Fluorescent penetrant inspection equipment

Abstract

To prevent false color in an imaging image from occurring and specify the position of a crack on a surface to be inspected.SOLUTION: A camera 40 being a monochrome camera performs first and second imagings in the state of fixing a relative position, a relative angle and a magnification to a surface 51 to be inspected. The first imaging performs imaging not through an ultraviolet cut filter 48; the second imaging performs imaging through the ultraviolet cut filter 48; and a CPU 11 (a processor) extracts an area exceeding a predetermined brightness threshold value as a fluorescent area 60 from an image by the second imaging, and superimposes the fluorescent area 60 on an image by the first imaging.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] Disclosed herein is a fluorescent penetrant inspection device.

[0002] Patent Document 1 discloses a surface defect inspection device. This defect inspection device performs fluorescent penetrant inspection. An inspection surface that has been subjected to a penetrant treatment using a luminescent penetrant is irradiated with ultraviolet light. An image of the inspection surface is captured by an image sensor. The image data of the image sensor is digitally converted into RGB data. The image data is then displayed as a color image. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Publication No. 2018-179621 Summary of the Invention [Problem to be solved by the invention]

[0004] In digital cameras, color filters are arranged on the image sensor to capture color images. For example, red, green, and blue filter elements are arranged in a Bayer pattern.

[0005] For example, a green or blue filter element is placed next to a red filter element. Where the green or blue filter element is placed, no red pixel value is detected. Therefore, the pixel value of the pixel sandwiched between the filter elements of the same color is estimated based on the pixel values ​​of the adjacent filter elements of the same color. This estimation is called an interpolation process.

[0006] For example, suppose that one of two adjacent red filter elements has a pixel value of 255 and the other has a pixel value of 0. In this case, 127 is calculated as the red pixel value of the area sandwiched between these filter elements.

[0007] In high-contrast areas, a pixel value of 255 may be adjacent to a pixel value of 0. In such cases, the red pixel value of 127 calculated by interpolation results in a so-called false color. For example, false color streaks appear along the edges of objects in a captured image.

[0008] Here, among the constituent colors of the color filter, red, green, and blue, the false green color may be similar to the fluorescent color in fluorescent penetrant flaw detection. In other words, the false color may cause an erroneous determination that a flaw is present.

[0009] Therefore, this specification discloses a fluorescent penetrant flaw detection device that can suppress the occurrence of false colors in captured images and identify the location of flaws on an inspected surface. [Means for solving the problem]

[0010] This specification discloses a fluorescent penetrant flaw detection device. The device includes a camera, a flaw detection light, an ultraviolet cutoff filter, and a processor. The camera captures an image of the surface to be inspected. The search light irradiates the surface to be inspected with ultraviolet light. The ultraviolet cutoff filter can be positioned on the optical axis of the camera. The processor processes the image captured by the camera. The camera is a monochrome camera. The camera performs first and second image capture with a fixed relative position, relative angle, and magnification with respect to the surface to be inspected. The first image capture is performed without using the ultraviolet cutoff filter. The second image capture is performed using the ultraviolet cutoff filter. The processor extracts, from the image captured by the second image, an area exceeding a predetermined brightness threshold as a fluorescent area. The processor also superimposes the fluorescent area on the image captured by the first image.

[0011] According to the above configuration, the surface to be inspected is imaged using a monochrome camera. This eliminates the need for interpolation, thereby suppressing false colors. When the surface to be inspected is imaged using a monochrome camera, it becomes difficult to distinguish between the reflected light from the searchlight and the fluorescence emitted from the flaw detection agent. Therefore, a first image is taken without using an ultraviolet-cut filter, and a second image is taken using an ultraviolet-cut filter. The first image allows the surface shape of the surface to be inspected to be visible. Furthermore, the second image allows the fluorescent area on the surface to be identified. Furthermore, by superimposing the fluorescent area on the first image, it is possible to determine where the fluorescent area appears on the surface to be inspected.

[0012] In the above configuration, the processor may also color the fluorescent region.

[0013] According to the above configuration, the operator can easily find the fluorescent region in the black and white image.

[0014] In the above configuration, when the ultraviolet cut filter is disposed on the optical axis, the camera and the ultraviolet cut filter may be spaced apart along the optical axis.

[0015] For example, when attaching an ultraviolet filter to a lens barrel, there is a risk that the optical axis of the camera may become unstable when the filter is attached. By making the camera and the ultraviolet cut filter non-contact, this fluctuation in the optical axis can be suppressed.

[0016] In the above configuration, the fluorescent penetrant flaw detection device may include a filter arm that can move the ultraviolet cut filter on and off the optical axis of the camera.

[0017] According to the above configuration, there is no need for an operator to replace the filter.

[0018] In the above configuration, the flaw detection lamp may have a higher irradiation intensity during the second image capture than during the first image capture.

[0019] According to the above configuration, the second image capture is performed under high-intensity illumination, so that the fluorescent region can be clearly viewed. [Effects of the Invention]

[0020] The fluorescent penetrant flaw detection device disclosed in this specification can suppress the occurrence of false colors in captured images and can identify the position of flaws on the surface to be inspected. [Brief explanation of the drawings]

[0021] [Figure 1] 1 is a perspective view illustrating the configuration of a fluorescent penetrant flaw detection device according to an embodiment of the present invention. [Figure 2] FIG. 2 is a perspective view illustrating the structure of the periphery of the camera. [Figure 3] FIG. 1 is a diagram illustrating an example of a hardware configuration of a computer device. [Figure 4] FIG. 2 is a diagram illustrating an example of functional blocks of a computer device. [Figure 5] FIG. 10 is a diagram illustrating a flaw detection flow. [Figure 6] FIG. 2 is a diagram illustrating an example of a first captured image. [Figure 7] FIG. 10 is a diagram illustrating an example of a second captured image. [Figure 8] FIG. 10 is a diagram illustrating an example of a composite image. DETAILED DESCRIPTION OF THE INVENTION

[0022] 1. Test subject 1 illustrates a fluorescent penetrant flaw detection device according to this embodiment, and also illustrates an object 50 to be inspected.

[0023] 1 and 2 show a Cartesian coordinate system. This Cartesian coordinate system is composed of a U-axis, an L-axis, and a W-axis. The U-axis is a vertical axis. The L-axis and the W-axis extend on a horizontal plane. The L-axis and the W-axis are perpendicular to each other. The L-axis is parallel to the 0° line of the base 22 of the robot arm 20.

[0024] The object 50 to be inspected is, for example, an aluminum casting. The object 50 to be inspected constitutes, for example, a part of a vehicle body. The object 50 to be inspected is a large cast product in which the frame and panels of the vehicle body are integrally molded.

[0025] Since the object 50 to be inspected is part of the vehicle body frame, the object 50 to be inspected is required to have a certain level of strength. For example, the object 50 to be inspected is provided with a plurality of ribs 52 and holes 54. Accordingly, the surface 51 to be inspected has an uneven shape. For example, when the object 50 to be inspected is assembled to a vehicle body, the surface 51 to be inspected becomes part of the side of the vehicle body. During flaw detection inspection, the object 50 to be inspected is laid down, with the surface 51 to be inspected facing upward.

[0026] During flaw detection testing, the bottom surface of test surface 51 serves as opposing surface 53 for camera 40. In other words, opposing surface 53 is disposed approximately perpendicular to the optical axis of camera 40. For example, opposing surface 53 intersects with the optical axis of the camera at an angle in the range of 80° to 100°.

[0027] Known fluorescent flaw detection methods include fluorescent magnetic particle testing and fluorescent penetrant testing. Fluorescent magnetic particle testing is not suitable for aluminum, which is a non-magnetic material. Therefore, fluorescent penetrant testing is applied to inspected aluminum castings.

[0028] In fluorescent penetrant testing, a penetrant is applied to the surface 51 to be inspected. A fluorescent substance is added to the penetrant. After the penetrant is applied, the penetrant is removed from the surface 51 to be inspected after a predetermined time. When ultraviolet light is further irradiated onto the surface 51 to be inspected, the penetrant that has soaked into scratches and cracks (i.e., that has not been removed) emits light.

[0029] 2. Fluorescent penetrant inspection equipment 1 and 2 illustrate a robot arm 20 and devices supported by the arm. The robot arm 20 supports a camera 40, a ring light 44, and an ultraviolet cut filter 48. The robot arm 20 is, for example, a four-axis articulated robot. The robot arm 20 includes a base 22, a first arm 24, a second arm 26, and an attachment bar 28.

[0030] The base 22 includes a lower base portion 22A and an upper base portion 22B. The upper base portion 22B rotates about an axis L1 relative to the lower base portion 22A. The axis L1 is parallel to the U-axis (vertical axis). For example, the upper base portion 22B rotates relative to the lower base portion 22A by a servo motor (not shown).

[0031] One end of a first arm 24 is connected to the upper base portion 22B. One end of a second arm 26 is connected to the other end of the first arm 24. Furthermore, an attachment bar 28 is connected to the other end of the second arm 26.

[0032] The first arm 24 rotates about a rotation axis L2 relative to the upper base portion 22B. The second arm 26 rotates about a rotation axis L3 relative to the first arm 24. Furthermore, the attachment bar 28 rotates about a rotation axis L4 relative to the second arm 26. The rotation axes L2, L3, and L4 each extend horizontally. Servo motors (not shown) are provided on the rotation axes L2, L3, and L4.

[0033] The attachment bar 28 extends on an extension line of the second arm 26. A camera 40, a ring light 44, and an ultraviolet cut filter 48 are attached to the attachment bar 28. That is, the robot arm 20 moves the camera 40 and the ultraviolet cut filter 48 relative to the object 50 to be inspected.

[0034] Camera 40 captures an image of a wide area of ​​surface 51 to be inspected (see FIG. 1). For example, camera 40 captures an image of the entire surface 51 to be inspected in one shot. For example, camera 40 is a camera with interchangeable lenses. A lens barrel 42 is attached to the lens mount of camera 40. The tip of lens barrel 42 supports an objective lens 46.

[0035] Furthermore, the camera 40 is a monochrome camera. For example, the camera 40 does not have a color filter placed on the image sensor. The image sensor outputs pixel values ​​ranging from 0 (black) to 255 (white) depending on the intensity of the received light. Because the camera 40 is a monochrome camera, the interpolation process that is performed when generating a digital color image is not performed. In this way, the fluorescent penetrant flaw detection device according to this embodiment uses monochrome images, which in principle do not produce false colors, instead of color digital images, which may produce false colors.

[0036] 2, a ring light 44 is attached to the attachment bar 28 so as to surround the objective lens 46. The ring light 44 is a ring-shaped ultraviolet flaw detection lamp. The ring light 44 irradiates the surface 51 to be inspected with ultraviolet light. The camera 40 and the ring light 44 are positioned so that the center of the ring light 44 passes through the optical axis of the camera 40. The relative positions of the camera 40 and the ring light 44 are fixed.

[0037] For example, the ring light 44 is a so-called UV ring light. The ring light 44 includes a plurality of ultraviolet LEDs 45 as a light source. The plurality of ultraviolet LEDs 45 are arranged in a circular ring shape. The optical axis of the ultraviolet LEDs 45 is parallel to the optical axis of the camera 40, for example.

[0038] For example, UV-A rays are emitted from the ultraviolet LED 45. The wavelength band of UV-A rays is 315 nm to 400 nm. In contrast, the lower limit of the wavelength of visible light is known to be 360 ​​nm. In other words, the light emitted from the ring light 44 contains components in the wavelength band of visible light.

[0039] The illumination intensity of the ring light 44 may be variable. For example, the illumination intensity of the ring light 44 varies between 0% and 100%. For example, the ring light 44 can be switched between two intensities: high intensity (for example, 100%) and low intensity (for example, 20%). For example, the intensity refers to luminous intensity [W / sr], which is a unit of radiation intensity. Alternatively, the irradiance [W / m 2 ] may also be used.

[0040] 1 and 2, an ultraviolet cut filter 48 is attached to the attachment bar 28. The ultraviolet cut filter 48 is attached to the attachment bar 28 via a filter arm 47.

[0041] For example, filter arm 47 extends linearly in parallel with the optical axis of camera 40. Filter arm 47 further includes a movable portion 47A and a fixed portion 47B. Filter arm 47 also includes a servo motor (not shown) that rotates movable portion 47A. Movable portion 47A rotates relative to fixed portion 47B around rotation axis L5.

[0042] An ultraviolet cut filter 48 is attached to the end of the movable part 47A. The filter arm 47 is capable of moving the ultraviolet cut filter 48 off the optical axis (FIG. 1) and on the optical axis (FIG. 2) of the camera 40. In the on-axis arrangement in FIG. 2, the main surface of the ultraviolet cut filter 48 and the optical axis of the camera 40 are perpendicular to each other.

[0043] Here, the ultraviolet cut filter 48 is spaced apart from the camera 40 and the lens barrel 42 along the optical axis. In other words, the lens barrel 42 and the ultraviolet cut filter 48 are not in contact with each other. For example, if the ultraviolet cut filter 48 were to be attached to the lens barrel 42 using a fitting mechanism such as a tab, there is a risk that the optical axis of the camera 40 would be shaken during the flaw detection flow described below. By making the ultraviolet cut filter 48 not in contact with the camera 40 and the lens barrel 42, it is possible to prevent the optical axis of the camera 40 from being shaken between the first and second imaging described below. Referring to FIG. 2, for example, the ultraviolet cut filter 48 is disposed between the objective lens 46 and the ring light 44.

[0044] The ultraviolet cut filter 48 blocks wavelengths of, for example, 400 nm or less. As described above, the wavelength band of UV-A rays is 315 nm to 400 nm. In other words, the ultraviolet cut filter 48 removes the visible light component of UV-A rays.

[0045] Referring to Fig. 1, a computer device 10 is connected to a robot arm 20, a camera 40, a ring light 44, and a filter arm 47. The computer device 10 controls these devices. The computer device 10 is also connected to a display unit 16 and an input device 17. Referring to Fig. 1, the display unit 16 is, for example, a display device. The input device 17 is, for example, a keyboard or a mouse.

[0046] Referring to FIG. 3, the computer device 10 includes a CPU 11, a RAM 12, a ROM 13, a storage 14, and an input / output controller 15.

[0047] The CPU 11 is a central processing unit, also called a processor. The RAM 12 is a volatile storage device that temporarily stores data during operation. The ROM 13 is a storage device from which data can be read. The storage 14 is a storage device from which data can be written and read. The storage 14 is configured, for example, by an HDD (Hard Disk Drive) or an SSD (Solid State Drive).

[0048] When the CPU 11 executes a program stored in the storage 14 or the ROM 13, the functional blocks illustrated in Fig. 4 are constructed in the computer device 10. The CPU 11 (processor) includes a camera control unit 18A, a robot control unit 18B, a lighting control unit 18C, and an image processing unit 18E. These functional blocks execute the flaw detection flow illustrated in Fig. 5. In other words, the CPU 11 processes the image captured by the camera 40 in accordance with the flaw detection flow.

[0049] 3. Flaw detection flow 1, the object 50 to be inspected is placed at a fixed position for flaw detection inspection. For example, a stage (not shown) is provided in front of the robot arm 20. The object 50 to be inspected is placed on this stage. For flaw detection inspection, the surface 51 to be inspected of the object 50 to be inspected faces upward.

[0050] Furthermore, a penetrant is applied to the surface 51 to be inspected by an operator or a painting robot. After the penetrant is applied, the penetrant is removed from the surface 51 to be inspected after a predetermined time. Once the penetrant is removed, a command to start the flaw detection inspection is sent from the input device 17 to the camera control unit 18A. This command triggers the start of the flaw detection flow shown in FIG. 5.

[0051] 1, 4, and 5, camera control unit 18A transmits to robot control unit 18B a movement command to capture an image of inspection surface 51. Robot control unit 18B controls robot arm 20 to move camera 40 to the imaging point (S10).

[0052] For example, the robot control unit 18B aligns the optical axis of the camera 40 with the geometric center of the surface 51 to be inspected. Furthermore, the robot control unit 18B makes the optical axis of the camera 40 perpendicular to the opposing surface 53 of the surface 51 to be inspected (see FIG. 1). The camera control unit 18A also separates the camera 40 from the surface 51 to be inspected. For example, the camera control unit 18A separates the camera 40 from the surface 51 to be inspected by, for example, a distance of 50 cm to 150 cm from the top end of the surface 51 to be inspected. Furthermore, the camera control unit 18A sets the magnification ratio so that the entire surface 51 to be inspected is within the field of view of the camera 40.

[0053] When the camera 40 is placed at the imaging point and the imaging setting is completed, the robot control unit 18B transmits a movement completion notification to the camera control unit 18A. Upon receiving the movement completion notification, the camera control unit 18A transmits a command to the camera 40 and the illumination control unit 18C to image the inspection surface 51.

[0054] The camera 40 captures an image of the inspection surface 51 at least twice. In capturing images multiple times, the relative position, relative angle, and imaging magnification of the camera 40 are fixed with respect to the inspection surface 51. For example, from step S12 to step S16, the relative position, relative angle, and imaging magnification of the camera 40 with respect to the inspection surface 51 are fixed.

[0055] The fluorescent penetrant flaw detection device performs imaging without using the ultraviolet cut filter 48 (first imaging) and imaging through the ultraviolet cut filter 48 (second imaging). For example, the imaging conditions for the first imaging and the second imaging are the same except for the presence or absence of the ultraviolet cut filter 48. Alternatively, in addition to the presence or absence of the ultraviolet cut filter 48, the illumination intensity of the ring light 44 may be changed between the first imaging and the second imaging.

[0056] For example, in the first image capture, the filter arm 47 moves the ultraviolet cut filter 48 off the optical axis of the camera 40. In this state, the camera 40 captures an image of the inspection surface 51 (S12). After the first image capture, the camera control unit 18A (see FIG. 4) sends a movement command to the robot control unit 18B. In response to the movement command, the robot control unit 18B rotates the filter arm 47. This positions the ultraviolet cut filter 48 on the optical axis of the camera 40 (S14).

[0057] The robot control unit 18B sends a movement completion command to the camera control unit 18A. In response to this, the camera control unit 18A sends an imaging command to the lighting control unit 18C and the camera 40. Then, the camera 40 performs a second imaging operation.

[0058] 6 shows an example of an image captured by the first imaging. In the first imaging, the inspection surface is illuminated by fluorescence from the flaw detection agent and the visible light component of the ring light 44. Therefore, the image of the inspection surface 51 appears relatively clear (compared to the second imaging).

[0059] As described above, the camera 40 is a monochrome camera. Therefore, the image obtained by the first capture (first captured image) is a grayscale image. In this case, it is difficult to distinguish between the fluorescence emitted by the flaw detection agent and the visible light component of the ring light 44.

[0060] In order to make it possible to distinguish between the fluorescence due to the flaw detection agent and the visible light component of the ring light 44, a second image is taken (S16). As illustrated in Fig. 7, in the second image, the fluorescent region 60 has a visible brightness. On the other hand, the region other than the fluorescent region 60 is a dark region that is difficult to see.

[0061] Here, in the second image capture, the illumination intensity of the ring light 44 may be higher than that in the first image capture. By illuminating the surface 51 to be inspected with high intensity, the fluorescent region 60 can be clearly imaged. Furthermore, in the second image capture, the visible light component of the ring light 44 is filtered out by the ultraviolet cut filter 48. Therefore, halation of the surface 51 to be inspected due to the visible light component is suppressed.

[0062] In this way, the first captured image is obtained by the first capturing. In the first captured image, a clear image of the surface 51 to be inspected is obtained, but it is difficult to distinguish between the visible light component of the ring light 44 and the fluorescence of the flaw detection agent. In addition, the second captured image is obtained by the second capturing. In the second captured image, the fluorescence of the flaw detection agent is visible, but it is unclear where on the surface 51 to be inspected it (fluorescence) appears.

[0063] As described above, throughout the first and second images, the relative position, relative angle, and imaging magnification of the camera 40 with respect to the inspection surface 51 are fixed. Therefore, by superimposing the first captured image and the second captured image, it is possible to identify the area of ​​the inspection surface 51 in which fluorescence is present.

[0064] 5, image processing unit 18E extracts fluorescent region 60 from the second captured image (S18). For example, image processing unit 18E extracts, as fluorescent region 60, a region of the second captured image that exceeds a predetermined brightness threshold.

[0065] Furthermore, image processing unit 18E superimposes fluorescent region 60 on the first captured image (S20). As described above, the relative position, relative angle, and imaging magnification of camera 40 with respect to inspection surface 51 are fixed throughout the first and second captured images. In other words, the origin position and coordinate scale of the local coordinate system in the first captured image are the same in the world coordinate system as the origin position and coordinate scale of the local coordinate system in the second captured image. Therefore, by plotting fluorescent region 60 on the coordinate plane of the first captured image, a composite image 65 such as that shown in FIG. 8 is generated.

[0066] Note that a coloring process for the fluorescent region 60 may be performed between step S18 and step S20. For example, image processing unit 18E imparts a predetermined fluorescent color to the fluorescent region 60. As a result, in the composite image 65, the fluorescent region 60 in the fluorescent color is superimposed on the grayscale first captured image. That is, the visibility of the fluorescent region 60 is improved.

[0067] The image processing unit 18E displays the composite image 65 on the display unit 16 (see FIG. 1) (S22). The composite image 65 allows the operator to visually identify the location of the flaw on the inspection surface 51. [Explanation of symbols]

[0068] 10 Computer device, 11 CPU (processor), 18A Camera control unit, 18B Robot control unit, 18C Lighting control unit, 18E Image processing unit, 20 Robot arm, 28 Attachment bar, 40 Camera, 42 Lens barrel, 44 Ring light, 45 UV LED, 46 Objective lens, 47 Filter arm, 48 UV cut filter, 50 Test object, 51 Test surface, 60 Fluorescent area, 65 Composite image.

Claims

1. a camera for capturing an image of the surface to be inspected; a flaw detection lamp that irradiates ultraviolet light onto the surface to be inspected; an ultraviolet cut filter that can be placed on the optical axis of the camera; a processor that processes the captured image of the camera; Equipped with the camera is a monochrome camera, the camera performs a first image capture without using the ultraviolet cut filter and a second image capture through the ultraviolet cut filter with the relative position, relative angle, and magnification fixed with respect to the inspection surface; The processor: extracting a region exceeding a predetermined brightness threshold as a fluorescent region from the image obtained by the second capture; superimposing the fluorescent region on the image obtained by the first imaging; Fluorescent penetrant inspection equipment.

2. The fluorescent penetrant flaw detection device according to claim 1, The processor processes the fluorescent region to have a color. Fluorescent penetrant inspection equipment.

3. The fluorescent penetrant flaw detection device according to claim 1, When the ultraviolet cut filter is disposed on the optical axis, the camera and the ultraviolet cut filter are spaced apart along the optical axis. Fluorescent penetrant inspection equipment.

4. The fluorescent penetrant flaw detection device according to claim 3, a filter arm that can move the ultraviolet cut filter on and off the optical axis of the camera; Fluorescent penetrant inspection equipment.

5. The fluorescent penetrant flaw detection device according to any one of claims 1 to 4, The flaw detection lamp has a higher irradiation intensity during the second imaging compared to the first imaging. Fluorescent penetrant inspection equipment.

Images