spherical object B obtained by the second processing unit in this manner expands the curved surface of the spherical object B into a plane, there is no deformation of marks or the like caused by the curvature of the spherical object B. In addition, since it is an image viewed substantially from a direction perpendicular to the surface of the sphere, there is no deformation of the marks or the like caused by the recesses on the golf ball surface. Therefore, it is suitable for inspection of stains on the surface of the spherical object B, inspection of marks on the surface of the spherical object B, and the like.

In this embodiment, the color in the width direction of the light cut line is averaged. However, similar to step S5A-2 (see FIG. 9 (c)), the luminance value is calculated for the point where the X coordinate is x. The center of gravity point, the center of gravity point adds the brightness values of a predetermined width above and below and calculates the average brightness. By obtaining the average brightness centered on the center of gravity point in this manner, it is possible to more accurately remove deformations such as marks caused by unevenness on the golf ball surface. The reason is that by using only the color information near the light cut-off line, color information that accurately corresponds to the height position of the irradiated surface can be obtained, and unevenness can be corrected without noise. In addition, instead of averaging the colors, the light-cut straight lines may be connected in the shooting order to obtain a plane-corrected image. In this case, compared with a plane-corrected image in which chromaticity straight lines are connected, deformation due to unevenness on the surface of a sphere cannot be suppressed, but calculation can be simplified. Furthermore, in this case, the light-cut straight line itself becomes chromaticity information on the surface of the sphere.

In the plane-corrected image obtained in this manner, deformations such as marks corresponding to the curved surface of the spherical object and the unevenness of the surface do not appear. Therefore, the pattern on the surface can be more accurately discriminated, and the color defect on the surface of the spherical object B is more suitable for detection. In particular, inspection of printed marks, patterns, etc. can be performed with high accuracy. For example, when inspecting a mark or a logo printed on a golf ball having a concave portion on the surface, the deformation caused by the concave portion can be removed to perform accurate pattern matching.

Next, an operation of the defect detection unit 18 (step S6) will be described. The operation of the defect detection unit 18 includes the operation of the first detection unit 18a and the operation of the second detection unit 18b.

The first detection unit 18a detects a defect in the three-dimensional shape of the surface of the spherical object B based on the surface height image.

Next, an example of a detection method by the first detection unit 18a is given. The first detection unit 18a checks the shape of the depressions (or protrusions), the size of the depressions (or protrusions), and the interval of the depressions (or protrusions) based on the surface height image. The size, shape, and the like can be checked by comparing with a preset inspection threshold, or by comparing with a previously obtained reference sample surface height image. For example, when the inspection object is a golf ball and the surface height image is a 3D distance image obtained by gradation, a concave portion (cavity) is detected as black and other than white . The surface inspection based on the 3D distance image is preferably performed as (1) to (6) below. (1) General inspection There is no black or white part in the general inspection. In the 3D distance image (surface height image), large burrs or flaws are detected as extremely dark portions or white portions. In the case of existing 2D images, light is blocked by large burrs or flaws, or diffuse reflection occurs, which cannot be accurately detected. (2) Inspection of the gap between recesses Check whether the shape or area of the gap between recesses is abnormal. By checking the interval of the recessed portions, it is possible to detect a shallow portion of the recessed portion (for example, a friction mark) or a portion in which the recessed portion is completely buried. (3) Inspection of the shape of the concave part Check whether the shape of the concave part is deformed. Deformed recesses can be detected. (4) Inspection in the recessed portion The recessed portion that is not detected during the inspection of the recessed shape is taken as an object, and the presence of a white portion or an extremely dark portion in the recessed portion is checked. In the 3D distance image, the bulge in the concave portion is detected as a white portion in the concave portion, and the hole (for example, a fused hole) in the concave portion is detected as an extremely dark portion in the concave portion. (5) Inspection outside the recessed part Examine the presence of extremely dark or white parts for the parts other than the recessed part. In the 3D distance image, the bulge outside the concave portion is detected as an extremely white portion, and the depression outside the concave portion is detected as a black portion. (6) Inspection around the recessed portion The term "periphery of the recessed portion" refers to the area between the inside of the recessed portion and the outside of the recessed portion or the boundary area. Check the area for extremely white areas. In the 3D distance image, a bulge (for example, a burr) around the recess is detected as an extremely white portion in the region. In addition, in order to assist the bump inspection of the 3D distance image, the bump inspection can be performed with a two-dimensional (2D) chromaticity image (especially the overall inspection) at the same time as the inspection of the 3D distance image described above. .

The second detection unit 18b detects defects such as the shape and color of printed matter or marks provided on the surface of the spherical object B based on the surface chromaticity image, and further inspects the surface of the spherical object B for stains and the like. Specifically, the inspection is performed in comparison with a preset sample threshold or a sample image of a reference obtained in advance.

This inspection apparatus 10 can perform defect detection of the shape of the surface of the spherical object B and defect detection of the color of the surface of the spherical object substantially simultaneously. In addition, since the surface height image is used for defect detection of the shape of the surface of the spherical object B, the surface chromaticity image is used for defect detection of the color of the surface of the spherical object B, and inspection data suitable for each defect detection are obtained and performed. Inspection, so inspection can be performed accurately and quickly. Furthermore, since the surface height image and the surface chromaticity image for different inspections are obtained from the same two-dimensional image, the illumination unit and the imaging unit can be shared, and the inspection apparatus can be miniaturized. As described above, the inspection device 10 is most suitable as a final inspection device for a sphere (golf) B.

Next, a second embodiment (second processing unit 17b1) of the second processing unit of the inspection apparatus 10 will be described.

The second processing unit 17b1 of the image processing unit 17 obtains an orthographic projection image of a spherical object centered on a portion to be inspected. In detail, a three-dimensional model with color information is created from the light cut-off line of the two-dimensional image, and an orthographic projection image centered on the part to be inspected is obtained from the three-dimensional model with color information. . In this embodiment, the orthographic projection image becomes a "surface chromaticity image".

An example of the operation sequence of step S5B1 is shown in FIG. The coordinates (x, y, z) and chromaticity information of each point are obtained from the light cut-off line of the two-dimensional image, and a three-dimensional model with color information is obtained from these (step S5B1-1). Based on the three-dimensional model with color information, the portion to be inspected is taken as a center-oriented orthographic projection image (step S5B1-2).

Next, each operation will be described in detail. The three-dimensional coordinates and color information of each point are obtained from the light cut-off line of the two-dimensional image (step S5B1-1). The two-dimensional coordinates (x, y) and color information on the two-dimensional image are directly obtained from the light cut-off line of the two-dimensional image. On the other hand, each two-dimensional image is obtained by rotating a sphere at a certain speed and shooting at a certain shutter speed. Therefore, the two-dimensional coordinates on each two-dimensional image can be transformed based on the information directly obtained from the light-cut line of the two-dimensional image and the acquisition time of the two-dimensional image on which the light-cut line is drawn. And find the three-dimensional coordinates of the sphere. Then, based on the obtained three-dimensional coordinates and color information, a three-dimensional model with color information is obtained.

Based on the three-dimensional model with color information, the portion to be inspected is taken as a center-oriented orthographic projection image (step S5B1-2). The orthographic image of the three-dimensional model is obtained by projecting the three-dimensional model on a plane by using an infinity light source. As the orthographic image of the three-dimensional model moves away from the center, a deformation corresponding to the curvature of the sphere is generated. Therefore, an orthographic projection image is obtained so that the portion to be inspected becomes the center.

Since the ortho-projection image of the spherical object B obtained by the second processing unit 17b1 in this way is centered on the portion to be inspected, the distortion corresponding to the curved surface and unevenness of the surface of the spherical object can be minimized.

Like colored golf balls such as yellow, there are spheres having a transparent layer containing a pigment. The flake-shaped white light irradiated on the surface of such a spherical object not only irradiates the surface, but also enters the transparent layer, diffuses and reflects, and is received by the imaging unit 13. In this case, the width of the light cut line of the two-dimensional image is enlarged, and the edges also appear blurred. Therefore, it is difficult to accurately determine the center of gravity (or center point) of the light cut line and the brightness of the light cut line. average value. In the case where such a spherical object is used as an inspection object, a predetermined width is selectively cut out from a bright portion (high-brightness portion) in a width direction of a light cut line of a two-dimensional image, and other portions are cut. The brightness is set to 0 and the arithmetic processing is performed, whereby the average value of the brightness of the barycenter point (or center point) of the light cut line and the light cut line can be obtained. This method can also be applied to the inspection apparatus of any of the embodiments described above. In addition, it can also be used for the inspection of the spherical object B whose white light is reflected on the surface, which is preferable.

10‧‧‧ Inspection device

11‧‧‧Rotating part

12‧‧‧Lighting Department

13‧‧‧ Camera Department

14‧‧‧Direction adjustment section

15‧‧‧ Supply Department

15a‧‧‧hole

15b‧‧‧Support Department

15b1‧‧‧upper surface

16‧‧‧Control Department

17‧‧‧Image Processing Department

17a‧‧‧First Processing Department

17b, 17b1‧‧‧Second Processing Department

18‧‧‧ Defect Inspection Department

18a‧‧‧First detection department

18b‧‧‧Second Detection Department

21a, 21b‧‧‧rotation support

21a1, 21b1, 31a1, 31b1‧‧‧ shaft core

21a2, 21b2, 31a2, 31b2 ‧‧‧ inside of the front end

22, 32‧‧‧Driver

23a, 23b, 33a, 33b ‧‧‧ actuators

25‧‧‧Frame

25a, 25b, 25c, 25d

31a, 31b‧‧‧ Support Department

A‧‧‧ Shooting range

B‧‧‧ Sphere

B1‧‧‧ datum surface

B2‧‧‧Concave

CL‧‧‧ Chroma Straight Line

H‧‧‧ Depth

L‧‧‧ illuminated surface

L1‧‧‧light cutting line

P, P'‧‧‧ points

R‧‧‧ Distance

S‧‧‧ curve

S1 ～ S8, S5A, S5A-1 ～ S5A-4, S5B, S5B-1 ～ S5B-3, S5B1, S5B1-1, S5B1-2‧‧‧Steps

Z‧‧‧ Holding area (area)

Z1‧‧‧rotation axis

Z2‧‧‧rotation axis (line)

cL1‧‧‧light cut straight

cL _g ‧‧‧line

cd ₁ ‧‧‧brightness value

y _H , y _L , y _g , y _m ‧‧‧Y coordinates

α, θ‧‧‧ angle

Δ‧‧‧ displacement

FIG. 1 is a perspective view showing an embodiment of an inspection apparatus according to the present invention. FIG. 2 is a block diagram showing a configuration of a control unit of the inspection apparatus of FIG. 1. 3 (a) and 3 (b) are side views respectively showing a standby state of a rotating portion of the inspection device of FIG. 1 and a state of holding a spherical object. FIG. 4 (a) is a schematic diagram illustrating a relationship between an illumination section and an imaging section of the inspection apparatus of FIG. 1, and FIG. 4 (b) is a schematic diagram illustrating an optical path difference between a reference surface of a spherical object and a concave section. FIGS. 5 (a) and 5 (b) are side views respectively showing a standby state of a direction adjusting unit of the inspection device of FIG. 1 and a sandwiched state of a spherical object. FIGS. 6 (a), 6 (b), and 6 (c) are perspective views showing a spheroid supply step, a spheroid check step, and a spheroid direction adjustment step of the inspection apparatus of FIG. 1, respectively. 7 (a) is a flowchart showing an operation sequence of a control unit of the inspection device of FIG. 1, and FIG. 7 (b) is a flowchart showing an operation sequence of a first processing unit of an image processing unit of the control unit; 7 (c) is a flowchart showing an operation procedure of a second processing unit of an image processing unit of the control unit. Fig. 8 (a) is a two-dimensional image of a golf ball, Fig. 8 (b) is a schematic view of a light cut-off line, and Figs. 8 (c) and 8 (d) are schematic views showing the concept of coordinate transformation. Fig. 9 (a) is a two-dimensional image after polar coordinate transformation, Fig. 9 (b) is a schematic diagram of a light cut straight line after polar coordinate transformation, and Fig. 9 (c) is a point of gravity showing a width direction of the light cut straight line FIG. 9 (d) is a schematic diagram showing the concept of the center point in the width direction of the light-cut straight line. FIG. 10 (a) is an image showing a high-density straight line (height information of the irradiation surface), and FIG. 10 (b) is a surface height image. FIG. 11 (a) is a schematic diagram showing a method of obtaining a straightness line from a light cut straight line, and FIG. 11 (b) is a surface chromaticity image. FIG. 12 is a flowchart showing an operation procedure of a second processing unit of an inspection apparatus according to another embodiment of the present invention.

Claims (14)

An inspection device for the surface of a spherical object, comprising: a rotating part that rotates the spherical object; an illumination part that irradiates the surface of the spherical object with a sheet-like white light parallel to the rotation axis; and an imaging part that obtains light emitted from the surface of the spherical object. A two-dimensional image obtained by photographing the sheet-like white light; an image processing unit having a first processing unit and a second processing unit, wherein the first processing unit performs a light cutting line of the two-dimensional image Processing to obtain a surface height image of the spherical object, the second processing unit processing a light cut line of the two-dimensional image to obtain a surface chromaticity image of the spherical object; and a defect detection unit Having a first detection unit and a second detection unit, the first detection unit detects a defect on the surface of the spherical object based on the surface height image, and the second detection unit detects the defect based on the surface chromaticity image Detecting defects on the surface of the sphere. The inspection device for the surface of a spherical object according to item 1 of the patent application scope, wherein the rotating portion has a pair of rotation supporting portions that sandwich the spherical object from both sides. The inspection device for a spherical object surface according to item 1 or item 2 of the patent application scope, wherein the light amount at both ends of the sheet-like white light is greater than the light amount at the center of the sheet-like white light. The inspection device for the surface of a spherical object according to any one of claims 1 to 3, wherein the spherical object has a concave portion and / or a convex portion on the surface. The inspection device for the surface of a spherical object according to item 4 of the scope of patent application, wherein the spherical object is a golf ball having a plurality of recesses on the surface. The inspection device for a spherical object surface according to any one of claims 1 to 5, wherein the surface height image is a direction perpendicular to the reference surface of the spherical object in shades. An image of the distance between the surface of the sphere and the reference surface. The inspection device for the surface of a spherical object according to any one of claims 1 to 6, wherein the surface chromaticity image is a surface of the spherical object that is unfolded into a plane, and indicates the spherical object. Chromaticity of the surface corrects the image. The inspection device for a spherical object surface according to any one of claims 1 to 6, wherein the surface chromaticity image is an orthographic projection of the spherical object with the part to be inspected as a center. image. A method for inspecting the surface of a spherical object includes: a step of rotating the spherical object; a step of irradiating a surface of the spherical object with a sheet-shaped white light parallel to a rotation axis; and obtaining the sheet-shaped white light irradiated on the surface of the spherical object A step of photographing a two-dimensional image obtained by light; a step of processing a light cut-off line of the two-dimensional image to obtain a surface height image of the spherical object; a light section of the two-dimensional image A step of processing a broken line to obtain a surface chroma image of the spheroid; a step of detecting a defect on the surface of the spheroid based on the surface height image; and detecting the chroma image based on the surface chroma image Describe the steps on the surface of the sphere. The method for inspecting the surface of a spherical object according to item 9 of the scope of patent application, wherein the step of rotating the spherical object is to support the spherical object from both sides by a pair of rotation support portions and rotate the spherical object at a constant speed. A step of. The method for inspecting the surface of a spherical object according to item 10 of the scope of patent application, includes the step of changing the direction of the spherical object. The method for inspecting the surface of a spherical object according to any one of claims 9 to 11, wherein the light cut-off line of the two-dimensional image is processed to obtain a surface height map of the spherical object. The step of the image is the following steps: bending correction is performed on the light cut line of the two-dimensional image to create a light cut straight line, and the height information of the surface of the spherical object is obtained based on the light cut straight line, and according to the The height information is used to obtain a surface height image representing the distance between the surface of the sphere and the reference surface in a direction perpendicular to the reference surface of the sphere in shades. The method for inspecting the surface of a spherical object according to any one of claims 9 to 12, wherein the light cut-off line of the two-dimensional image is processed to obtain the surface chromaticity of the spherical object. In the image step, the light cut line of the two-dimensional image is subjected to bending correction to create a light cut straight line, and the chromaticity information of the surface of the spherical object is obtained based on the light cut straight line. The chromaticity information is used to obtain the plane-corrected image obtained by expanding the surface of the spherical object into a plane and indicating the chromaticity of the surface of the spherical object. The method for inspecting the surface of a spherical object according to any one of claims 9 to 12, wherein the light cut-off line of the two-dimensional image is processed to obtain the surface chromaticity of the spherical object. The step of the image is the following step: a three-dimensional model with color information of the sphere is made according to the light cut-off line of the two-dimensional image, and the three-dimensional model with color information is obtained for inspection The orthographic image is centered on the part.