Pipe inspection system, pipe inspection method, and pipe inspection program

Abstract

To provide a system that calculates the outer diameter and thickness of a tubular object and determines whether the tubular object is a suitable product even if the acquired image is not an accurate circle. [Solution] An inspection system including an inspection device that generates an image by cutting out an area including the end face of the pipe from an image of the end face of the pipe captured by an imaging device, identifies inner circumference pixels indicating the inner circumference of the pipe and outer circumference pixels indicating the outer circumference of the pipe in the image, calculates the outer diameter of the pipe based on the outer circumference pixels, calculates the thickness of the pipe based on the outer circumference pixels and the inner circumference pixels, calculates the ratio of the pipe thickness to the outer diameter of the pipe, and determines whether the ratio is within a predetermined range.

Description

[Technical Field]

[0001] The present disclosure relates to a pipe inspection system, a pipe inspection method, and a pipe inspection program. [Background technology]

[0002] Conventionally, methods for calculating the outer diameter and thickness of a tubular object have been known. For example, Patent Document 1 describes a method for calculating the outer diameter and thickness of a tubular object by identifying the outer contour and inner contour of the tubular object from an image of the end face of the tubular object.

[0003] Another known method for estimating the thickness of an object is to estimate the thickness based on the number of pixels in an image of the object. For example, Patent Document 2 describes a method for estimating the thickness of a tire by measuring the number of pixels from the center of the tire toward the outer periphery in a cross-sectional image of the tire. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Patent No. 4816817 [Patent Document 2] Japanese Patent Application Publication No. 2023-180954 Summary of the Invention [Problem to be solved by the invention]

[0005] The method described in Patent Document 1 is based on the premise that the image is an accurate circle, and therefore has the problem that it cannot be applied when the image is unclear or when the target tubular object is not an accurate circle.

[0006] Furthermore, when the method described in Patent Document 2 is applied to a tubular object, there is a problem in that unless the image is an accurate circle, it is not possible to identify the pixel that is the center of the circle.

[0007] The object of the present disclosure is to provide a system that can calculate the outer diameter and pipe thickness of a tubular object and determine whether the tubular object is a suitable product even if the acquired image is not exactly circular. [Means for solving the problem]

[0008] (1) A pipe inspection system according to an embodiment of the present disclosure includes: 1. A pipe inspection system comprising: the pipe inspection system includes an imaging device and an inspection device; the imaging device images the end surface of the pipe; The inspection device includes: a region including the end face of the pipe is cut out from the image captured by the imaging device to generate a cut-out image; In the cutout image, an internal region that is the inside of the pipe, an external region that is the outside of the pipe, and an end surface region that is the end surface of the pipe are identified; Identifying inner periphery pixels in the end face region that indicate an inner periphery of the pipe and outer periphery pixels in the end face region that indicate an outer periphery of the pipe; Calculating a first length and a second length based on the inner perimeter pixels and the outer perimeter pixels; calculating a ratio of the second length to the first length; It is determined whether the ratio is within a predetermined range.

[0009] (2) A pipe inspection system according to one embodiment of the present disclosure is the pipe inspection system according to (1), the first length is an outer diameter of the tube; the second length is the thickness of the pipe; The inspection device includes: Calculating the length of a line segment connecting two points on the perimeter pixels of the pipe based on the perimeter pixels; Calculating a pipe thickness of the pipe based on the outer peripheral pixels and the inner peripheral pixels; A ratio of the pipe thickness to the first length is calculated.

[0010] (3) A pipe inspection system according to one embodiment of the present disclosure is the pipe inspection system according to (1) or (2), The inspection device identifies, as the inner periphery pixels, pixels in the end face region that have a side that contacts the inner region.

[0011] (4) A pipe inspection system according to an embodiment of the present disclosure is the pipe inspection system according to any one of (1) to (3), The inspection device identifies, among the pixels in the end face region, pixels having a side that contacts the internal region and pixels having only a vertex that contacts the internal region as the internal perimeter pixels.

[0012] (5) A pipe inspection system according to an embodiment of the present disclosure is the pipe inspection system according to any one of (1) to (4), The inspection device includes: Refer to pixels in a predetermined range centered on the inner periphery pixel, It is determined whether the inner perimeter pixels are pixels having a side that contacts the internal region or pixels having only a vertex that contacts the internal region.

[0013] (6) A pipe inspection system according to an embodiment of the present disclosure is the pipe inspection system according to any one of (1) to (5), The inspection device includes: Calculating vectors from the inner periphery pixel to two inner periphery pixels adjacent to the inner periphery pixel; Calculate the absolute value of the dot product of the two calculated vectors, Statistical processing is performed on the calculated absolute values ​​to identify whether the inner perimeter pixels are pixels having sides in contact with the internal region or pixels having only vertices in contact with the internal region.

[0014] (7) A pipe inspection system according to an embodiment of the present disclosure is the pipe inspection system according to any one of (1) to (6), The inspection device identifies, as the peripheral pixels, pixels in the end face region that have a side that contacts the external region.

[0015] (8) A pipe inspection system according to an embodiment of the present disclosure is the pipe inspection system according to any one of (1) to (7), The inspection device identifies, among the pixels in the end face region, pixels having a side that contacts the external region and pixels having only a vertex that contacts the external region as the peripheral pixels.

[0016] (9) A pipe inspection system according to one embodiment of the present disclosure is the pipe inspection system according to any one of (1) to (8), The inspection device includes: Refer to pixels in a predetermined range centered on the peripheral pixel, It is determined whether the perimeter pixels are pixels having sides that contact the external region or pixels having only vertices that contact the external region.

[0017] (10) A pipe inspection system according to one embodiment of the present disclosure is the pipe inspection system according to any one of (1) to (9), The inspection device includes: Calculating vectors from the peripheral pixel to two peripheral pixels adjacent to the peripheral pixel; Calculate the absolute value of the dot product of the two calculated vectors, A statistical process is performed on the calculated absolute values ​​to identify whether the peripheral pixels are pixels having a side in contact with the external region or pixels having only a vertex in contact with the external region.

[0018] (11) A pipe inspection system according to one embodiment of the present disclosure is the pipe inspection system according to (5) or (9), The inspection device includes: If noise is present in the extracted image, the range of pixels to be referenced is widened.

[0019] (12) A pipe inspection system according to an embodiment of the present disclosure is the pipe inspection system according to any one of (2) to (11), The inspection device includes: One arbitrarily selected outer pixel is set as the outer edge point of interest, Calculating the distance between the target perimeter point and each of the perimeter pixels other than the target perimeter point; Identifying the farthest perimeter point, which is the perimeter pixel that is the longest distance from the target perimeter point; calculating a distance between the target perimeter point and the farthest perimeter point as a line segment length of the target perimeter point; storing the target perimeter point and the line segment length of the target perimeter point in association with each other; Calculating the line segment lengths of the target perimeter points for at least half of the perimeter pixels; The first length is calculated based on a statistical value of the line segment lengths of the target perimeter points.

[0020] (13) A pipe inspection system according to an embodiment of the present disclosure is the pipe inspection system according to any one of (2) to (12), The inspection device includes: One arbitrarily selected inner pixel is set as the inner point of interest. Calculating the distance between the inner perimeter point of interest and each of all outer perimeter pixels; Identifying a nearest perimeter point, which is a perimeter pixel that is the shortest distance from the inner perimeter point of interest; Calculating the distance between the inner peripheral point of interest and the nearest outer peripheral point as the pipe thickness of the inner peripheral point of interest; storing the target inner circumferential point and the pipe thickness at the target inner circumferential point in association with each other; Calculating the pipe thickness of the inner peripheral point of interest for at least half of the inner peripheral pixels; The pipe thickness of the pipe is calculated based on the statistical values ​​of the pipe thickness at the target inner circumferential point.

[0021] (14) A pipe inspection system according to an embodiment of the present disclosure is the pipe inspection system according to any one of (1) to (13), If the tube is circular, the first length is an outer diameter of the tube; the second length is the thickness of the pipe; The inspection device includes: One arbitrarily selected outer pixel is set as the outer edge point of interest, Calculating the distance between the target perimeter point and each of the perimeter pixels other than the target perimeter point; Identifying the farthest perimeter point, which is the perimeter pixel that is the longest distance from the target perimeter point; calculating a distance between the target outer periphery point and the farthest outer periphery point as an outer diameter of the target outer periphery point; storing the target outer periphery point and the outer diameter of the target outer periphery point in association with each other; Calculating the outer diameter of the target outer periphery point for at least half of the outer periphery pixels; Identifying the outer diameter of the pipe based on statistics of the outer diameter of the outer periphery point of interest; Calculating the ratio of the wall thickness of the pipe to the outer diameter of the pipe; It is determined whether the ratio is within a predetermined range.

[0022] (15) A pipe inspection system according to an embodiment of the present disclosure is the pipe inspection system according to any one of (1) to (13), If the tube is rectangular, the inspection device One arbitrarily selected outer pixel is set as the outer edge point of interest, Calculating the distance between the target perimeter point and each of the perimeter pixels other than the target perimeter point; Identifying the farthest perimeter point, which is the perimeter pixel that is the longest distance from the target perimeter point; calculating a distance between the target perimeter point and the farthest perimeter point as a line segment length of the target perimeter point; storing the target perimeter point and the line segment length of the target perimeter point in association with each other; Calculating the line segment lengths of the target perimeter points for at least half of the perimeter pixels; Identifying the longest line segment length of the target outer periphery point as the diagonal length of the pipe; Calculating the ratio of the wall thickness of the tube to the diagonal length of the tube; It is determined whether the ratio is within a predetermined range.

[0023] (16) A pipe inspection system according to an embodiment of the present disclosure is the pipe inspection system according to any one of (1) to (13), If the tube is triangular, the inspection device One arbitrarily selected outer pixel is set as the outer edge point of interest, Calculating the distance between the target perimeter point and each of the perimeter pixels other than the target perimeter point; Identifying the farthest perimeter point, which is the perimeter pixel that is the longest distance from the target perimeter point; calculating a distance between the target perimeter point and the farthest perimeter point as a line segment length of the target perimeter point; storing the target perimeter point and the line segment length of the target perimeter point in association with each other; Calculating the line segment lengths of the target perimeter points for at least half of the perimeter pixels; Identifying the longest line segment length of the target outer periphery point as the side length of the pipe; Calculating the ratio of the pipe thickness to the side length of the pipe; It is determined whether the ratio is within a predetermined range.

[0024] (17) A pipe inspection system according to one embodiment of the present disclosure is the pipe inspection system according to (13), If the tube is triangular or rectangular, the inspection device: When calculating the pipe thickness of the pipe, it is determined whether the target inner peripheral point is a corner pixel indicating a corner of the triangle or the quadrangle; If the target inner periphery point is not a corner pixel, calculate the pipe thickness for at least half of the inner periphery pixels; Identifying the inner perimeter point of interest, the nearest outer perimeter point, and pixels on a line connecting the inner perimeter point of interest and the nearest outer perimeter point as line pixels; If the target inner perimeter point is a corner pixel, a remaining pixel indicating a pixel that does not belong to the line pixel is identified; It is determined whether the degree of match between the remaining pixels and pre-stored corner area information is within a predetermined range.

[0025] (18) A method for inspecting a pipe according to one embodiment of the present disclosure, comprising: An imaging device photographs the end surface of the pipe; A step of generating a cut-out image by cutting out an area including the end face of the pipe from the image captured by the imaging device; identifying an internal region that is the inside of the pipe, an external region that is the outside of the pipe, and an end surface region that is the end surface of the pipe in the extracted image; identifying inner perimeter pixels in the end face region that represent an inner perimeter of the tube and outer perimeter pixels in the end face region that represent an outer perimeter of the tube; calculating an outer diameter of the tube based on the perimeter pixels; calculating a thickness of the pipe based on the outer periphery pixels and the inner periphery pixels; calculating a ratio of the pipe wall thickness to the outer diameter of the pipe; and determining whether the ratio is within a predetermined range.

[0026] (19) A pipe inspection program according to an embodiment of the present disclosure, comprising: On the computer, taking an image of the end surface of the pipe; extracting an area including the end face of the pipe from the captured image to generate an extracted image; Identifying an internal region that is an interior of the pipe, an external region that is an exterior of the pipe, and an end surface region that is an end surface of the pipe in the cut-out image; identifying inner perimeter pixels in the end face region that represent an inner perimeter of the tube and outer perimeter pixels in the end face region that represent an outer perimeter of the tube; calculating an outer diameter of the tube based on the perimeter pixels; Calculating a pipe thickness of the pipe based on the outer peripheral pixels and the inner peripheral pixels; Calculating a ratio of the pipe thickness to the outer diameter of the pipe; and determining whether the ratio is within a predetermined range. [Effects of the Invention]

[0027] According to one embodiment of the present disclosure, even if the acquired image is not exactly circular, the outer diameter and thickness of the tubular object can be calculated and it can be determined whether the tubular object is a suitable product. [Brief explanation of the drawings]

[0028] [Figure 1] 1 is a schematic diagram illustrating a pipe inspection system according to an embodiment of the present disclosure. [Figure 2] FIG. 1 is a block diagram illustrating a configuration of an imaging device according to an embodiment of the present disclosure. [Figure 3] 1 is a block diagram showing a configuration of an inspection device according to an embodiment of the present disclosure. [Figure 4] 10 is a flowchart illustrating an example of the operation of a pipe inspection system according to an embodiment of the present disclosure. [Figure 5A] FIG. 1 is an image diagram showing processing details of a pipe inspection system according to an embodiment of the present disclosure. [Figure 5B] FIG. 1 is an image diagram showing processing details of a pipe inspection system according to an embodiment of the present disclosure. [Figure 6A] FIG. 1 is an image diagram showing processing details of a pipe inspection system according to an embodiment of the present disclosure. [Figure 6B] FIG. 1 is an image diagram showing processing details of a pipe inspection system according to an embodiment of the present disclosure. [Figure 6C] FIG. 1 is an image diagram showing processing details of a pipe inspection system according to an embodiment of the present disclosure. [Figure 7] 10 is a flowchart illustrating an example of the operation of a pipe inspection system according to an embodiment of the present disclosure. [Figure 8] FIG. 1 is an image diagram showing processing details of a pipe inspection system according to an embodiment of the present disclosure. [Figure 9] 10 is a flowchart illustrating an example of the operation of a pipe inspection system according to an embodiment of the present disclosure. [Figure 10]10 is a flowchart illustrating an example of the operation of a pipe inspection system according to an embodiment of the present disclosure. [Figure 11A] FIG. 1 is an image diagram showing processing details of a pipe inspection system according to an embodiment of the present disclosure. [Figure 11B] FIG. 1 is an image diagram showing processing details of a pipe inspection system according to an embodiment of the present disclosure. [Figure 12] 10 is a flowchart illustrating an example of the operation of a pipe inspection system according to an embodiment of the present disclosure. [Figure 13] 10 is a flowchart illustrating an example of the operation of a pipe inspection system according to an embodiment of the present disclosure. [Figure 14A] FIG. 1 is an image diagram showing processing details of a pipe inspection system according to an embodiment of the present disclosure. [Figure 14B] FIG. 1 is an image diagram showing processing details of a pipe inspection system according to an embodiment of the present disclosure. DETAILED DESCRIPTION OF THE INVENTION

[0029] Hereinafter, an embodiment of the present disclosure will be described with reference to the drawings. In each drawing, parts having the same configuration or function are denoted by the same reference numerals. In the description of this embodiment, duplicated descriptions of the same parts may be omitted or simplified as appropriate.

[0030] FIG. 1 is a diagram schematically illustrating the configuration of an inspection system 1 according to the present disclosure. The inspection system 1 includes an imaging device 10 and an inspection device 20. The inspection system 1 determines whether an object to be inspected is within a predetermined range of conforming products and whether the object to be inspected is mixed with other objects of a different type. The following description will be given using an example in which the object to be inspected is a pipe 30 and the pipe 30 is flowing down a production line.

[0031] 2 is a block diagram showing the configuration of the imaging device 10. The imaging device 10 captures an image of the end face of a pipe 30 flowing down a production line. The imaging device 10 includes an imaging unit 11, a storage unit 12, and a communication unit 13.

[0032] The photographing unit 11 photographs the end face of the tube 30 as it flows down the production line. The photographing unit 11 is not particularly limited as long as it can photograph the end face of the tube 30 with high accuracy, and may be, for example, a camera.

[0033] The storage unit 12 includes one or more semiconductor memories, one or more magnetic memories, one or more optical memories, or a combination of at least two of these. The semiconductor memory is, for example, a random access memory (RAM) or a read-only memory (ROM). The RAM is, for example, a static RAM (SRAM) or a dynamic RAM (DRAM). The ROM is, for example, an electrically erasable programmable read-only memory (EEPROM). The storage unit 12 functions as, for example, a main storage device, an auxiliary storage device, or a cache memory. The storage unit 12 stores images captured by the image capture unit 11. The storage unit 12 may also store programs and data used in the operation of the image capture device 10.

[0034] The communication unit 13 includes one or more communication interfaces. The communication interfaces may be either wired or wireless communication interfaces. The communication interfaces may be, for example, interfaces compatible with mobile communication standards such as Bluetooth (registered trademark), LTE (Long Term Evolution), 4G (4th Generation), or 5G (5th Generation), or a LAN interface. The communication unit 13 may communicate with the inspection device 20 directly or via a network.

[0035] 3 is a block diagram showing the configuration of the inspection device 20. The inspection device 20 calculates the outer diameter and pipe thickness of the pipe 30 based on an image of the end face of the pipe 30 captured by the imaging device 10, and determines whether the pipe 30 is within a predetermined range of conforming products and whether a different type of pipe 30 has been mixed in. The inspection device 20 may be a general-purpose computer such as a workstation or a personal computer, or may be a dedicated computer configured to function as the inspection device 20. The inspection device 20 includes a control unit 21, a memory unit 22, a communication unit 23, an input unit 24, and an output unit 25.

[0036] The control unit 21 includes one or more processors, one or more dedicated circuits, or a combination thereof. The processor is a general-purpose processor such as a CPU (Central Processing Unit) or a GPU (Graphics Processing Unit), or a dedicated processor specialized for a specific process. The dedicated circuit is, for example, an FPGA (Field-Programmable Gate Array) or an ASIC (Application Specific Integrated Circuit). The control unit 21 executes various processes related to the operation of the inspection device 20 and controls each part of the inspection device 20. The control unit 21 may realize the control function by executing a program stored in the memory unit 22. The control unit 21 may control the imaging device 10 via the communication unit 23.

[0037] The memory unit 22 includes one or more semiconductor memories, one or more magnetic memories, one or more optical memories, or a combination of at least two of these. The semiconductor memory is, for example, a random access memory (RAM) or a read-only memory (ROM). The RAM is, for example, a static RAM (SRAM) or a dynamic RAM (DRAM). The ROM is, for example, an electrically erasable programmable read-only memory (EEPROM). The memory unit 22 functions as, for example, a main memory device, an auxiliary memory device, or a cache memory. The memory unit 22 may store a compatibility range, which is the range of compatible products for the pipe 30. The memory unit 22 may also store programs and data used in the operation of the inspection device 20.

[0038] When AI (Artificial Intelligence) is used in the inspection process, the storage unit 22 may store a machine learning model for the process using AI.

[0039] The communication unit 23 includes one or more communication interfaces. The communication interfaces may be either wired or wireless communication interfaces. The communication interfaces may be, for example, interfaces compatible with mobile communication standards such as Bluetooth (registered trademark), LTE (Long Term Evolution), 4G (4th Generation), or 5G (5th Generation), or a LAN interface. The communication unit 23 may communicate with the image capture device 10 directly or via a network.

[0040] The input unit 24 includes one or more input interfaces. The input interface may be, for example, a physical key, a capacitance key, a pointing device, a touch screen integrated with a display, or a microphone that accepts voice input. The input interface may further include a camera that captures captured images or image codes, or an IC card reader. The input unit 24 accepts operations to input information used in the operation of the inspection device 20 and sends the input information to the control unit 21.

[0041] The output unit 25 includes one or more output interfaces. The output interface is, for example, an external or built-in display that outputs information as an image or video, a speaker that outputs information as sound, or an interface for connecting to an external output device. The display is, for example, an LCD (Liquid Crystal Display) or an organic EL (Electro Luminescence) display. The output unit 25 outputs information obtained by the operation of the inspection device 20.

[0042] The operation of the inspection device 20 is realized by a processor included in the control unit 21 executing a program. The program may be recorded on a computer-readable recording medium. The computer-readable recording medium is, for example, a magnetic recording device, an optical disc, a magneto-optical recording medium, or a semiconductor memory. The program is distributed in a state recorded on a portable recording medium such as a DVD (Digital Versatile Disc) or a CD (Compact Disc)-ROM on which the program is recorded. Some or all of the operation of the inspection device 20 may be executed by a dedicated circuit included in the control unit 21.

[0043] Next, a method for inspecting an object to be inspected by the inspection system 1 will be described. In the inspection system 1 according to an embodiment of the present disclosure, the shape of the object to be inspected is identified based on the ratio of two reference lengths, and whether the ratio is within a predetermined range (the conformance range) is determined to determine whether the object is conforming. For example, if the object to be inspected is a pipe 30, if the distance between the image capture device 10 and the pipe 30 is always constant, the size of the pipe 30 captured in the captured image can be matched to the actual size of the pipe 30, and the absolute values ​​of the outer diameter and pipe thickness of the pipe 30 can be measured based on the captured image. However, in practice, it is difficult to always maintain a constant distance between the image capture device 10 and the pipe 30, and an error of at least several millimeters may occur. Therefore, it is difficult to measure the absolute values ​​of the outer diameter and pipe thickness of the pipe 30. In the inspection system 1 according to an embodiment of the present disclosure, if the object to be inspected is a pipe 30, the ratio of the pipe thickness to the outer diameter of the pipe is calculated, and therefore the ratio can be accurately calculated even if the distance between the image capture device 10 and the pipe 30 is not constant. By using this method, even if it is difficult to maintain a constant distance between the imaging device and the object to be inspected, it is possible to accurately determine whether the product is suitable or not.

[0044] The following description explains how the inspection system 1 according to one embodiment of the present disclosure determines whether the ratio of the pipe thickness to the outer diameter of the pipe is within a conforming range, thereby determining whether the pipe 30 being inspected is a conforming product and whether a different type of pipe 30 has been mixed in.

[0045] (Setting the compatibility range) The control unit 21 of the inspection device 20 receives information about the compatibility range via the input unit 24. The control unit 21 stores the received information about the compatibility range in the storage unit 22. The control unit 21 may store the information about the compatibility range in the storage unit 22 in association with the type of pipe.

[0046] (Compliance assessment) Next, a method for calculating the ratio of the pipe thickness to the pipe outer diameter in the inspection system 1 according to an embodiment of the present disclosure, and a method for determining whether the pipe being inspected is a conforming product, will be described with reference to Figures 4 to 9. The following describes a determination method particularly for circular pipes. Figure 4 is a flowchart showing the procedure by which the control unit 21 of the inspection device 20 determines conformance of the pipe 30.

[0047] The imaging device 10 captures an image of the end face of the tube 30, which is the object of inspection and is flowing through the production line (step S101). The image captured by the imaging device 10 may be stored in the memory unit 12 and then transmitted to the inspection device 20 via the communication unit 13, or may be transmitted directly to the inspection device 20 without going through the memory unit 12.

[0048] When the control unit 21 of the inspection device 20 acquires the end face image of the pipe 30 photographed by the photographing device 10, it generates a cropped image by cropping out an area including the end face of the pipe 30 (step S102). The control unit 21 of the inspection device 20 may generate the cropped image using, for example, AI.

[0049] The control unit 21 of the inspection device 20 identifies an internal region that is the inside of the pipe 30, an external region that is the outside of the pipe 30, and an end face region that is the end face of the pipe 30 from the cut-out image generated in step S102 (step S103). AI may also be used to identify each region.

[0050] The control unit 21 of the inspection device 20 identifies inner peripheral pixels that indicate the inner peripheral portion of the pipe and outer peripheral pixels that indicate the outer peripheral portion of the pipe in the end face region of the image data whose respective regions have been identified in step S103 (step S104). The inner peripheral portion of the pipe refers to the portion of the end face region that is in contact with the internal region, and the outer peripheral portion of the pipe refers to the portion of the end face region that is in contact with the external region.

[0051] The control unit 21 of the inspection device 20 identifies, among the pixels present in the end face region, pixels having a side that contacts the internal region as inner peripheral pixels. Similarly, the control unit 21 of the inspection device 20 identifies, among the pixels present in the end face region, pixels having a side that contacts the external region as outer peripheral pixels. Furthermore, the control unit 21 of the inspection device 20 identifies, among the pixels present in the end face region, pixels that do not correspond to either inner peripheral pixels or outer peripheral pixels as end face pixels. The control unit 21 of the inspection device 20 may store the identified inner peripheral pixels, outer peripheral pixels, and end face pixels in the memory unit 22.

[0052] 5A is an image diagram of an end face region image in which inner peripheral pixels, outer peripheral pixels, and end face pixels have been identified by the control unit 21 of the inspection device 20. In FIG. 5A, light hatching with diagonal lines slanting upward to the right indicates inner peripheral pixels, light hatching with diagonal lines slanting upward to the left indicates outer peripheral pixels, dark hatching with diagonal lines slanting upward to the right indicates end face pixels, and white indicates an internal region or an external region.

[0053] The control unit 21 of the inspection device 20 may specify, among the pixels present in the end face region, not only pixels having sides that contact the internal region but also pixels having only a vertex that contacts the internal region as inner peripheral pixels. A pixel having only a vertex that contacts the internal region refers to a pixel that does not have a side that contacts the internal region but has a vertex that contacts the internal region. Similarly, the control unit 21 of the inspection device 20 may specify, among the pixels present in the end face region, not only pixels having sides that contact the external region but also pixels having only a vertex that contacts the external region as outer peripheral pixels. A pixel having only a vertex that contacts the external region refers to a pixel that does not have a side that contacts the external region but has a vertex that contacts the external region.

[0054] Figure 5B is an illustration of the case where pixels that only have vertices that contact the internal region in Figure 5A are additionally identified as inner periphery pixels. Dark hatching with diagonal lines slanting upward to the left indicates pixels that only have vertices that contact the internal region and are additionally identified as inner periphery pixels. Other pixels are similar to Figure 5A.

[0055] If an error occurs during image processing or AI processing such as steps S102 and S103, or if noise or the like gets into the image, distortion or the like may occur in the shape of the edge region image. If such distortion occurs, it may be impossible to accurately identify inner or outer peripheral pixels simply by determining whether they are in contact with the inner or outer region.

[0056] In such a case, the control unit 21 of the inspection device 20 may perform pattern matching by referring to pixels in a predetermined range centered on one of the identified inner perimeter pixels. If the relationship between the inner perimeter pixel and the adjacent inner perimeter pixel is a straight line or a curve, the control unit 21 of the inspection device 20 may determine that the inner perimeter pixel is a pixel that has a side that contacts the internal region. If the relationship between the inner perimeter pixel and the adjacent pixel is a corner, the control unit 21 of the inspection device 20 may determine that the inner perimeter pixel is a pixel that has only a vertex that contacts the internal region.

[0057] Similarly, the control unit 21 of the inspection device 20 may perform pattern matching by referring to pixels in a predetermined range centered on one of the identified perimeter pixels. If the relationship between the perimeter pixel and the adjacent perimeter pixel is a straight line or a curve, the control unit 21 of the inspection device 20 may determine that the perimeter pixel is a pixel that has a side that contacts the external region. If the relationship between the perimeter pixel and the adjacent pixel is a corner, the control unit 21 of the inspection device 20 may determine that the perimeter pixel is a pixel that has only a vertex that contacts the external region.

[0058] Note that a linear pixel relationship refers to a state in which adjacent pixels are lined up in a straight line in either the vertical, horizontal, or diagonal direction. A curved pixel relationship refers to a state in which one adjacent pixel is adjacent in the vertical or horizontal direction, and the other adjacent pixel is adjacent in a diagonal direction without contacting the other adjacent pixel. A corner pixel relationship refers to a state in which one adjacent pixel is adjacent in the vertical direction, and the other adjacent pixel is adjacent in the horizontal direction.

[0059] 6A to 6C are diagrams illustrating the relationship between any inner or outer peripheral pixel and its adjacent pixels through pattern matching. In FIGS. 6A to 6C, the pixel in question is indicated by light hatching, and the adjacent pixel is indicated by dark hatching. When the pixel in question and the adjacent pixel are arranged as shown in FIG. 6A, the control unit 21 of the inspection device 20 determines that the pixel relationship is a corner. When the pixel in question and the adjacent pixel are arranged as shown in FIG. 6B, the control unit 21 of the inspection device 20 determines that the pixel relationship is a curve. When the pixel in question and the adjacent pixel are arranged as shown in FIG. 6C, the control unit 21 of the inspection device 20 determines that the pixel relationship is a straight line.

[0060] If noise is present in the cropped image generated in step S102 or if the inner periphery of the cropped image is not smooth, the control unit 21 of the inspection device 20 may widen the range of pixels to be referenced. For example, although a 3×3 range of pixels is referenced in FIGS. 6A to 6C, the range of pixels to be referenced may be widened to a 5×5 range, a 7×7 range, or the like.

[0061] Alternatively, the control unit 21 of the inspection device 20 may calculate a vector from one of the identified inner periphery pixels to two inner periphery pixels adjacent to the identified inner periphery pixel, and then calculate the absolute value of the dot product of the two calculated vectors. If the inner periphery pixels are arranged in a straight line, the absolute value of the calculated dot product will be close to 1. If the identified inner periphery pixel is near a corner, i.e., if the angle between the adjacent pixels is close to 90°, the absolute value of the calculated dot product will be close to 0. The control unit 21 of the inspection device 20 performs statistical processing on a combination of multiple vectors to determine whether the relationship between the inner periphery pixel and the adjacent pixels is a corner. If the relationship between the inner periphery pixel and the adjacent pixels is a corner, the control unit 21 of the inspection device 20 may determine that the inner periphery pixel is a pixel that has only vertices that contact the internal region.

[0062] The control unit 21 of the inspection device 20 may use a similar method to determine whether the relationship between one of the identified peripheral pixels and an adjacent pixel is a corner. If the relationship between the peripheral pixel and the adjacent pixel is a corner, the control unit 21 of the inspection device 20 may determine that the peripheral pixel is a pixel that has only a vertex that contacts the external region.

[0063] The control unit 21 of the inspection device 20 calculates the outer diameter of the pipe 30 based on the outer periphery pixels identified in step S104 (step S105).

[0064] The processing of step S105 will be described with reference to Fig. 7. Fig. 7 is a flowchart showing the specific procedure of step S105. Fig. 8 is an image diagram in which the pixels of the cut-out image generated in step S102 shown in Fig. 4 by the control unit 21 of the inspection device 20 are visualized. In describing the flowchart shown in Fig. 7, Fig. 8 will be referred to as appropriate.

[0065] The control unit 21 of the inspection device 20 sets one arbitrarily selected perimeter pixel as a target perimeter point (step S201).

[0066] The control unit 21 of the inspection device 20 measures the distance between the target perimeter point set in step S201 and each of the perimeter pixels other than the target perimeter point. After measuring the distances to all perimeter pixels, the control unit 21 of the inspection device 20 identifies the farthest perimeter point, which is the perimeter pixel that is the longest distance from the target perimeter point (step S202). For example, in Figure 8, if B6 is set as the target perimeter point, the farthest perimeter point is N10.

[0067] The control unit 21 of the inspection device 20 calculates the distance between the target outer periphery point set in step S201 and the farthest outer periphery point identified in step S202 as the outer diameter of the target outer periphery point (step S203). The control unit 21 of the inspection device 20 stores the calculated outer diameter of the target outer periphery point in the memory unit 22.

[0068] The control unit 21 of the inspection device 20 refers to the memory unit 22 and checks whether the outer diameters of the target outer periphery points have been calculated for a predetermined number of target outer periphery points (step S204). The predetermined number may be at least half the number of outer periphery pixels. If the outer diameters of the target outer periphery points have not been calculated for the predetermined number of target outer periphery points (step S204: No), the control unit 21 of the inspection device 20 performs steps S201 to S203 again.

[0069] If the outer diameters of the predetermined number of target outer periphery points have been calculated (step S204: Yes), the control unit 21 of the inspection device 20 calculates the outer diameter of the pipe 30 based on the statistical value of the outer diameters of the predetermined number of target outer periphery points calculated in step S203 (step S205). The statistical value may be, for example, the average value of the calculated outer diameters of the target outer periphery points. The statistical value may be, for example, the average value of the calculated outer diameters of the target outer periphery points excluding 20% ​​of the top and bottom ends, or may be the median of the calculated outer diameters of the target outer periphery points.

[0070] After calculating the outer diameter of the pipe 30 in step S205, the control unit 21 of the inspection device 20 calculates the pipe thickness of the pipe 30 based on the inner peripheral pixels and outer peripheral pixels identified in step S104 (step S106).

[0071] The process of step S106 will be described with reference to Fig. 9. Fig. 9 is a flowchart showing the specific procedure of step S106.

[0072] The control unit 21 of the inspection device 20 sets one arbitrarily selected inner periphery pixel as a target inner periphery point (step S301).

[0073] The control unit 21 of the inspection device 20 measures the distance between the inner perimeter point of interest set in step S301 and each of the outer perimeter pixels. After measuring the distances to all the outer perimeter pixels, the control unit 21 of the inspection device 20 identifies the nearest outer perimeter point, which is the outer perimeter pixel that is the shortest distance from the inner perimeter point of interest (step S302). For example, in Figure 8, if E8 is set as the inner perimeter point of interest, the nearest outer perimeter point is B8.

[0074] The control unit 21 of the inspection device 20 calculates the distance between the inner peripheral point of interest set in step S301 and the nearest outer peripheral point identified in step S302 as the pipe thickness of the inner peripheral point of interest (step S303). The control unit 21 of the inspection device 20 stores the calculated pipe thickness of the inner peripheral point of interest in the memory unit 22.

[0075] The control unit 21 of the inspection device 20 refers to the memory unit 22 and checks whether the pipe thicknesses of the target inner peripheral points have been calculated for a predetermined number of target inner peripheral points (step S304). The predetermined number may be at least half of the total number of inner peripheral pixels. More preferably, it is all of the inner peripheral pixels. If the pipe thicknesses of the target inner peripheral points have not been calculated for the predetermined number of target inner peripheral points (step S304: No), the control unit 21 of the inspection device 20 performs steps S301 to S303 again.

[0076] If the pipe thicknesses of the predetermined number of target inner periphery points have been calculated (step S304: Yes), the control unit 21 of the inspection device 20 calculates the pipe thickness of the pipe 30 based on the statistical values ​​of the pipe thicknesses of the predetermined number of target inner periphery points calculated in step S303 (step S305). The statistical value may be, for example, the average value of the calculated pipe thicknesses of the target inner periphery points. The statistical value may be, for example, the average value of the calculated pipe thicknesses of the target inner periphery points excluding 20% ​​of the top and bottom ends, or the median value of the calculated pipe thicknesses of the target inner periphery points. In this way, by performing statistical processing on at least half of the inner periphery pixels, the pipe thickness of the pipe 30 can be calculated more accurately even if the image captured by the imaging device 10 is distorted, such as when the pipe 30 is captured from a slightly oblique angle. Furthermore, by performing statistical processing on all inner periphery pixels, the pipe thickness of the pipe 30 can be calculated with even greater accuracy.

[0077] In step S301, the control unit 21 of the inspection device 20 may set one arbitrarily selected outer periphery pixel as the outer periphery point of interest instead of the inner periphery point of interest. In this case, in step S302, the distance between the outer periphery point of interest and each of the inner periphery pixels is measured. After measuring the distances to all the inner periphery pixels, the control unit 21 of the inspection device 20 identifies the nearest inner periphery point, which is the inner periphery pixel that is the shortest distance from the outer periphery point of interest. In step S303, the control unit 21 of the inspection device 20 calculates the distance between the outer periphery point of interest and the nearest inner periphery point as the pipe thickness of the outer periphery point of interest. After calculating the pipe thickness of the inner periphery point of interest for a predetermined number of outer periphery points of interest, the control unit 21 of the inspection device 20 calculates the pipe thickness of the pipe 30 based on the statistics of the pipe thicknesses of the predetermined number of outer periphery points of interest.

[0078] After calculating the outer diameter of the pipe 30 in step S205 and the thickness of the pipe 30 in step S305, the control unit 21 of the inspection device 20 calculates the ratio of the pipe thickness to the outer diameter of the pipe 30 (step S107).

[0079] The control unit 21 of the inspection device 20 refers to the storage unit 22 and determines whether the ratio of the pipe thickness to the outer diameter of the pipe 30 calculated in step S107 is within a preset range for conforming products (step S108).

[0080] (Conformity assessment when the pipe is triangular or rectangular) Next, a method for determining whether a pipe to be inspected conforms to the standard when the pipe is triangular or rectangular will be described with reference to Figures 10 to 14B. Figure 10 is a flowchart showing the procedure by which the control unit 21 of the inspection device 20 determines conformance of a triangular or rectangular pipe 30. When the pipe is triangular, conformance is determined based on the ratio of the pipe thickness to the pipe side length. When the pipe is rectangular, conformance is determined based on the ratio of the pipe thickness to the pipe diagonal length.

[0081] The imaging device 10 captures an image of the end face of the tube 30, which is the object of inspection and is flowing through the production line (step S401). The image captured by the imaging device 10 may be stored in the memory unit 12 and then transmitted to the inspection device 20 via the communication unit 13, or may be transmitted directly to the inspection device 20 without going through the memory unit 12.

[0082] When the control unit 21 of the inspection device 20 acquires the image of the end face of the pipe 30 photographed by the photographing device 10, it generates a cut-out image by cutting out an area including the end face of the pipe 30 (step S402). The control unit 21 of the inspection device 20 may generate the cut-out image using, for example, AI.

[0083] The control unit 21 of the inspection device 20 identifies an internal region that is the inside of the pipe 30, an external region that is the outside of the pipe 30, and an end face region that is the end face of the pipe 30 from the cut-out image generated in step S402 (step S403). AI may also be used to identify each region.

[0084] The control unit 21 of the inspection device 20 identifies inner peripheral pixels indicating the inner peripheral portion of the pipe and outer peripheral pixels indicating the outer peripheral portion of the pipe in the end face regions of the image data whose respective regions have been identified in step S103 (step S404).

[0085] The control unit 21 of the inspection device 20 identifies, among the pixels present in the end face region, pixels having a side that contacts the internal region as inner peripheral pixels. Similarly, the control unit 21 of the inspection device 20 identifies, among the pixels present in the end face region, pixels having a side that contacts the external region as outer peripheral pixels. Furthermore, the control unit 21 of the inspection device 20 identifies, among the pixels present in the end face region, pixels that do not correspond to either inner peripheral pixels or outer peripheral pixels as end face pixels. The control unit 21 of the inspection device 20 may store the identified inner peripheral pixels, outer peripheral pixels, and end face pixels in the memory unit 22.

[0086] Fig. 11A is an image of an end face region image for a rectangular pipe, in which the inner peripheral pixels, outer peripheral pixels, and end face pixels have been identified by the control unit 21 of the inspection device 20. Fig. 11B is an image of an end face region image for a triangular pipe, in which the inner peripheral pixels, outer peripheral pixels, and end face pixels have been identified by the control unit 21 of the inspection device 20. In Figs. 11A and 11B, light hatching with diagonal lines slanting upward to the right indicates inner peripheral pixels, hatching with diagonal lines slanting upward to the left indicates outer peripheral pixels, dark hatching with diagonal lines slanting upward to the right indicates end face pixels, and white indicates an internal region or an external region.

[0087] The positional relationship between two adjacent inner peripheral pixels may be determined by pattern matching, calculating the absolute value of the dot product of vectors, or the like, in the same manner as in step S104.

[0088] The control unit 21 of the inspection device 20 determines the diagonal length or the side length of the pipe 30 based on the peripheral pixels determined in step S404 (step S405).

[0089] The process of step S405 will be described with reference to Fig. 12. Fig. 12 is a flowchart showing the specific procedure of step S405.

[0090] The control unit 21 of the inspection device 20 sets one arbitrarily selected perimeter pixel as a target perimeter point (step S501).

[0091] The control unit 21 of the inspection device 20 measures the distance between the target perimeter point set in step S501 and each of the perimeter pixels other than the target perimeter point. After measuring the distances to all perimeter pixels, the control unit 21 of the inspection device 20 identifies the farthest perimeter point, which is the perimeter pixel that is the longest distance from the target perimeter point (step S502). For example, in FIG. 11A, if B2 is set as the target perimeter point, the farthest perimeter point is M13. Also, in FIG. 11B, if C9 is set as the target perimeter point, the farthest perimeter point is N2.

[0092] The control unit 21 of the inspection device 20 calculates the distance between the target perimeter point set in step S501 and the farthest perimeter point identified in step S502 as the line segment length of the target perimeter point (step S503). The control unit 21 of the inspection device 20 stores the calculated line segment length of the target perimeter point in the memory unit 22.

[0093] The control unit 21 of the inspection device 20 refers to the memory unit 22 and checks whether the line segment lengths of the target perimeter points have been calculated for a predetermined number of target perimeter points (step S504). The predetermined number may be at least half the number of perimeter pixels. If the line segment lengths of the target perimeter points have not been calculated for the predetermined number of target perimeter points (step S504: No), the control unit 21 of the inspection device 20 performs steps S501 to S503 again.

[0094] If the line segment lengths of the predetermined number of target outer periphery points have been calculated (step S504: Yes), the control unit 21 of the inspection device 20 identifies the longest line segment length of the predetermined number of target outer periphery points calculated in step S503 as the diagonal length or side length of the pipe 30 (step S505).

[0095] When the diagonal length or side length of the pipe 30 is identified in step S505, the control unit 21 of the inspection device 20 calculates the pipe thickness of the pipe 30 based on the inner peripheral pixels and outer peripheral pixels identified in step S404 (step S406).

[0096] The process of step S406 will be described with reference to Fig. 13. Fig. 13 is a flowchart showing the specific procedure of step S406.

[0097] The control unit 21 of the inspection device 20 sets one arbitrarily selected inner periphery pixel as a target inner periphery point (step S601).

[0098] The control unit 21 of the inspection device 20 determines whether the target inner periphery point set in step S601 is a corner pixel (step S602). A corner pixel is a pixel that includes a part that corresponds to a corner of the inner or outer periphery of the pipe 30. If the target inner periphery point is a corner pixel (step S602: Yes), processing for the target inner periphery point is interrupted, and a different inner periphery pixel is set as the target inner periphery point.

[0099] If the inner perimeter point of interest is not a corner pixel (step S602: No), the control unit 21 of the inspection device 20 measures the distance between the inner perimeter point of interest set in step S601 and each of the outer perimeter pixels. After measuring the distances to all the outer perimeter pixels, the control unit 21 of the inspection device 20 identifies the nearest outer perimeter point, which is the outer perimeter pixel that is the shortest distance from the inner perimeter point of interest (step S603). For example, in FIG. 11A, if D6 is set as the inner perimeter point of interest, the nearest outer perimeter point is B6. Also, in FIG. 11B, if H9 is set as the inner perimeter point of interest, the nearest outer perimeter point is G11.

[0100] The control unit 21 of the inspection device 20 calculates the distance between the inner peripheral point of interest set in step S601 and the nearest outer peripheral point identified in step S603 as the pipe thickness of the inner peripheral point of interest (step S604). The control unit 21 of the inspection device 20 stores the calculated pipe thickness of the inner peripheral point of interest in the memory unit 22.

[0101] The control unit 21 of the inspection device 20 refers to the memory unit 22 and checks whether the pipe thicknesses of the target inner periphery points have been calculated for a predetermined number of target inner periphery points (step S605). The predetermined number may be at least half of the total number of inner periphery pixels excluding the corner pixels. More preferably, it is all of the inner periphery pixels. If the pipe thicknesses of the target inner periphery points have not been calculated for the predetermined number of target inner periphery points (step S605: No), the control unit 21 of the inspection device 20 performs steps S601 to S604 again.

[0102] If the pipe thicknesses of the predetermined number of target inner peripheral points have been calculated (step S605: Yes), the control unit 21 of the inspection device 20 calculates the pipe thickness of the pipe 30 based on the statistical values ​​of the pipe thicknesses of the predetermined number of target inner peripheral points calculated in step S604 (step S606). The statistical value may be, for example, the average value of the calculated pipe thicknesses of the target inner peripheral points. The statistical value may be, for example, the average value of the calculated pipe thicknesses of the target inner peripheral points excluding 20% ​​of the top and bottom ends, or the median value of the calculated pipe thicknesses of the target inner peripheral points. In this way, by performing statistical processing on at least half of the inner peripheral pixels, the pipe thickness of the pipe 30 can be calculated more accurately even if the image captured by the imaging device 10 is distorted, such as when the pipe 30 is captured from a slightly oblique angle. Furthermore, by performing statistical processing on all inner peripheral pixels, the pipe thickness of the pipe 30 can be calculated with even greater accuracy.

[0103] The control unit 21 of the inspection device 20 identifies one inner periphery pixel that is not a corner pixel, the nearest outer periphery point corresponding to the inner periphery pixel, and pixels that exist on the line connecting the inner periphery pixel and the nearest outer periphery point as on-line pixels (step S607). The control unit 21 of the inspection device 20 repeatedly executes step S607 for all inner periphery pixels except for the corner pixels until on-line pixels are identified.

[0104] Once line pixels have been identified for all inner peripheral pixels excluding corner pixels, the control unit 21 of the inspection device 20 identifies the inner peripheral pixels, outer peripheral pixels, and end surface pixels that were not identified as line pixels as remaining pixels (step S608). Figure 14A is an image of an end surface region image in which line pixels and remaining pixels have been identified for a rectangular pipe. Figure 14B is an image of an end surface region image in which line pixels and remaining pixels have been identified for a triangular pipe. In Figures 14A and 14B, light hatching indicates line pixels, and dark hatching indicates remaining pixels.

[0105] However, if an arbitrarily selected outer pixel is set as the outer periphery point of interest instead of the inner periphery point of interest and the thickness of the triangular or rectangular pipe 30 is calculated based on the thickness of the outer periphery point of interest, the distance from the outer pixel corresponding to the remaining pixel to the nearest inner periphery point will be included as noise data. Therefore, when calculating the thickness of the triangular or rectangular pipe 30, it is necessary to calculate it based on the thickness of the inner periphery point of interest.

[0106] After calculating the outer diameter of the pipe 30 in step S505, calculating the thickness of the pipe 30 in step S606, and identifying the remaining pixels of the pipe 30 in step S608, the control unit 21 of the inspection device 20 calculates the ratio of the pipe thickness to the outer diameter of the pipe 30 (step S407).

[0107] The control unit 21 of the inspection device 20 refers to the storage unit 22 and determines whether the ratio of the pipe thickness to the outer diameter of the pipe 30 calculated in step S407 is within a preset range for conforming products (step S408).

[0108] The control unit 21 of the inspection device 20 determines whether there is a chip in the corner region of the pipe 30 (step S409). The corner region refers to a region of a predetermined range that includes the corner of the pipe 30. In one embodiment of the present disclosure, the corner region may match the remaining pixels. The control unit 21 of the inspection device 20 refers to the memory unit 22 and determines whether the degree of match between the remaining pixels identified in step S608 and the pre-stored corner region information is within a predetermined range for conforming products. If the degree of match between the remaining pixels and the corner region information is within the conforming product range, the control unit 21 of the inspection device 20 determines that there is no chip in the corner region of the pipe 30. If the degree of match between the remaining pixels and the corner region information is outside the conforming product range, the control unit 21 of the inspection device 20 determines that there is a chip in the corner region of the pipe 30.

[0109] The present disclosure is not limited to the above-described embodiments. For example, multiple blocks shown in the block diagrams may be integrated, or a single block may be divided. Instead of executing multiple steps shown in the flowcharts in chronological order as described, steps may be executed in parallel or in a different order depending on the processing capabilities of the device executing each step, or as needed. Other modifications are possible within the scope of the present disclosure.

[0110] In the present disclosure, the ratio of the pipe thickness to the outer diameter of the pipe is calculated for a circular shape, the ratio of the pipe thickness to the side length of the pipe is calculated for a triangular shape, and the ratio of the pipe thickness to the diagonal length of the pipe is calculated for a rectangular shape. However, the calculated ratios are not limited to these. Arbitrary first and second lengths may be calculated based on the inner and outer perimeter pixels, and the ratio of the second length to the first length may be calculated. The first length may be, for example, the length of a line segment connecting two points on the outer perimeter pixels of the pipe. Furthermore, the second length may be, for example, the thickness of the pipe.

[0111] Although the present disclosure has been described using examples of circular, triangular, and rectangular pipes, the shape of the pipe is not limited to these. The present disclosure can also be applied to pipes having polygonal shapes with more sides than a triangle, such as pentagons and hexagons. [Explanation of symbols]

[0112] 1. Inspection system 10 Imaging equipment 11 Filming Department 12 Storage section 13 Communications Department 20 Inspection equipment 21 Control section 22 Memory section 23 Communications Department 24 Input section 25 Output section 30 tubes

Claims

1. 1. A pipe inspection system comprising: the pipe inspection system includes an imaging device and an inspection device; the imaging device images the end surface of the pipe; The inspection device includes: a region including the end face of the pipe is cut out from the image captured by the imaging device to generate a cut-out image; In the cutout image, an internal region that is the inside of the pipe, an external region that is the outside of the pipe, and an end surface region that is the end surface of the pipe are identified; Identifying inner periphery pixels in the end face region that indicate an inner periphery of the pipe and outer periphery pixels in the end face region that indicate an outer periphery of the pipe; calculating a first length and a second length based on the inner periphery pixels and the outer periphery pixels; calculating a ratio of the second length to the first length; determining whether the ratio is within a predetermined range; Pipe inspection system.

2. 10. The pipe inspection system of claim 1, the first length is the length of a line segment connecting two points on the perimeter pixels of the tube; the second length is a wall thickness of the tube; The inspection device includes: Calculating the first length based on the perimeter pixels; Calculating a pipe thickness of the pipe based on the outer peripheral pixels and the inner peripheral pixels; Calculating a ratio of the pipe thickness to the first length. Pipe inspection system.

3. 3. The pipe inspection system of claim 2, the inspection device identifies, among the pixels in the end face region, pixels having a side that contacts the internal region as the internal perimeter pixels; Pipe inspection system.

4. 3. The pipe inspection system of claim 2, the inspection device identifies, among the pixels in the end face region, pixels having a side contacting the internal region and pixels having only a vertex contacting the internal region as the internal perimeter pixels; Pipe inspection system.

5. 3. The pipe inspection system of claim 2, The inspection device includes: Refer to pixels in a predetermined range centered on the inner periphery pixel, Identifying whether the inner perimeter pixels are pixels having an edge that contacts the internal region or pixels having only a vertex that contacts the internal region; Pipe inspection system.

6. 3. The pipe inspection system of claim 2, The inspection device includes: Calculating a vector from the inner periphery pixel to two inner periphery pixels adjacent to the inner periphery pixel; Calculate the absolute value of the dot product of the two calculated vectors, performing statistical processing on the calculated absolute values ​​to identify whether the inner periphery pixels are pixels having a side in contact with the internal region or pixels having only a vertex in contact with the internal region; Pipe inspection system.

7. 3. The pipe inspection system of claim 2, the inspection device identifies, as the peripheral pixels, pixels in the end face region that have a side that contacts the external region; Pipe inspection system.

8. 3. The pipe inspection system of claim 2, the inspection device identifies, among the pixels in the end face region, pixels having a side contacting the external region and pixels having only a vertex contacting the external region as the peripheral pixels; Pipe inspection system.

9. 3. The pipe inspection system of claim 2, The inspection device includes: Refer to pixels in a predetermined range centered on the peripheral pixel, Identifying whether the perimeter pixels are pixels having sides that contact the external region or pixels having only vertices that contact the external region; Pipe inspection system.

10. 3. The pipe inspection system of claim 2, The inspection device includes: Calculating a vector from the peripheral pixel to two peripheral pixels adjacent to the peripheral pixel; Calculate the absolute value of the dot product of the two calculated vectors, performing statistical processing on the calculated absolute values ​​to identify whether the peripheral pixels are pixels having sides in contact with the external region or pixels having only vertices in contact with the external region; Pipe inspection system.

11. 10. A pipe inspection system according to claim 5 or 9, The inspection device includes: If noise is present in the extracted image, the range of pixels to be referenced is widened. Pipe inspection system.

12. 3. The pipe inspection system of claim 2, The inspection device includes: One arbitrarily selected perimeter pixel is set as the target perimeter point, Calculating the distance between the target perimeter point and each of the perimeter pixels other than the target perimeter point; Identifying the farthest perimeter point, which is the perimeter pixel that is the longest distance from the target perimeter point; calculating a distance between the target perimeter point and the farthest perimeter point as a line segment length of the target perimeter point; storing the target perimeter point and the line segment length of the target perimeter point in association with each other; Calculating the line segment lengths of the target perimeter points for at least half of the perimeter pixels; calculating the first length based on a statistical value of the line segment length of the target perimeter point; Pipe inspection system.

13. 3. The pipe inspection system of claim 2, The inspection device includes: One arbitrarily selected inner pixel is set as the inner point of interest, Calculating the distance between the inner perimeter point of interest and each of all outer perimeter pixels; Identifying a nearest perimeter point, which is a perimeter pixel that is the shortest distance from the inner perimeter point of interest; Calculating the distance between the inner peripheral point of interest and the nearest outer peripheral point as the pipe thickness of the inner peripheral point of interest; storing the target inner circumferential point and the pipe thickness at the target inner circumferential point in association with each other; Calculating the pipe thickness of the inner peripheral point of interest for at least half of the inner peripheral pixels; calculating a pipe thickness of the pipe based on a statistical value of the pipe thickness at the target inner circumferential point; Pipe inspection system.

14. 10. The pipe inspection system of claim 1, If the tube is circular, the first length is an outer diameter of the tube; the second length is a wall thickness of the tube; The inspection device includes: One arbitrarily selected perimeter pixel is set as the target perimeter point, Calculating the distance between the target perimeter point and each of the perimeter pixels other than the target perimeter point; Identifying the farthest perimeter point, which is the perimeter pixel that is the longest distance from the target perimeter point; calculating a distance between the target outer periphery point and the farthest outer periphery point as an outer diameter of the target outer periphery point; storing the target outer periphery point and the outer diameter of the target outer periphery point in association with each other; Calculating the outer diameter of the target outer periphery point for at least half of the outer periphery pixels; Identifying the outer diameter of the pipe based on statistics of the outer diameter of the outer periphery point of interest; Calculating the ratio of the wall thickness of the pipe to the outer diameter of the pipe; determining whether the ratio is within a predetermined range; Pipe inspection system.

15. 10. The pipe inspection system of claim 1, If the tube is rectangular, the first length is a diagonal length of the tube; the second length is a wall thickness of the tube; The inspection device includes: One arbitrarily selected perimeter pixel is set as the target perimeter point, Calculating the distance between the target perimeter point and each of the perimeter pixels other than the target perimeter point; Identifying the farthest perimeter point, which is the perimeter pixel that is the longest distance from the target perimeter point; calculating a distance between the target perimeter point and the farthest perimeter point as a line segment length of the target perimeter point; storing the target perimeter point and the line segment length of the target perimeter point in association with each other; Calculating the line segment lengths of the target perimeter points for at least half of the perimeter pixels; Identifying the longest line segment length of the target outer periphery point as the diagonal length of the pipe; Calculating the ratio of the wall thickness of the tube to the diagonal length of the tube; determining whether the ratio is within a predetermined range; Pipe inspection system.

16. 10. The pipe inspection system of claim 1, If the tube is triangular, the first length is a side length of the tube; the second length is a wall thickness of the tube; The inspection device includes: One arbitrarily selected perimeter pixel is set as the target perimeter point, Calculating the distance between the target perimeter point and each of the perimeter pixels other than the target perimeter point; Identifying the farthest perimeter point, which is the perimeter pixel that is the longest distance from the target perimeter point; calculating a distance between the target perimeter point and the farthest perimeter point as a line segment length of the target perimeter point; storing the target perimeter point and the line segment length of the target perimeter point in association with each other; Calculating the line segment lengths of the target perimeter points for at least half of the perimeter pixels; Identifying the longest line segment length of the target outer periphery point as the side length of the pipe; Calculating the ratio of the pipe thickness to the side length of the pipe; determining whether the ratio is within a predetermined range; Pipe inspection system.

17. 14. The tube inspection system of claim 13, If the tube is triangular or rectangular, the inspection device: When calculating the pipe thickness of the pipe, it is determined whether the target inner peripheral point is a corner pixel indicating a corner of the triangle or the quadrangle; If the target inner perimeter point is not a corner pixel, calculate the pipe thickness of the target inner perimeter point for at least half of the inner perimeter pixels; Identifying the inner perimeter point of interest, the nearest outer perimeter point, and pixels on a line connecting the inner perimeter point of interest and the nearest outer perimeter point as line pixels; If the target inner perimeter point is a corner pixel, a remaining pixel indicating a pixel that does not belong to the line pixel is identified; determining whether the degree of match between the remaining pixels and pre-stored corner region information is within a predetermined range; Pipe inspection system.

18. 1. A method for inspecting a pipe, comprising: An imaging device photographs the end surface of the pipe; A step of generating a cut-out image by cutting out an area including the end face of the pipe from the image captured by the imaging device; identifying an internal region that is the inside of the pipe, an external region that is the outside of the pipe, and an end surface region that is the end surface of the pipe in the extracted image; identifying inner perimeter pixels in the end face region that represent an inner perimeter of the tube and outer perimeter pixels in the end face region that represent an outer perimeter of the tube; calculating a first length and a second length based on the inner periphery pixels and the outer periphery pixels; calculating a ratio of the second length to the first length; and determining whether the ratio is within a predetermined range. Pipe inspection methods.

19. 1. A pipe inspection program comprising: On the computer, taking an image of the end surface of the pipe; extracting an area including the end face of the pipe from the captured image to generate an extracted image; Identifying an internal region that is an interior of the pipe, an external region that is an exterior of the pipe, and an end surface region that is an end surface of the pipe in the cut-out image; identifying inner perimeter pixels in the end face region that represent an inner perimeter of the tube and outer perimeter pixels in the end face region that represent an outer perimeter of the tube; calculating a first length and a second length based on the inner periphery pixels and the outer periphery pixels; calculating a ratio of the second length to the first length; determining whether the ratio is within a predetermined range; Pipe inspection program.

Images