Measurement system and method, and parts remanufacturing system and method

Abstract

To provide a measurement system capable of measuring a shape of a component by determining the necessity of removal of a film even if the surface of the component is covered with the film.SOLUTION: Disclosed is a measurement system for measuring a shape of a component whose surface is at least partially covered with a film, which includes: an imaging part which images the component to acquire the image data; an image analysis part for determining a state of the film to diagnose whether or not removal of the film is necessary and determine a measurement region based on the image data; and a measurement part for measuring a shape of the component in the measurement region. The measurement region includes at least a part of a region from which the film is removed when it is diagnosed that removal of the film is necessary.SELECTED DRAWING: Figure 3

Description

[Technical Field]

[0001] The present disclosure relates to metrology systems and methods, and component remanufacturing systems and methods. [Background technology]

[0002] With the recent rise in environmental awareness, there has been an increasing demand for recycling. One type of recycling is a business in which manufacturers collect products that have malfunctioned or deteriorated over time, and recycle and reuse the parts contained in the collected products. To recycle parts, it is necessary to identify the damaged areas that caused the malfunction or the areas damaged over time, and repair the damaged areas according to the extent of the damage.

[0003] Although it is not related to identifying and repairing damaged areas in parts included in a product, there is, for example, Patent Document 1, a technology for automatically inspecting corroded areas on the floor of a tank. The abstract and Figure 3 of Patent Document 1 state that "the main body 5 is provided with a measuring unit 10 that measures the distance to the floor 4, and the distance between the measuring unit 10 and the corroded area C and the floor 4 around it is measured."

[0004] Furthermore, paragraph 0043 and Figure 6 of Patent Document 1 state that "The distance between the corroded area C and the floor surface 4 around it and the measuring unit 10 is automatically measured as shown in Figure 3 (step S7). The depth of each corroded area C can be calculated as the difference in distance from the surrounding area, and the position of each corroded area C relative to the position of the main body 5 can be understood as the position of the measuring unit 10 when the corroded area C is detected as being deeper than the surrounding area."

[0005] Furthermore, although it is not related to identifying and repairing damaged areas in parts included in a product, Patent Document 2, for example, describes a technology for thermal spray repair of concave damaged areas that have occurred in the oven wall of a coke oven coking chamber. Paragraph 0100 of Patent Document 2 states, "Using a laser distance meter, for example, a laser profile meter 16, provided in a thermal spray repair device 10, the concave damaged area 4 in the oven wall 3 of the coking chamber 2 to be repaired is measured. Then, the controller 18 of the thermal spray repair device 10 processes the measurement values ​​of the laser profile meter 1 to obtain a three-dimensional coordinate diagram using contour lines as shown in Figure 7. Then, from this three-dimensional coordinate diagram, the damage depth H and the depth HM of the deepest part (for example, HM = 40 mm) of each coordinate position of the damaged area 4 are read." [Prior art documents] [Patent documents]

[0006] [Patent Document 1] Japanese Patent Application Publication No. 2019-200171 [Patent Document 2] Japanese Patent Application Laid-Open No. 2013-64083 Summary of the Invention [Problem to be solved by the invention]

[0007] However, in parts collected by manufacturers, the damaged area of ​​the part may be covered by corrosion such as rust or a coating such as a paint film and may not be exposed. Therefore, even if the techniques described in Patent Documents 1 and 2 are applied in such cases, it is not possible to calculate the extent of the damage, such as the depth of the damage. In other words, when the damaged area is covered by a coating, it is not possible to measure the shape of the part itself, which poses a problem in that it is not possible to identify the damaged area and the extent of the damage.

[0008] The present disclosure has been made in consideration of the above-mentioned problems, and aims to provide a measurement system and method, as well as a parts remanufacturing system and method, that can determine whether or not a coating needs to be removed and measure the shape of a part, even if the surface of the part is covered with a coating. [Means for solving the problem]

[0009] In order to solve the above problems, the measurement system of the present disclosure is, for example, a measurement system that measures the shape of a component at least part of whose surface is covered with a coating, and includes an imaging unit that images the component to obtain image data, an image analysis unit that determines the state of the coating based on the image data, diagnoses whether or not removal of the coating is necessary, and determines a measurement area, and a measurement unit that measures the shape of the component in the measurement area, and if it is determined that removal of the coating is necessary, the measurement area includes at least a portion of the area from which the coating has been removed. The image analysis unit estimates the degree of change in the shape of the part inside the coating from the image data as the degree of damage using a prediction model previously constructed based on the correlation between coating data and shape change of the part inside the coating, and diagnoses areas where the degree of damage is estimated to be large as requiring removal of the coating, and the measurement area includes areas where the coating has been removed from the areas where the degree of damage is estimated to be large. .

[0010] Furthermore, the part remanufacturing system of the present disclosure includes, for example, the measurement system, a repair diagnosis unit that determines whether the part can be repaired and the repair method based on the measurement results of the part shape by the measurement system and pre-recorded design data of the part, and a repair unit that repairs the part using the repair method when the repair diagnosis unit determines that the part can be repaired.

[0011] In order to solve the above problems, the measurement method disclosed herein is, for example, a measurement method for measuring the shape of a component having at least a portion of its surface covered with a coating, and includes an imaging step of imaging the component to obtain image data, an image analysis step of determining the state of the coating based on the image data, diagnosing whether or not removal of the coating is necessary, and determining a measurement area, a coating removal step of removing the coating diagnosed as requiring removal, and a measurement step of measuring the shape of the component in the measurement area after removal of the coating, and when it is diagnosed that removal of the coating is necessary, the measurement area includes at least a portion of the area from which the coating has been removed. The image analysis step uses a prediction model previously constructed based on the correlation between coating data and changes in the shape of the part inside the coating to estimate the degree of change in the shape of the part inside the coating from the image data as the degree of damage, and diagnoses areas where the degree of damage is estimated to be large as areas where the coating is required to be removed, and the measurement area includes areas where the coating is removed from the areas where the degree of damage is estimated to be large. .

[0012] Furthermore, the component remanufacturing method of the present disclosure includes, for example, a repair diagnosis process that, after the measurement method, determines whether the component can be repaired and how to repair it based on the measurement results of the component shape and pre-recorded design data for the component, and a repair process that repairs the component using the repair method if it is determined that the component can be repaired. [Effects of the Invention]

[0013] According to the present disclosure, it is possible to provide a measurement system and method, as well as a parts remanufacturing system and method, that can determine whether or not a coating needs to be removed and measure the shape of a part even if the surface of the part is covered with a coating. [Brief explanation of the drawings]

[0014] [Figure 1] FIG. 1 is a diagram showing the entire process of remanufacturing a part. [Figure 2] FIG. 1 is a flowchart of the entire process of remanufacturing parts. [Figure 3] FIG. 1 is a block diagram of a parts remanufacturing system. [Figure 4] 1 is a diagram showing an example of a coating according to the first embodiment. FIG. [Figure 5] FIG. 10 is a diagram showing an example of use of the component remanufacturing system according to the second embodiment. [Figure 6] 10A to 10C are diagrams illustrating an imaging process according to the third embodiment. [Figure 7] 10A to 10C are diagrams illustrating an imaging process according to the fourth embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0015] Hereinafter, embodiments of the present disclosure will be described with reference to the drawings. Note that the embodiments of the present disclosure are not limited to the embodiments described below, and various modifications are possible within the scope of the technical concept. Furthermore, corresponding parts in each drawing used to describe the embodiments described below are designated by the same reference numerals, and duplicated explanations will be omitted. (Embodiment 1) FIG. 1 illustrates the entire process of parts recycling. Specifically, FIG. 1 shows an example of the entire process from collecting malfunctioning or deteriorated products to reusing the parts contained in the collected products. The process shown in FIG. 1 includes a collection process 10 in which products are collected from users; a disassembly and cleaning process 20 in which the collected products are disassembled, the parts removed, and cleaned; a part measurement process 30 in which the shape and surface condition of the parts are measured; a repair feasibility assessment process 40 in which, based on the measurement results, the repair feasibility of the parts is assessed and, if repairable, a repair method is determined; a part repair process 50 in which repairs are performed based on the assessed repair method; a part inspection process 60 in which the parts are inspected to determine whether they have been repaired as expected; an assembly process 70 in which products are assembled using parts that pass inspection; a product inspection process 80 in which the assembled products are inspected; and a shipping process 90 in which products that pass product inspection are shipped. The process shown in FIG. 1 is merely an example and is not limiting. Of these processes, the present disclosure is primarily concerned with the component measurement process 30, the repair feasibility assessment process 40, and the component repair process 50.

[0016] Products for which recycled parts can be used include automobiles, trains, aircraft, ships, industrial equipment such as motors, power infrastructure such as turbines, and construction machinery such as bulldozers. These products are made up of many large parts, and repairing these parts each time they deteriorate often requires less total energy than replacing them with new parts. Furthermore, recycling resources reduces the environmental impact. As the transition to a recycling-oriented society becomes more widespread in the future, it will be important to improve the work efficiency of the parts recycling process and to recycle more parts in a shorter period of time.

[0017] Since the main material of the large parts is metal, this embodiment will be described assuming that metal parts are recycled. However, this disclosure is not limited to metal and can be applied to parts made of various materials, such as ceramic.

[0018] One of the criteria for determining whether a part can be repaired is the degree to which it has deviated from its design value. Therefore, in the part remanufacturing process disclosed herein, a part measurement process 30 is performed for a repairability assessment process 40 that assesses whether the part can be repaired and determines the repair method. In this disclosure, unless otherwise specified, a change in the shape of a part from its design value is referred to as damage, and the degree of change in shape is defined as the degree of damage. Furthermore, the area where damage is estimated to have occurred is referred to as the damaged area.

[0019] Many metal parts are painted, have a coating formed, or have corrosion such as rust covering their surfaces. In this disclosure, when a part is covered with a coating, a change in the shape of the part inside the coating (change from the design value) is called damage, the degree of change in the shape of the part inside the coating is defined as the degree of damage, and the area where damage is estimated to have occurred is called the damaged area.

[0020] Figure 2 is a flowchart of all the processes for part remanufacturing. Figure 3 is a block diagram of a part remanufacturing system. The part measurement process 30 shown in Figure 2 is a measurement method for measuring the shape of a part at least partially covered with a coating, and includes an imaging process 31, an image analysis process 32, a coating removal preparation process 33, a coating removal process 34, and a measurement process 35. Note that the coating removal preparation process 33 and the coating removal process 34 may be combined into one process. The part repair process 50 also includes a cleaning necessity diagnosis process 51, a cleaning process 52, a repair method diagnosis process 53, and a repair process 54.

[0021] 3 includes an imaging unit 2, an image analysis unit 3, a surface processing unit 4, a measurement unit 5, a repair diagnosis unit 6, and a repair unit 7. These are connected to a network 100 via interfaces I / F (2a to 7a). The measurement system according to this embodiment is a system that includes the imaging unit 2, the image analysis unit 3, the surface processing unit 4, and the measurement unit 5 of the part recycling system 1.

[0022] Each block (2-7) has a common interface I / F (2a-7a) for connecting to a network, a processor (CPU) (2b-7b) for executing programs in each block, memory (2c-7c) for storing programs executed by the processor, and storage units (2d-7d) for storing programs executed by the processor and data. These functions are typically implemented using a group of multiple servers and computer terminals, but they can also be implemented using a single server, or some or all of the functions can be implemented on the cloud (a cloud environment physically consisting of one or more servers), and various physical configurations are possible. The present disclosure is not limited to the physical configurations that implement these functions. Furthermore, some of the servers and computer terminals can be implemented using smartphones, tablets, personal computers (PCs), laptops, etc.

[0023] The memories (2c to 7c) include ROM (Read Only Memory), which is a non-volatile storage element, and RAM (Random Access Memory), which is a volatile storage element. The ROM stores unchanging programs. The RAM temporarily reads and stores programs and data corresponding to an application from the storage device when the application is started, etc.

[0024] The storage units (2d-7d) are large-capacity, nonvolatile storage devices such as hard disks, flash memory such as solid-state drives (SSDs), and optical storage devices. The storage units store data and programs required for each block to execute a program. The data and programs required for each block to execute a program are stored in each storage unit via a network, removable media (optical disks, flash memory, etc.), wireless communication, etc., and the servers (including smartphones and personal computers) corresponding to each block have interfaces connecting them to the storage units. They may also be equipped with mouse and keyboard interfaces for issuing installation instructions, etc. The part measurement process 30, repair feasibility assessment process 40, and part repair process 50 shown in Figure 2 are described in detail below using the block diagram shown in Figure 3.

[0025] In the imaging step 31 shown in FIG. 2, the imaging unit 2 shown in FIG. 3 captures an image of a component to acquire image data. The imaging unit 2 may be, for example, a camera mounted on a smartphone or tablet. The captured image data is transmitted to the image analysis unit 3 via the network 100. A smartphone will be used as an example of the imaging unit 2. Here, a smartphone is a computer that has imaging and communication functions and can be used by installing an app (application, software). To improve the accuracy of the diagnosis of the need for coating removal in the image analysis step 32 described later, it is preferable to capture the component in the imaging step 31 by adjusting imaging conditions such as the imaging direction, distance from the component, and lighting conditions. Therefore, for example, by installing an app for capturing component images, capturing images according to the app's instructions, and transmitting the image data, anyone can easily perform the work and improve work efficiency. Such an app is stored in the memory unit 2d. The captured image data is also stored in the memory unit 2d. In addition to acquiring image data, the imaging unit 2 may also acquire three-dimensional shape information using a LiDAR (Light Detection and Ranging) function, which has recently been installed in smartphones.

[0026] The imaging unit 2 may capture images of the component under multiple imaging conditions, such as a set of images captured from the front and from an oblique angle, a set of images captured from multiple angles, or a set of image data when the lighting direction is from the front and from an oblique angle.

[0027] FIG. 4(b) shows an example of a component whose surface is covered with an uneven coating. For example, by acquiring multiple image data captured under different imaging conditions, such as by capturing images of the coating region 150b on the surface of the component 150 using multiple cameras 2e, capturing images under multiple lighting conditions, or capturing images from multiple angles, it becomes possible to three-dimensionally capture the unevenness of the coating. In this case, the image analysis unit 3 can more accurately determine the condition of the coating, such as the unevenness of the coating region 150b. This allows the image analysis unit 3 to improve the accuracy of estimating the damaged region 150c and diagnosing whether or not the coating needs to be removed.

[0028] In the image analysis step 32 shown in Fig. 2, the image analysis unit 3 estimates the coating area based on the image data and determines the state of the coating. Examples of the state of the coating include the color, thickness, unevenness (surface roughness), area, etc. of the coating. Then, it is determined whether or not the coating needs to be removed.

[0029] The image analysis unit 3 also determines the measurement area while diagnosing whether or not the coating needs to be removed. For example, the image analysis unit 3 estimates the damaged area and the degree of damage based on the determined state of the coating, and determines the measurement area based on the degree of damage. This makes it possible to measure only areas with significant damage, thereby reducing the work time compared to measuring the entire part. Note that the method for determining the measurement area is not limited to this. For example, if the degree of damage in the area not covered by the coating is significant, the area not covered by the coating may be included in the measurement area, or the entire surface of the part may be the measurement area regardless of the degree of damage.

[0030] If the coating covers the measurement area and interferes with measurement, the image analysis unit 3 determines that coating removal is necessary, and the process proceeds to coating removal preparation step 33. If the coating does not cover the measurement area, or if the coating covers the measurement area but does not interfere with measurement, as in Figure 4(c) described below, for example, the image analysis unit 3 determines that coating removal is not necessary, and the process proceeds to measurement step 35. The image analysis unit 3 may also determine whether coating removal is necessary based on multiple image data acquired by the imaging unit 2 through imaging under multiple different imaging conditions.

[0031] A surface analysis program for carrying out the image analysis process 32 is stored in the memory unit 3d of the image analysis unit 3 shown in Figure 3, and when an instruction to start the surface analysis program is given, the image analysis unit 3 reads the image data and starts the image analysis process 32.

[0032] The image analysis unit 3 analyzes the surface condition of the part based on the image data and estimates the coating area using color, brightness (brightness difference, contrast within the image), the difference between specular and rough surfaces (difference in surface roughness, etc.), surface scattering, etc. For example, it can extract all areas with a color or brightness different from the background (areas that are not parts in the image data), and detect a coating area if the color or brightness differs from the background by more than a certain threshold. For example, the image data can be represented as 256-level RGB values, and areas where all R, G, and B values ​​are below a certain value can be considered a coating area.

[0033] In addition, the image analysis unit 3 estimates the approximate thickness and unevenness (surface roughness) of the coating by analyzing image data captured from two directions or from an oblique angle so that the lighting during imaging casts a shadow on the three-dimensional object.

[0034] FIG. 4(a) is a diagram showing an example of a part whose surface is covered with a flat coating. In FIG. 4(a), part 150 is an iron part, and the coating in coating region 150b is a flat coating applied to the surface of part 150. Based on image data acquired by imaging with camera 2e, if the coating in coating region 150b is in a clean state with no peeling or unevenness as shown in FIG. 4(a), it can be assumed that there are no areas with significant damage, and image analysis unit 3 may determine that removal of the coating in coating region 150b is unnecessary. Also, if the coating in coating region 150b is uniform and the shape of the part can be measured through the coating, it may be determined that removal of the coating is unnecessary. However, even if the coating in the coating region 150b is clean and uniform, if it is necessary to remove the coating in the coating region 150b from the material data of the part 150 and the coating in the coating region 150b and measure the part 150, for example, if it is desired to use the entire part as the measurement region or to compare it with a normal region, the image analysis unit 3 may diagnose that removal of the coating in the coating region 150b is necessary.

[0035] FIG. 4(c) shows a case where a thin oxide film or other coating is present instead of a paint film. If the coating is sufficiently thin, such as a thin oxide film, interference fringes 160 may appear in the image data, as shown in FIG. 4(c). In this case, the image analysis unit 3 can determine from the image data that the coating is a thin oxide film. Preferably, the image analysis unit 3 is equipped with AI (artificial intelligence) that has machine-learned the correlation between the interference fringes 160 and the coating condition. Furthermore, if the coating is sufficiently thin, the component may be visible through the image data. In this case, the image analysis unit 3 can also determine from the image data that the coating is a thin oxide film. If the coating is sufficiently thin, its thickness is on the order of nanometers, meaning that the coating is unlikely to affect measurement. Therefore, in this case, the image analysis unit 3 may determine that removal of the coating is unnecessary and proceed to the measurement step 35. If the type of oxide film is known from the component's material, the oxide film thickness may be accurately determined based on the principles of ellipsometry.

[0036] When the image analysis unit 3 diagnoses that removal of the coating is unnecessary, the number of coating removal steps is reduced, and the work efficiency of the part measurement step 30 is improved accordingly.

[0037] In the image analysis step 32 shown in Figure 2, the diagnosis of whether or not the coating needs to be removed can be made by each worker. However, if the decision is made using AI that has learned the correlation between image data and the state of the coating through machine learning, the decision will no longer depend on the person making the decision, and it will be possible to always make a diagnosis at the same level.

[0038] Next, in the image analysis step 32 shown in Fig. 2, the damaged area where the shape of the part has changed and the extent of the damage are estimated. Here, the estimation of the damaged area and the extent of the damage of the part inside the coating will be described. The method for estimating the damaged area and the extent of the damage in the area without the coating is not limited to this.

[0039] When estimating the damaged area and the extent of damage, the parameters used to estimate the coating area and the parameters obtained by estimating the coating area are used. Information related to the coating, such as the parameters used to estimate the coating area and the parameters obtained by estimating the coating area, as well as image data of the coating, is referred to as coating data. For example, parameters used to estimate the coating area include color, brightness (brightness difference within the image, contrast), the difference between a specular and rough surface (difference in surface roughness, etc.), and surface scattering, while parameters obtained by estimating the coating area include the shape of the coating area, the area area, and the maximum thickness of the coating.

[0040] Various methods for estimating the degree of damage are conceivable. For example, a threshold value can be set for coating data, such as the color, contrast, and area of ​​the coating region. This threshold value can be determined in advance based on the correlation between the coating data, which has been measured or calculated in advance, and the degree of damage. For example, if multiple samples are measured and the degree of damage is severe enough to require measurement, the R value threshold can be set to 128, since the R value of the RGB color of the coating region in all samples exceeds 128 gradations. In this way, the degree of damage can be estimated across the entire component surface, and the area where the degree of damage exceeds the predetermined threshold can be determined as the damaged area. Alternatively, the damaged area can be estimated based on the image data and the condition of the coating, and then the degree of damage can be estimated for the damaged area. In this case, the area for which the degree of damage is evaluated is limited to the damaged area, thereby improving work efficiency.

[0041] Furthermore, using AI techniques such as machine learning to estimate the degree of damage can enable more accurate estimation. The aforementioned estimation method of setting a threshold for coating data involves defining the color of the coating as a feature, calculating a threshold for each feature, and estimating the degree of damage based on the magnitude relationship with the threshold. This estimation can also be replaced by a machine learning prediction model. That is, the image analysis unit 3 can use a prediction model previously constructed based on the correlation between the coating data and the shape change of the component inside the coating to estimate the degree of damage from image data based on the degree of change in component shape. This allows for the quantity related to the degree of damage to be expressed numerically, such as by probability, rather than by classification as to whether or not measurement is required, as described below, enabling more detailed diagnosis and improving diagnostic accuracy. A machine learning prediction model can be obtained by training a specific machine learning algorithm to correlate the coating data with the degree of damage.

[0042] In machine learning methods, the model can be trained using image data of the coating. A prediction model can be obtained by training the correlation between the image data of the coating and the degree of damage. In this case, both the estimation of the damaged area and the degree of damage can be automated in one go.

[0043] In addition, in a method using machine learning, image data of the coating that is determined to require measurement may be classified in advance from images that do not require measurement, and then each may be trained to build a prediction model. In this case, too, it is possible to automate both the estimation of the damaged area and the extent of the damage in one go.

[0044] Estimating the degree of damage, including methods using AI technology, requires a prediction method that is pre-constructed based on the correlation between coating data and changes in the shape of the parts inside the coating. The AI ​​image analysis program may be called by the surface analysis program from a calculation server (cloud) or may be implemented within the surface analysis program. The image analysis unit 3 may have a function to receive input of a set of damage level data and coating data in order to construct or update a prediction model or method used to estimate the degree of damage.

[0045] Damaged areas may exist not only in areas covered by a coating on the surface of the part, but also in areas not covered by a coating, or in both. For example, if part of the surface of an iron part is covered with a rust coating, there may be areas damaged by the rust and scratches in areas without rust. In this case, it is necessary to diagnose whether both areas are repairable and determine whether they can be repaired. Therefore, the image analysis unit 3 checks the areas without a coating for damage that clearly makes them unrepairable. If it can diagnose the areas without a coating as clearly unrepairable due to severe damage, etc., the process proceeds to the abort step 200, where the product is reused as raw material (material recycling) or, if the damage is severe, is discarded.

[0046] Furthermore, if there is no damage that is clearly beyond repair, the damaged area and degree of damage in the uncoated area are estimated. In this case, the damaged area and degree of damage are estimated from the surface condition of the part based on image data. For example, a correlation between image data and the degree of damage may be obtained for the uncoated area as well, and the damaged area and degree of damage may be estimated from the correlation between the image data and the degree of damage. In the uncoated area, the degree of damage is correlated with the brightness in the image data, the brightness difference (contrast) from the surrounding area, the difference between a specular surface and a rough surface (difference in surface roughness, difference in surface scattering), etc.

[0047] Furthermore, if the correlation between image data and the degree of damage is obtained in advance for both coated and uncoated regions, and a prediction model or prediction method is constructed based on the obtained correlation, the same measurement system or parts reconditioning system can be used for both coated and uncoated regions. This solves the problems of complicated work for workers, increased learning required to understand the system, and an expansion in system scale that arise when different measurement systems or parts reconditioning systems are used for coated and uncoated regions.

[0048] The image analysis unit 3 classifies the damaged area as an area of ​​high damage level or low damage level. Alternatively, multiple ranks may be provided according to the degree of damage, and the estimated degree of damage may be classified into each rank. For example, a rank indicating no damage has occurred, a rank indicating damage has occurred but repair is not required, a rank indicating repair is required, a rank indicating repair is not possible, etc.

[0049] The image analysis unit 3 determines the measurement area based on the estimated damage level. For example, the measurement area may be determined so that it includes the area estimated to have the most severe damage among the estimated damaged areas. Furthermore, if the damage level is classified into ranks, the measurement area may be determined so that it includes areas with a specific damage level. Accurately estimating and measuring areas with severe damage improves the accuracy of determining whether repair is possible. Furthermore, when determining whether repair is possible, it is also important to estimate that there are no areas with severe damage. The measurement area may be determined automatically by the image analysis unit 3, or, if the damage level is quantified, the worker may determine it by looking at it. The worker may also determine it by checking the condition of the damaged area.

[0050] If the coating covers the measurement area and interferes with measurement, the image analysis unit 3 determines that the coating needs to be removed, and the process proceeds to coating removal preparation step 33. If the coating does not interfere with measurement, the image analysis unit 3 determines that the coating does not need to be removed, and the process proceeds to measurement step 35. The necessity of coating removal may also be determined based on the degree of damage. For example, the image analysis unit 3 may determine that removal of the coating is not necessary in an area where the degree of damage is estimated to be small. Furthermore, if the coating interferes with measurement in an area where the degree of damage is estimated to be large, the image analysis unit 3 determines that removal of the coating is necessary. If the change in shape of the part inside the coating can be measured from above the coating because the coating thickness is uniform, for example, measurements may be performed from above the coating without removing the coating, and the degree of damage to the part inside the coating may be estimated based on the measurement results to determine whether repair is possible. Note that if the measurement area is determined regardless of the degree of damage, estimation of the degree of damage and damaged area may be omitted.

[0051] As described above, by using a prediction method and a prediction model that are pre-established based on the correlation between coating data and changes in the shape of the components inside the coating, and estimating areas with high damage levels from captured image data, it is possible to automate the setting of measurement areas, thereby improving work efficiency. Note that various thresholds, parameters related to the prediction method, and coating data are stored in the storage unit 3d.

[0052] In the coating removal preparation step 33 shown in Figure 2, after it is diagnosed that the coating needs to be removed, the coating area diagnosed as needing removal is used as the coating removal area to prepare for the coating removal step 34, in which the coating is removed. Specifically, the processing conditions of the surface processing unit 4 when removing the coating are determined based on at least the state of the coating. The coating removal preparation step 33 may be performed by any of the imaging unit 2, the image analysis unit 3, the processor of the surface processing unit 4, etc., or may be performed in conjunction with each other. The surface processing conditions are determined based on information obtained from the image data, such as the state of the coating, coating data, damaged areas, the degree of damage, and measurement areas. The information required for the processing conditions depends on the processing device.

[0053] When the processing device 4e included in the surface processing unit 4 shown in Figure 3 is a polishing / grinding machine such as a grinder, the thickness of the coating to be removed is estimated from the degree of damage in the measurement area in the coating removal preparation process 33. Then, in the coating removal process 34, the processing device 4e polishes and grinds the coating according to the estimated thickness. The coating thickness and depth are estimated by the image analysis unit 3 and transmitted to the surface processing unit 4. However, the configuration is not limited to this; the estimation can be performed by a processor in the surface processing unit 4, a processor in the measurement unit 5, or a processor in the repair diagnosis unit 6. Other processors not described above are also acceptable. In particular, the image analysis unit 3 and the repair diagnosis unit 6 may share a processor, and various physical configurations are possible. The surface processing unit 4 may automatically process the coating based on the transmitted values, or an operator may remove the coating based on the estimated coating thickness. The parameters input to operate the processing device are stored in the memory of either the image analysis unit 3, the surface processing unit 4, or the repair diagnosis unit 6.

[0054] When the processing device 4e is a laser removal device, the processing conditions are laser irradiation conditions determined based on the material information of the part. Therefore, the laser irradiation conditions must be determined in the coating removal preparation step 33. The most important parameter among them is the energy density [J / m 2 ] and are related to energy [J], peak power [W], peak power density [W / m 2 ], pulse width [s], average power [W], wavelength [nm], etc. In laser removal equipment, the energy density is changed by changing any of these parameters. If the energy density does not exceed a certain value, the coating cannot be removed. On the other hand, if it exceeds a certain value, not only the coating but also the part will be damaged. Therefore, it is necessary to set the energy density to an appropriate value depending on the condition of the coating and the material of the part.

[0055] If information on the material of the part is available, it is possible to calculate the energy density at which ablation occurs and the part is damaged. Also, for the processing device 4e to be used, the relationship between the part material and the maximum energy density (or quantities related to energy density, such as peak power and pulse width) may be obtained in advance through experiments as part material information.

[0056] The lower limit of the energy density (or a quantity related to the energy density) may be determined experimentally, but it may also be determined based on coating data extracted from image data, such as the thickness of the coating, using a database stored in a memory unit such as the image analysis unit 3 or the surface processing unit 4. Once the lower limit of the energy density is determined, it is sufficient to set the energy density slightly higher than the lower limit.

[0057] Alternatively, the energy density may be set to half the maximum energy density calculated from the component's material information. Metals, in particular, have high reflectivity and therefore are less likely to absorb laser light, so the power density that causes damage is often significantly higher than that of corrosion films such as rust. Therefore, when the material is metal, there is a high degree of leeway in setting the power density, making it possible to determine a wide range of energy densities, as described above.

[0058] There are various other possible methods for determining the laser irradiation conditions, but if the laser irradiation conditions can be determined based on the material information of the part or information analyzed from the image data in the image analysis unit 3, there will be no need for the operator to consider and determine the conditions each time the part is used, and the effect will be that coating removal by laser can be automated.

[0059] It is desirable that these laser conditions be displayed on the monitor that controls the processing device 4e, or on a smartphone using the coating removal app, or recorded in a file, etc. This will help prevent processing errors and make it possible to check the processing conditions at a later date.

[0060] In the coating removal step 34 shown in FIG. 2, when the image analysis unit 3 determines that removal of the coating is necessary, the surface processing unit 4 removes the coating in accordance with the coating removal region and processing conditions.

[0061] Although all processes may be automated without human intervention, if human intervention is required, the use of an app or the like to assist in coating removal can reduce work errors and improve work efficiency. For example, an app to assist in coating removal can be installed on a smartphone or computer, and the app can assist in coating removal by displaying the removal area along with the captured image and the estimated degree of damage. In this case, the smartphone screen serves as the display unit. This improves the work efficiency of coating removal by humans. Furthermore, if processing conditions and the like are also displayed, it is possible to prevent accidental damage to parts during coating removal. The app can be run not only on a smartphone, but also on the CPU 4b of the surface processing unit 4, or on another CPU. It is sufficient that it can operate during coating removal.

[0062] In the measurement step 35 shown in FIG. 2, the measurement unit 5 measures the shape of the part in the measurement region. Note that if a coating present in the measurement region interferes with measurement, the image analysis step 32 determines that the coating needs to be removed. In this case, the coating present in the measurement region is removed in the coating removal step 34. Therefore, if the image analysis step 32 determines that a region estimated to have a high degree of damage needs to have the coating removed, the measurement region is determined to include that region, and the coating in the region estimated to have a high degree of damage is removed in the coating removal step 34. Therefore, the measurement region in the measurement step 35 includes the region where the coating from the region estimated to have a high degree of damage has been removed.

[0063] The measurement method may be performed by an operator using a caliper or the like, or three-dimensional measurement may be performed using a laser displacement meter, laser profile measuring instrument, etc. In particular, lasers can perform three-dimensional measurement of a large area all at once by scanning. The measuring device 5e is a caliper, laser displacement meter, laser profile measuring instrument, etc.

[0064] If you are only judging whether repairs are possible, it may be sufficient to measure a few points with a caliper, but to improve the accuracy of the repair judgment, 3D measurement is preferable. Similarly, when diagnosing repair methods, 3D measurement is preferable because the amount of information obtained by using 3D measurement and the resulting shape data increases the accuracy of diagnosing repair methods.

[0065] If the coating is partially removed due to the influence of the coating, the removed area can be measured. Furthermore, the area estimated to be heavily damaged may be designated the first measurement area, and the area estimated to be lightly damaged may be designated the second measurement area. In this case, it is desirable that the second measurement area be sufficiently damaged that it can be considered normal. The measurement unit measures the relative shape of the part in the first measurement area by measuring the shape of the part in each of the first and second measurement areas. This relative measurement is highly accurate. Furthermore, since the second measurement area has only a small degree of damage, absolute dimensions can be measured with high accuracy. Therefore, if the coating is also removed in the second measurement area, the relative shape of the part in the first measurement area can be measured with higher accuracy. As a result, the absolute shape of the part in the first measurement area can be measured with higher accuracy. Furthermore, in the case of a very thin coating, such as that shown in Figure 4(c), the coating does not interfere with measurement, so the second measurement area may be measured without removing the coating. Alternatively, a region that is only slightly damaged and originally has no coating may be used as the second measurement region.

[0066] Removing all of the coating to measure the part's shape takes time. Therefore, rather than removing all of the coating, the work time can be shortened by removing only the coating from areas estimated to be severely damaged and then measuring them. Furthermore, by partially removing the coating and then measuring the relative shapes of the removed areas and areas with less damage, the work time can be shortened while improving measurement accuracy.

[0067] In the repair feasibility assessment step 40 shown in FIG. 2, the repair assessment unit 6 determines whether the part can be repaired based on the shape data (measurement results of the part's shape) obtained in the measurement step 35 and the part's design data, such as pre-recorded 3D CAD data. Specifically, the part's shape is inspected to see if it has changed significantly from the design data, and the part's repair feasibility is determined based on the inspection results. Various data required for the diagnosis are stored in the memory unit 6d. Furthermore, the memory unit 6d of the repair assessment unit 6 may share part of the memory unit 3d of the image analysis unit, and the data used for image analysis may be shared.

[0068] It should be noted that the 3D CAD, design drawings, etc. do not necessarily have to be stored in the storage unit 6d, but may be read into the memory 6c from another server via the network 100.

[0069] The diagnostic inspection involves a detailed examination of areas estimated by the image analysis unit 3 to have particularly severe damage. Damaged areas can be inspected not only by simple shape difference but also by material component analysis using laser-induced breakdown spectroscopy (LIBS) to confirm that the components fall within material specifications. In other words, by performing multiple inspections, such as not only shape inspection but also component inspection, on areas with severe damage estimated by the image analysis unit 3, it becomes a targeted inspection that does not inspect the entire surface multiple times or mistakenly inspect only areas with less damage, which has the effect of improving work efficiency.

[0070] The repair diagnosis unit 6 determines whether the product can be repaired based on these inspections. If the product can be repaired, the process proceeds to the part repair process 50. If the product cannot be repaired, the process proceeds to the abort process 200, where the product is reused as a raw material (material recycling) or is discarded if the damage is severe.

[0071] The part repair process 50 shown in Figure 2 will now be described. First, in the cleaning necessity diagnosis process 51, the repair diagnosis unit 6 diagnoses whether cleaning is necessary, and if necessary, proceeds to the cleaning process 52. Here, cleaning refers to the surface processing unit 4 removing all of the coating remaining on the part surface when the image analysis process 32 diagnoses that coating removal is unnecessary or when the coating covering the part surface is partially removed in the coating removal process 34. In the cleaning necessity diagnosis process 51, diagnosis can be made based on previously acquired data, such as the coating data obtained by the image data and image analysis, the coating area, damaged area, damage level, measured surface shape, and inspection data used for repair diagnosis. In addition, in the cleaning necessity diagnosis process 51, the reprocessing conditions of the surface processing unit 4 when removing the coating are determined based at least on the condition of the coating. Furthermore, if the surface processing unit 4 is a laser removal device, the reprocessing conditions may be the same laser irradiation conditions as the processing conditions.

[0072] In this way, the type of processing device and processing conditions (reprocessing conditions) for removing the coating in the cleaning step 52 may be the same as those used in the coating removal step 34 before measurement.

[0073] The repair method diagnosis process 53 shown in Figure 2 will now be described. Similar to the repair feasibility diagnosis process 40, the repair diagnosis unit 6 determines the part repair method based on the part shape measurement results and pre-recorded part design data. Repair methods include processes such as thermal spraying, build-up, cutting, remelting, and heat treatment. A single process or multiple processes may be performed. The selection of the process, the process conditions, and the process order are determined in the repair method diagnosis process 53. The repair method diagnosis can be performed based on previously acquired data, such as the image data and coating data obtained through image analysis, the coating area, damaged area, damage level, measured surface shape, and inspection data used for repair diagnosis. In particular, when repair is performed using a laser, utilizing image data and the optical characteristics obtained from it, such as color and brightness, simplifies the setting of laser irradiation conditions. Once the repair method is determined, the process proceeds to repair process 54.

[0074] 2, the repairing section 7 repairs the part using the previously determined repairing method. Specifically, the repairing section 7 repairs the part using a repairing device 7e.

[0075] 2, the cleaning necessity diagnosis step 51 and the repair method diagnosis step 53 are separate steps, but the present invention is not limited to this, and it is also possible to determine the necessity of cleaning and the repair method at the timing of the cleaning necessity diagnosis step 51, and then proceed to repair including cleaning. There are various possibilities for the order from the cleaning necessity diagnosis step 51 to the repair step 54.

[0076] After the repair, the part proceeds to the parts inspection process 60, where the repaired part is inspected. This parts inspection is an inspection to determine whether the repaired part has the characteristics and dimensions expected at the time of diagnosis in the repair method diagnosis process 53, and whether there are any problems when it is assembled as part of a product.

[0077] If it is determined in the component inspection process 60 that assembly is not possible, the process returns to the repair method diagnosis process 53 and the repair is redone. In some cases, the process may return to the cleaning necessity diagnosis process 51.

[0078] If it is determined in the part inspection process 60 that assembly is possible, an assembly process 70 is carried out to assemble the parts, and then a product inspection process 80 is carried out to inspect whether there are any problems with the product. If there are no problems in the product inspection process 80, the product proceeds to a shipping process 90. If the product inspection process 80 determines that it cannot be shipped, the product is reassembled, possibly using new parts.

[0079] In the past, when there was corrosion such as rust or a coating such as paint, workers would determine the condition of the coating to check for damage and measure the extent of the damage, and if removal of the coating was necessary, they would remove it by dissolving or scraping it with a chemical solution and then measure the part.

[0080] However, in the present disclosure, the determination of the state of the coating is automated, thereby ensuring the accuracy of the diagnosis of whether or not the coating needs to be removed and reducing the workload of the worker.

[0081] Furthermore, in recent years, lasers have become widely used for a variety of tasks, from measurement to component cleaning and build-up welding. However, because high-intensity lasers are used, direct viewing and even scattered light can be extremely dangerous. This disclosure also makes it possible to automate coating removal, reducing the number of times workers have to work directly with a laser, thereby improving worker safety. Furthermore, automation from the imaging process 31 to the repair process 54 is possible, improving work efficiency.

[0082] As described above, the present disclosure makes it possible to provide a measurement system and method, as well as a parts remanufacturing system and method, that can determine whether or not a coating needs to be removed and measure the shape of a part, even if the surface of the part is covered with a coating. (Embodiment 2) Fig. 5 is a diagram showing an example of use of the parts recycling system according to embodiment 2. A case where rust corrosion is present as a coating on an iron part will be described using Fig. 5. Note that a description of the same configuration as in embodiment 1 will be omitted.

[0083] FIG. 5(a) is a diagram showing an example of the imaging step 31. An iron part 150 has a normal area 150a where the part is not covered with rust and a coated area 150b where the part is covered with rust. It is assumed that the normal area 150a is only slightly damaged. An example will be described in which the imaging unit 2 is a smartphone, and the imaging step 31 is performed using an imaging app. It is assumed that in this embodiment, the screen of the smartphone functions as the display unit.

[0084] To perform the imaging process 31, the worker launches an imaging app on the smartphone. Following the instructions of the app displayed on the smartphone screen, the worker images the component 150 using the smartphone's camera function. If the image is captured properly, the app displays "OK" on the smartphone screen, saves the image data of the image 101, and transmits it to the image analysis unit 3 via the wireless interface 2a and the network 100. If the image is not proper, the app instructs the worker to retake the image.

[0085] FIG. 5(b) shows an example in which a damaged area 150c estimated by the image analysis unit 3 is displayed in the image 101. The display unit displays the damaged area on the image captured by the imaging unit 2, as shown in FIG. 5(b). FIG. 5(b) shows an example in which three damaged areas 150c exist, and the degree of damage to the part due to rust corrosion is displayed numerically next to each damaged area 150c. The operator may set the measurement area based on the numerical value of the degree of damage, or a threshold value for the numerical value of the degree of damage may be set in advance, and the area where the numerical value of the degree of damage is greater than the threshold may be set as the measurement area. Alternatively, the measurement area may be automatically determined without displaying the damaged area, and only the measurement area may be displayed, or both the damaged area and the measurement area may be displayed. The process may proceed to the coating removal preparation step 33 or the measurement step 35 without displaying such information. However, by displaying the damaged area or measurement area, the worker can visually check the relevant parts of the part, so displaying the damaged area or measurement area has the effect of providing a double check by both the machine and the person, preventing coating removal or measurement in the wrong places.

[0086] In this embodiment, the image analysis unit 3 diagnoses in the image analysis step 32 that at least a portion of the component 150 does not require coating removal, and diagnoses areas where the degree of damage is estimated to be large as requiring coating removal. In this embodiment, the image analysis unit 3 diagnoses only the portions of the damaged area 150c where the damage degrees shown in FIG. 5(b) are 1.2 and 1.0 as requiring coating removal, and diagnoses the other coated areas 150b as not requiring coating removal. In this case, the image analysis unit 3 designates the two damaged areas 150c diagnosed as requiring coating removal as measurement areas 150d.

[0087] FIG. 5(c) is a diagram showing an example of the coating removal step 34. The processing device 4e according to the second embodiment is a laser removal device. In FIG. 5(c), the processing device 4e removes the coating in the measurement area 150d, which is an area diagnosed as needing removal of the coating. Note that, in the coating removal preparation step 33, the output of the laser light 4eRay irradiated by the processing device 4e is set sufficiently lower than the power that would cause ablation and damage to the iron, based on the material information of the part that the part is made of iron. In addition, the output of the laser light 4eRay is adjusted based on the state of the coating estimated from the color of the rust.

[0088] 5(d) is a diagram showing an example of the state of the measurement process 35. The measurement device 5e according to the second embodiment is a laser shape measurement device that performs three-dimensional measurement by irradiating the component surface with laser light 5e Ray. By simultaneously measuring the normal region 150a, in which the dimensional difference from the design value can be determined with high precision, and the measurement region 150d within the coating region 150b, the relative difference between the simultaneously measured regions can be determined with high precision, thereby achieving the effect of enabling the degree of damage in the damaged region 150c to be estimated with high precision.

[0089] FIG. 5(e) is a diagram showing an example of the cleaning step 52. In the cleaning step 52, if the repair diagnosis unit 6 determines that the component can be repaired, the surface processing unit 4 removes the coating that the image analysis unit 3 has determined not to require removal. In this embodiment, the coating region 150b present on the surface of the component 150 in FIG. 5(d) is removed. In FIG. 5(e), the coating region 150b is removed under the same laser irradiation conditions as in the coating removal step 34 shown in FIG. 5(c) using the laser removal device 4e.

[0090] FIG. 5(f) is a diagram showing an example of the repair process 54. The repair device 7e of the repair unit 7 repairs the part by welding, depositing, or the like using the laser light 7eRay. Here, the conditions for the laser light 7eRay can be determined based on, for example, the image 101 and the results obtained from it. Alternatively, they may be determined based on machine-learned data stored on the cloud.

[0091] After the repair, measurements are taken again, and the repair method used in the repair process 54 and the results (image 101, 3D shape, etc.) are transferred to the server. The server builds a database linking the repair method and the results, and periodically rebuilds the machine learning model. By sequentially updating the machine learning model using the repair method and the results, it is possible to determine the repair method with greater accuracy. The flow of transferring the results to the server after measurement and using them for machine learning can be implemented as a function in an app, and setting the flow conditions, etc. according to the app's GUI makes the work easier and improves work efficiency. (Embodiment 3) FIG. 6 is a diagram illustrating an imaging process according to the third embodiment. To accurately determine the coating condition and estimate the extent of damage based on image data, it is necessary to image the component under appropriate imaging conditions according to the coating condition. Therefore, the imaging unit 2 according to this embodiment has a function to assist in imaging so that an image that makes it easy to determine the coating condition is captured under appropriate imaging conditions. If the imaging unit 2 is a computer, smartphone, or the like, the imaging assistance function may be provided as an app. In other words, the measurement system or component remanufacturing system may include a computer or smartphone with an app for assisting in imaging installed. The imaging assistance will be specifically described below with reference to FIG. 6. Note that a description of the same configuration as in the first or second embodiment will be omitted.

[0092] An example in which the imaging unit 2 interactively (interactively / two-way) supports imaging will be described. As a function of supporting imaging, for example, the imaging unit 2 transmits image data obtained by imaging to the image analysis unit 3, and if the image analysis unit 3 determines that the image data is not good, it instructs the imaging unit 2 to retake the image under appropriate imaging conditions. In this embodiment, an example in which the imaging unit 2 has a display unit and instructs the imaging unit 2 through the display on the display unit will be described. However, this is not limited to this, and instructions may also be given by voice or the like. The quality of the image data is determined based on whether the information contained in the image data is sufficient for determining the state of the coating, estimating the degree of damage, estimating the damaged area, etc. For example, if the thickness of the coating cannot be determined from the image data or if the lighting environment is too dark to determine the color of the coating, the image data is determined to be bad.

[0093] The display unit according to this embodiment displays the image captured by the imaging unit 2, as well as the quality of the image data and instructions for re-imaging. Examples of the quality of the image data include a graded evaluation display such as Excellent, Good, and Poor. The instructions for re-imaging may be a display simply urging the user to take the image again, such as "Try taking another image," or a display giving advice on what is lacking, such as "Try enlarging the image and taking an image" or "Try taking an image from multiple directions."

[0094] When the imaging unit 2 assists in imaging and sufficient image data is acquired, the accuracy of determining the state of the coating based on the acquired image data improves, and the accuracy of estimating the degree of damage and the measurement area based on the state of the coating improves.

[0095] The imaging unit 2 supports imaging, making it possible to acquire appropriate image data without relying on the operator. Furthermore, if image data is acquired that does not allow the state of the coating to be determined, the imaging is performed again, which reduces unnecessary coating removal or repair based on image data that does not allow the state of the coating to be determined, improving work efficiency.

[0096] Fig. 6 is a diagram illustrating an imaging process 31 according to embodiment 3. In embodiment 3, a specific example will be described using Fig. 6 in which a part reproduction system and a person interactively collaborate to capture an image of a part via an app installed in the imaging unit 2.

[0097] Fig. 6(a) is a diagram showing an example of the imaging step 31. Fig. 6(a) shows the state in which a component 150 having a normal region 150a and a coated region 150b is imaged by the camera 2e of the imaging unit 2.

[0098] FIG. 6(b) is a diagram showing an example of a display by the display unit. Here, an example will be described in which the imaging unit 2, which is a smartphone, has a display unit, but this is not limiting. As shown in FIG. 6(b), the display unit displays an image 170a captured by the imaging unit 2, as well as text 170b asking whether or not to transfer the image 170a to the image analysis unit 3, and a button 170c for responding to the text 170b. If the Yes button 170c is pressed, the image data of the captured image 170a is transferred to the image analysis unit 3. If the No button 170c is pressed, the image is captured again. In this way, the operator is asked to confirm whether or not to transfer the image data before transferring the image data to the image analysis unit 3, but this may be omitted.

[0099] FIG. 6(c) shows that after image 170a is transferred in FIG. 6(b), the image analysis unit 3 determines that the information contained in the image data is insufficient to estimate the damaged area, and displays text 170d and an arrow 170e on the display unit to instruct the imaging unit 2 to capture an image from a different direction.

[0100] FIG. 6(d) shows the state in which the camera 2e of the imaging unit 2 captures an image of the component 150 from the direction of the arrow 170e shown in FIG. 6(c).

[0101] FIG. 6(e) is a diagram showing an example of a display by the display unit. Image 170f shown in FIG. 6(e) is an image captured by camera 2e of imaging unit 2. Imaging unit 2 transfers the image data of image 170f to image analysis unit 3, which then estimates the damaged area 150c and the degree of damage in each damaged area 150c. The display unit displays the damaged area 150c estimated by image analysis unit 3 on image 170f, as well as the degree of damage in each damaged area 150c as a numerical value. At this time, text 170g indicating whether the imaging was performed accurately and whether other damaged areas have been visually confirmed by the operator is displayed on the display unit to confirm that the image data is correct. A button 170h is also displayed for responding to text 170g. Pressing "YES" on button 170h causes the imaging to be redone; pressing "No" causes the system to proceed to a diagnosis of whether coating removal is necessary and to the determination of the measurement area.

[0102] FIG. 6(f) shows the display unit instructing the operator to enlarge and capture one of the damaged areas 170i. Here, the display unit displays not the captured image but the video captured in real time by the camera 2e of the imaging unit 2. The upper left corner of the display unit displays the position to be captured on a sub-screen 170h. For example, the sub-screen 170h displays the position to be captured as a dotted circle enlarged on an image previously captured by the imaging unit 2. When the focus is correct, text 170j indicates that the focus is correct. The operator confirms the text 170j and, if deemed acceptable, presses the image capture button 170k to capture the image. Enlarged imaging is not necessarily required, but may be performed by default to improve the accuracy of the diagnosis of whether or not the capsule needs to be removed. Enlarged imaging may also be performed when the image analysis unit 3 determines that the diagnostic probability (diagnostic reliability) of whether or not the capsule needs to be removed is low. The diagnostic probability of whether or not the coating needs to be removed may be displayed (not shown), and the operator may make a decision based on the displayed diagnostic probability to perform enlarged imaging.

[0103] As shown in this embodiment, by using marks such as letters and arrows on the display unit to assist in imaging, it is possible to obtain an image that makes it easier to determine the condition of the coating and improves the accuracy of diagnosing whether or not the coating needs to be removed. In this embodiment, the imaging direction is indicated by the display unit of the imaging unit 2, but this is not limited to this, and the lighting direction or lighting environment may also be indicated. By using such a measurement system, the quality of the image data is less dependent on the operator, and work rework is less likely to occur, improving work efficiency.

[0104] In this embodiment, an example has been shown in which the imaging unit 2 instructs the operator to take images while actually changing the position of the imaging unit 2, but the imaging unit 2 may also be automatically moved by a stage, robot arm, or the like. By creating a flowchart including the operations shown in Figures 6(a) to 6(f) and operating the measurement system according to the created flowchart, full automation, including re-imaging, is possible. In other words, a flowchart can be created with some or all of the components, such as imaging, image confirmation, imaging from another direction, estimation of the damaged area and extent, and enlarged imaging of the damaged area, and branching, such as skipping enlarged imaging, can be included.

[0105] Although a smartphone is used as an example of a device for giving instructions to a worker in this embodiment, smart glasses may also be used. Smart glasses are glasses-type devices equipped with a camera and an image display unit. When worn, they use AR (Augmented Reality) technology to display text, icons, images, and other information as if they were right in front of the worker's eyes. The use of AR technology allows workers to intuitively adjust the imaging position, thereby improving work efficiency. Furthermore, workers can work with both hands free and can check their surroundings while working, thereby improving safety.

[0106] As described above, if imaging is performed according to instructions from the imaging unit 2, it is easy to reproduce the best imaging conditions, thereby ensuring diagnostic accuracy regarding the need for capsule removal regardless of who performs the imaging, and further improving work efficiency. Furthermore, by configuring imaging to be performed automatically according to imaging instructions from the imaging unit 2, the imaging process can be automated. (Embodiment 4) FIG. 7 is a diagram illustrating an imaging process according to the fourth embodiment. In this embodiment, the imaging unit 2 simultaneously images a reference and a component. Here, the reference represents a standard for color, brightness, etc., and an example of such a reference is a color sample. An example of a reference is a card printed with R (red), G (green), B (blue), white, and black. By correcting the image data based on the color and brightness of the reference imaged simultaneously with the component, the influence of the lighting environment can be reduced, and the accuracy of diagnosing whether or not the coating needs to be removed can be improved.

[0107] Therefore, in this embodiment, the imaging step 31 for simultaneously imaging the reference and the component will be specifically described with reference to Fig. 7. In Fig. 7, 180 is the reference symbol. Note that the description of the same configuration as in the first to third embodiments will be omitted.

[0108] 7(a) is a diagram showing an example of the imaging step 31 in which a reference 180 is placed on the surface of the component 150 and an image is taken. The location where the reference 180 is placed is not limited to this, but by placing the reference 180 near the coating region 150b that is the target of diagnosis for determining whether or not the coating needs to be removed, the lighting environment of the target of diagnosis can be accurately corrected.

[0109] A method for correcting an image obtained by simultaneously capturing the reference 180 and the component 150 will now be described. First, a color transformation matrix is ​​calculated so that the color of the reference in the image obtained by simultaneously capturing the reference 180 and the component 150 in the imaging step 31 matches the color of the reference (the vector (R, G, B, I) of RGB values ​​and luminance I) that was previously set as a standard. Then, a correction is performed by converting the entire captured image using the calculated color transformation matrix. By using the image data of the corrected image to diagnose whether or not the coating needs to be removed, the diagnosis can be made while minimizing the effects of differences in lighting environments, particularly differences in the color of the illumination light.

[0110] Furthermore, when obtaining the correlation between image data and the degree of damage, the component is imaged with the reference 180 placed. Then, correction is performed in the same manner as before, and the correlation between the image data of the corrected image and the degree of damage is obtained. This makes it possible to obtain the correlation between image data and the degree of damage while suppressing the effects of differences in lighting environments, particularly differences in the color of the lighting light.

[0111] In this way, by determining a reference color (e.g., R, G, B, I), simultaneously capturing an image of the reference and the component, and correcting the image data based on a comparison between the reference color and the captured reference color, differences in the lighting environment, particularly the color of the illumination light, can be suppressed, resulting in improved accuracy in diagnosing whether or not coating removal is necessary. Note that although color data (R, G, B, I) including luminance is used as the reference, RGB values ​​(R, G, B) alone may also be used. Other color indices, such as chromaticity or wavelength dispersion, may also be used.

[0112] Figure 7(b) shows an example of a reference 180, which is a highly reflective white paper. A standard diffuse reflector that uniformly scatters and reflects light in all directions may also be used. If a material with high reflectivity over a wide wavelength range is used, it will reflect the color of the ambient light as it is, so differences in ambient light can be reflected in detail.

[0113] FIG. 7(c) is a diagram showing an example of a reference 180, in which a black portion 180bl is present within a white portion 180w. Using the reference 180 shown in FIG. 7(c) makes it easy to extract changes in contrast. Furthermore, if there are marks such as black dots in the four corners, the position of the reference can be easily found within the image. If there are marks not only in the four corners, it is easy to extract the position of the reference.

[0114] FIG. 7(d) shows an example of a reference 180, with a white section 180w in the center, surrounded by a black section 180bl, a red section 180r, a green section 180g, and a blue section 180b. Using the variations in each color makes it easier to derive a color transformation matrix, shortening the time required to generate corrected image data and reducing work time. While it is best to have all colors, it is also effective to have only white, red, and green, as well as a few other colors. The colors can be selected depending on the printing environment in which the reference will be produced. [Explanation of symbols]

[0115] 1...Parts remanufacturing system 2...Image capture unit 2a...Imaging unit interface (I / F) 2b...Camera processor (CPU) 2c...Memory of the imaging unit 2d...Memory section of the imaging section 2e…Camera 3...Image analysis section 3a...Image analysis unit interface (I / F) 3b...Image analysis processor (CPU) 3c...Image analysis unit memory 3d...Memory section of image analysis section 4...Surface processing section 4a...Interface (I / F) of surface processing part 4b...Processor (CPU) of the surface processing unit 4c…Memory of surface processing part 4d…Memory section of surface processing section 4e…Surface processing equipment 5...Measuring section 5a...Measuring unit interface (I / F) 5b...Measuring unit processor (CPU) 5c…Measuring unit memory 5d…Measuring unit memory section 5e…Measuring equipment 6... Repair and Diagnosis Department 6a...Interface for repair and diagnosis unit (I / F) 6b... Repair and diagnosis unit processor (CPU) 6c...Memory of repair diagnosis section 6d...Memory section of repair diagnosis section 7...Repair Department 7a... Repair department interface (I / F) 7b... Repair department processor (CPU) 7c...Memory of repaired part 7d…Memory section of repair section 7e...Repair equipment 10...Recovery process 20...Dismantling and cleaning process 30...Parts measurement process 31...Imaging process 32...Image analysis process 33...Preparation process for coating removal 34…Film removal process 35...Measurement process 40... Repair feasibility diagnosis process 50...Parts repair process 51...Cleaning necessity diagnosis process 52...Cleaning process 53... Repair method diagnosis process 54...Repair process 60...Parts inspection process 70...Assembly process 80...Product inspection process 90...Shipping process 100…Network 101...Image 150…Parts 150a…normal area 150b…Coating area 150c…damage area 150d…Measurement area 160...Interference fringes 180...Reference 200…Cancelled process

Claims

1. 1. A measurement system for measuring a shape of a part having at least a portion of a surface covered with a coating, comprising: an imaging unit that captures an image of the component and acquires image data; an image analysis unit that determines the state of the coating based on the image data, diagnoses whether or not removal of the coating is necessary, and determines a measurement region; a measurement unit that measures the shape of the part in the measurement area, When it is diagnosed that removal of the coating is necessary, the measurement region includes at least a part of the region from which the coating has been removed; The image analysis unit estimates the degree of change in the shape of the part inside the coating from the image data as the degree of damage using a prediction model previously constructed based on the correlation between coating data and changes in the shape of the part inside the coating, and diagnoses areas where the degree of damage is estimated to be large as areas where the coating is required to be removed, and the measurement area includes areas where the coating is removed from the areas where the degree of damage is estimated to be large. A measurement system characterized by:

2. In claim 1, the image analysis unit determines the region where the degree of damage is estimated to be large as a first measurement region, and the region where the degree of damage is estimated to be small as a second measurement region; The measurement unit measures the shape of the part in each of the first measurement area and the second measurement area, thereby measuring the relative shape of the part in the first measurement area. A measurement system characterized by:

3. In claim 1, The method further includes a surface processing unit that removes the coating when the image analysis unit determines that removal of the coating is necessary. A measurement system characterized by:

4. In claim 1, The method further includes a surface processing unit that, when the image analysis unit diagnoses that removal of the coating is necessary, removes the coating according to processing conditions determined based on at least the state of the coating. A measurement system characterized by:

5. In claim 1, The image analysis unit diagnoses that removal of the coating is not necessary for at least a part of the component, and diagnoses that removal of the coating is necessary for an area where the degree of damage is estimated to be large. A measurement system characterized by:

6. In claim 1, a display unit that displays an image captured by the imaging unit; the image analysis unit estimates a region where a change in shape of the component has occurred from the image data as a damaged region; The display unit displays the damaged area of ​​the component on the image. A measurement system characterized by:

7. In claim 1, the imaging unit images the component under a plurality of different imaging conditions to obtain a plurality of image data; The measurement system is characterized in that the image analysis unit diagnoses whether or not removal of the coating is necessary based on the plurality of image data.

8. In claim 4, the surface processing unit is a laser removal device, The processing conditions are laser irradiation conditions determined based on the material information of the part. A measurement system characterized by:

9. In claim 1, The measurement system is characterized in that the imaging unit assists in capturing an image that makes it easy to determine the state of the coating.

10. In claim 6, The display unit displays a display on the image to assist in removing the coating. A measurement system characterized by:

11. The measurement system according to claim 1 ; a repair diagnosis unit that determines whether the part can be repaired and how to repair it based on the measurement results of the part shape by the measurement system and pre-recorded design data of the part; a repair unit that repairs the part by the repair method when the repair diagnosis unit determines that the part can be repaired. A parts remanufacturing system.

12. The measurement system according to claim 5 ; a repair diagnosis unit that determines whether the part can be repaired and how to repair it based on the measurement results of the part's shape by the measurement system and pre-recorded design data of the part.

13. In claim 12, The method further includes a surface processing unit that, when the image analysis unit diagnoses that removal of the coating is necessary, removes the coating in accordance with processing conditions determined based on at least the state of the coating, When the repair diagnosis unit determines that the part can be repaired, the surface processing unit removes the coating that the image analysis unit determines does not need to be removed. A parts remanufacturing system.

14. In claim 13, When the repair diagnosis unit determines that the part can be repaired, the surface processing unit removes the coating that the image analysis unit determines does not require removal in accordance with reprocessing conditions determined based on at least the state of the coating. A parts remanufacturing system.

15. In claim 14, the surface processing unit is a laser removal device, the processing conditions are laser irradiation conditions determined based on material information of the part, The reprocessing conditions are the same laser irradiation conditions as the processing conditions. A parts remanufacturing system.

16. A measurement method for measuring a shape of a component having at least a portion of a surface covered with a coating, comprising: an imaging step of imaging the component to acquire image data; an image analysis step of determining the state of the coating based on the image data, diagnosing whether or not removal of the coating is necessary, and determining a measurement area; a coating removal step of removing the coating diagnosed as needing removal; a measurement step of measuring a shape of the component in the measurement area after removing the coating, When it is diagnosed that removal of the coating is necessary, the measurement region includes at least a part of the region from which the coating has been removed; The image analysis step uses a prediction model previously constructed based on the correlation between coating data and changes in the shape of the part inside the coating to estimate the degree of change in the shape of the part inside the coating from the image data as the degree of damage, and diagnoses a region where the degree of damage is estimated to be large as requiring removal of the coating, and the measurement region includes a region where the coating is removed from the region where the degree of damage is estimated to be large. A measuring method characterized by:

17. 17. The measurement method according to claim 16, further comprising the steps of: a repair diagnosis step of determining whether the part can be repaired and how to repair it based on the measurement results of the part shape and pre-recorded design data of the part; and a repairing step of repairing the part by the repairing method when it is determined that the part can be repaired. A component regeneration method.

Images