WO2023112497A1 - 情報処理装置、情報処理方法、プログラム及び記録媒体 - Google Patents

情報処理装置、情報処理方法、プログラム及び記録媒体 Download PDF

Info

Publication number
WO2023112497A1
WO2023112497A1 PCT/JP2022/039845 JP2022039845W WO2023112497A1 WO 2023112497 A1 WO2023112497 A1 WO 2023112497A1 JP 2022039845 W JP2022039845 W JP 2022039845W WO 2023112497 A1 WO2023112497 A1 WO 2023112497A1
Authority
WO
WIPO (PCT)
Prior art keywords
detection target
information processing
priority
feature amount
area
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/JP2022/039845
Other languages
English (en)
French (fr)
Japanese (ja)
Inventor
遼 池田
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Fujifilm Corp
Original Assignee
Fujifilm Corp
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Fujifilm Corp filed Critical Fujifilm Corp
Priority to EP22907031.3A priority Critical patent/EP4450959A4/en
Priority to CN202280081559.4A priority patent/CN118382799A/zh
Priority to JP2023567583A priority patent/JPWO2023112497A1/ja
Publication of WO2023112497A1 publication Critical patent/WO2023112497A1/ja
Priority to US18/738,416 priority patent/US20240331339A1/en
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N23/00—Investigating or analysing materials by the use of wave or particle radiation, e.g. X-rays or neutrons, not covered by groups G01N3/00 – G01N17/00, G01N21/00 or G01N22/00
    • G01N23/02—Investigating or analysing materials by the use of wave or particle radiation, e.g. X-rays or neutrons, not covered by groups G01N3/00 – G01N17/00, G01N21/00 or G01N22/00 by transmitting the radiation through the material
    • G01N23/06—Investigating or analysing materials by the use of wave or particle radiation, e.g. X-rays or neutrons, not covered by groups G01N3/00 – G01N17/00, G01N21/00 or G01N22/00 by transmitting the radiation through the material and measuring the absorption
    • G01N23/18—Investigating the presence of flaws defects or foreign matter
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N23/00—Investigating or analysing materials by the use of wave or particle radiation, e.g. X-rays or neutrons, not covered by groups G01N3/00 – G01N17/00, G01N21/00 or G01N22/00
    • G01N23/02—Investigating or analysing materials by the use of wave or particle radiation, e.g. X-rays or neutrons, not covered by groups G01N3/00 – G01N17/00, G01N21/00 or G01N22/00 by transmitting the radiation through the material
    • G01N23/04—Investigating or analysing materials by the use of wave or particle radiation, e.g. X-rays or neutrons, not covered by groups G01N3/00 – G01N17/00, G01N21/00 or G01N22/00 by transmitting the radiation through the material and forming images of the material
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/0002—Inspection of images, e.g. flaw detection
    • G06T7/0004—Industrial image inspection
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/50—Depth or shape recovery
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/20—Image preprocessing
    • G06V10/25—Determination of region of interest [ROI] or a volume of interest [VOI]
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/20—Image preprocessing
    • G06V10/255—Detecting or recognising potential candidate objects based on visual cues, e.g. shapes
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/40—Extraction of image or video features
    • G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
    • G06V10/764—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10116—X-ray image
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30108—Industrial image inspection
    • G06T2207/30136—Metal
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30168—Image quality inspection
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V2201/00—Indexing scheme relating to image or video recognition or understanding
    • G06V2201/06—Recognition of objects for industrial automation
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V2201/00—Indexing scheme relating to image or video recognition or understanding
    • G06V2201/07—Target detection

Definitions

  • the present disclosure relates to an information processing device, an information processing method, and a program, and more particularly, to information processing technology applied to processing for displaying the result of image processing performed on an image transmitted through radiation.
  • each of a plurality of flaws occurring on the surface of a steel plate detected from image data obtained by imaging the surface of the steel plate is detected as harmful based on the feature amount obtained from the brightness, shape, and size of the flaw in the image data.
  • a flaw inspection device has been proposed that classifies the flaws into either flaws or harmless flaws, and classifies the classified harmful flaws by ranks representing the severity of flaws according to at least the type and size of the flaw (Patent Document 1). .
  • One embodiment according to the technology of the present disclosure is an information processing device, an information processing method, and an information processing method that can extract a detection target area from a radiation image of an inspection object and optimize the workflow for the detection result of the detection target area. Offer a program.
  • An information processing apparatus comprising a processor according to the first aspect, wherein the processor acquires a radiation image of an inspection object transmitted through radiation, extracts a detection target area from the transmission image, and acquires a feature amount for the extracted detection target area. Then, the priority of the detection target area is determined based on the feature amount.
  • the amount of radiation irradiated to the inspection object is radiation in the X-ray wavelength region within a range of tube voltage of 60 kV or more and 450 kV or less and a range of tube current of 1 mA or more and 10 mA or less. corresponds to the amount of radiation when irradiated for 1 second or more and 5 minutes or less.
  • the transparent image satisfies the required image quality level.
  • the inspection object is metal
  • the feature quantity includes the depth on the transmission image in the extracted detection target area or the thickness of the inspection target.
  • the depth is acquired from the pixel value of the transmission image, the imaging conditions when acquiring the transmission image, and the three-dimensional model data of the inspection object.
  • the feature quantity is a pixel value.
  • the feature amount includes pixel values in the peripheral area of the extracted detection target area.
  • the feature amount includes any one of the major axis, area, and shape of the detection target area.
  • the feature amount is a numerical value calculated by combining any one of the major axis, area, shape, and pixel value of the detection target area, and pixel values in the surrounding area of the detection target area.
  • the feature amount is a numerical value calculated from distribution information of two or more detection target areas.
  • the processor controls the display content of the detection target area displayed on the display screen based on the determined priority.
  • the processor acquires the quality standard required for the inspection object, determines whether the inspection object satisfies the quality standard, and if the quality standard is satisfied, detects based on the feature amount. Determine the priority of the region of interest.
  • An information processing method executed by an information processing apparatus having a processor includes the steps of obtaining a transmission image of an inspection object by radiation, extracting a detection target area from the transmission image, and extracting the detection target area. and determining the priority of the detection target area based on the feature amount.
  • a program for executing an information processing method executed by an information processing apparatus having a processor includes the steps of acquiring a radiographic image of an inspection object, extracting a detection target region from the radiographic image, extracting a step of acquiring a feature amount for the detected detection target area; and a step of determining the priority of the detection target area based on the feature amount.
  • the present invention it is possible to extract the detection target area from the transmitted radiation image of the inspection object and optimize the workflow for the detection result of the detection target area.
  • FIG. 1 is a block diagram showing an information processing device according to an embodiment.
  • FIG. 2 is a block diagram showing an example of object image data.
  • FIG. 3 is a block diagram showing an example of product data.
  • FIG. 4 is a block diagram showing processing functions realized by the control unit.
  • FIG. 5 is a block diagram showing a configuration example of an imaging system.
  • FIG. 6 is a flow chart showing an information processing method using an information processing device.
  • FIG. 7 is a diagram illustrating an example of processing related to priority determination according to the embodiment.
  • FIG. 8 is a diagram illustrating another example of processing related to priority determination according to the embodiment.
  • FIG. 9 is a diagram illustrating another example of processing related to priority determination according to the embodiment.
  • FIG. 10 is a diagram illustrating another example of processing related to priority determination according to the embodiment.
  • FIG. 11 is a diagram showing an example of the display content of the detection target area obtained through processing by the information processing device.
  • FIG. 1 is a functional block diagram schematically showing the functional configuration of the information processing device 10 according to the embodiment of the present disclosure.
  • the information processing apparatus 10 is a device that executes a process of extracting a defect area, which is a detection target area, from a radiographic transmission image of an industrial product that is an inspection target, and determines the priority of the detection target area. Further, the information processing device 10 is a device that causes the display device to display according to the priority of the detection target area.
  • the information processing apparatus 10 mainly includes a control unit 12, a storage unit 14, an input/output interface (hereinafter referred to as I/F (interface)) 16, and a RAM (Random Access Memory) 22 and ROM (Read Only Memory) 24 .
  • An operation unit 18 and a display device 20 are connected to the information processing device 10 .
  • the operation unit 18 gives necessary commands to the information processing apparatus 10 based on the operator's operation.
  • the display device 20 displays various information under the control of the control section 12 .
  • the control section 12 includes a CPU (Central Processing Unit) that controls the operation of each section of the information processing device 10 .
  • the control unit 12 receives an operation input from an operator via the operation unit 18 and transmits a control signal corresponding to the operation input to each unit of the information processing apparatus 10 to control the operation of each unit.
  • the control unit 12 reads various programs stored in the storage unit 14, the ROM 24, or the like, develops them in the RAM 22, executes processing, and centrally controls each unit.
  • the control unit 12 controls the entire information processing device 10 .
  • the control unit 12 receives an operation input from an operator via the operation unit 18 and transmits a control signal corresponding to the operation input to each unit of the information processing apparatus 10 to control the operation of each unit.
  • the storage unit 14 is means for storing data including the operating system and control programs used by the control unit 12 .
  • a device including a magnetic disk such as HDD (Hard Disk Drive), a device including flash memory such as eMMC (embedded Multi Media Card), SSD (Solid State Drive), etc. can be used.
  • the storage unit 14 may store object photographed data D100 and product data D200, which will be described later.
  • the I/F 16 inputs and outputs various data (information) to and from the information processing device 10 .
  • the I/F 16 communicates with external devices, for example, via the network NW.
  • external devices for example, via the network NW.
  • wired communication for example, LAN (Local Area Network), WAN (Wide Area Network), Internet connection, etc.
  • wireless communication for example, LAN, WAN , Internet connection, etc.
  • the operation unit 18 is an input device that receives operation inputs from an operator, and includes a keyboard for character input, a pointer displayed on the display device 20, and a pointing device (mouse, trackball, etc.) for operating icons and the like. include.
  • a touch panel may be provided on the surface of the display device 20 instead of or in addition to the means enumerated above.
  • the display device 20 is, for example, a device such as a liquid crystal display, and can display various information such as a transmission image of an inspection target.
  • the information processing apparatus 10 can acquire the object photographed data D100 including the data of the transmission image of the object OBJ, which is the inspection object, photographed by the photographing system 100 via the I/F 16 (see FIG. 5).
  • the target object OBJ is, for example, an industrial product and is a metal.
  • the method of inputting the object photographed data D100 from the photographing system 100 to the information processing apparatus 10 is not limited to communication via the network NW listed above.
  • the information processing device 10 and the imaging system 100 may be connected by a USB (Universal Serial Bus) cable, Bluetooth (registered trademark), infrared communication, or the like, and the object photography data D100 may be attached to, detached from, and attached to the information processing device 10.
  • the photographed image data may be stored in a readable memory card and input to the information processing apparatus 10 via this memory card.
  • the information processing device 10 can communicate with a product database (product DB (database)) 200 via the network NW.
  • product DB 200 stores product data D200 for each target industrial product.
  • the control unit 12 retrieves and reads object specifying information for specifying an object from the photographed image data of the object OBJ acquired from the photographing system 100, and outputs product data D200 corresponding to the read object specifying information to the product. It becomes possible to obtain from the DB 200 . By using this product data D200, it becomes possible to detect defects, which are inspection objects, according to the type or characteristics of the object OBJ.
  • the product DB 200 may be installed on the network NW as in the embodiment so that the manufacturer name of the product data D200 can be updated, or may be installed in the information processing apparatus 10 .
  • FIG. 2 is a block diagram showing an example of object photographed data.
  • the object photographing data D100 includes object specifying information, a transmission image, and photographing parameters.
  • the object specifying information is information for specifying the object OBJ, and includes, for example, information indicating the product name, product number, ID (identification) information, manufacturer name, and technical classification of the object OBJ.
  • a transmission image is a transmission image of an object OBJ to be inspected, and includes pixel value or resolution information.
  • the shooting parameters are stored in association with each transmission image of the object OBJ, and include the shooting date and time of each transmission image, the location of the shooting target, the distance between the object OBJ and the shooting device at the time of shooting, and the angle with respect to the shooting device. Contains information about exposure time. Further, the imaging parameters include radiation irradiation start time, irradiation duration, irradiation angle, irradiation intensity, and the like, and conditions related to irradiation intensity include tube current and tube voltage.
  • FIG. 3 is a block diagram showing an example of product data.
  • product data D200 includes product identification information, product attribute information, and inspection area designation information.
  • the product data D200 may be stored in the storage unit 14 in association with the object image data D100 via the object specifying information and the product specifying information, or may be acquired from the product DB 200 each time a defect inspection is performed. You may do so.
  • Product identification information is information for identifying a product, and includes, for example, information indicating the product name, product number, manufacturer name, and technical classification.
  • Product attribute information includes, for example, information indicating the material, dimensions, and usage of each part of the product.
  • Product dimensions can be obtained, for example, from two-dimensional drawings and/or three-dimensional model data.
  • the information indicating the use of the product includes, for example, information on the name, type, processing state, and mounting method (for example, joints, welds, screwing, fitting, and soldering) of equipment to which the product is attached.
  • product attribute information includes defect occurrence information related to quality standards and acceptance criteria for inspections.
  • the defect occurrence information includes, for example, past inspection history and time series, past inspection date and time, material of the object OBJ, types of defects that occurred in the past (for example, foreign matter, cracks, scratches, air bubbles, welding gas defects, wear, rust, etc.), position, shape, size, depth, location of occurrence (coordinates of location, thickness of material, processing state (e.g. joints, welds, etc.)), information on frequency of defect occurrence, defects at least one of the captured images of
  • acceptance criteria may include standards such as ASTM standards.
  • the inspection area specification information is information indicating the inspection area specified by the manufacturer of each product (for example, information including the position of the inspection area, whether or not defects have occurred in the past, and defect occurrence frequency information such as defect occurrence frequency). based on information).
  • the inspection area designation information is created, for example, by specifying locations where defects are likely to occur statistically and structurally, based on information obtained when the manufacturer or the like repaired the product in the past.
  • the defect occurrence information includes, for example, at least one of the past inspection date and time, the material of the object OBJ, the type, shape, size, depth, occurrence location, and defect capture image of the defect that occurred in the past.
  • FIG. 4 is a block diagram showing an example of the control unit 12.
  • the control unit 12 includes a transparent image acquisition unit 121 , a detection target extraction unit 122 , a feature amount acquisition unit 123 , a priority determination unit 124 and a display control unit 125 .
  • the transmission image acquisition unit 121 acquires a radiation transmission image of the object OBJ.
  • the transmission image acquisition unit 121 can acquire the transmission image included in the object photographed data D100 stored in the storage unit 14 . Further, the transmission image acquisition unit 121 may acquire the transmission image included in the object image data D100 from an external device such as the imaging system 100 or an image management server (not shown) via the I/F 16 .
  • the transparent image acquisition unit 121 can acquire a transparent image from removable media such as a memory card.
  • the detection target extraction unit 122 extracts a detection target region from the transparent image. For example, the detection target extraction unit 122 performs region extraction processing on the transparent image using a segmentation model, and extracts a defect region as a detection target region from the transparent image. This identifies the position and shape of the defect.
  • a segmentation model is a learning model trained using machine learning to perform the task of image segmentation.
  • the transmissive image is divided into regions by classifying whether it is a region of or not.
  • the segmentation model may be a model that performs two-class classification detection for each pixel (pixel), whether it is a defect area or not, or a model that performs multi-class classification detection, such as what kind of defect it is for each pixel. It may be a model that performs classification detection.
  • the segmentation model is constructed using, for example, a convolutional neural network (CNN) with convolutional layers.
  • CNN convolutional neural network
  • FCN Fully Convolution Network
  • FCN Fully Convolution Network
  • the feature quantity acquisition unit 123 acquires the feature quantity for the extracted detection target area.
  • the feature values include elements such as the length, area, shape, position, pixel value and depth of the detection regions, the distance between the detection regions, the number of detection regions within a certain test field, and the thickness of the detection object. .
  • the distance between the detection areas can be, for example, the distance between the centers of gravity (or centers) of the two areas, or the shortest distance among line segments connecting points on the boundary line of each area.
  • the feature quantity acquisition unit 123 acquires at least one of these elements.
  • the feature amount acquisition unit 123 can acquire these elements from the transmission image acquired by the transmission image acquisition unit 121 .
  • the feature amount acquisition unit 123 may acquire the object specifying information and the shooting parameters included in the object shooting data D100 and acquire them as feature amounts. In addition to the feature amount for the extracted detection target area, the feature amount acquisition unit 123 may acquire the feature amount for the surrounding area of the extracted detection target area. Also, the feature amount of the peripheral area may be a pixel value.
  • the priority determination unit 124 determines the priority of the detection target area based on the feature amount acquired by the feature amount acquisition unit 123 .
  • the priority determination unit 124 determines, for example, the priority of diagnosing a defect that is a detection target area, that is, the importance of the necessity of visual recognition by an operator. A higher priority is given to an area with a higher importance of necessity of visual recognition.
  • the priority determination unit 124 further acquires any of the product identification information, product attribute information, and inspection area designation information included in the product data D200, and refers to these information to determine the priority of the detection target area. You may
  • the priority determination unit 124 determines the priority of the detection target area, so that the display or non-display of the output result for the defect area can be filtered based on the priority result. can be sorted based on the results of As a result, by optimizing the inspection workflow, it is possible to efficiently execute inspections, such as determining whether or not shipment is possible.
  • the priority determination unit 124 may determine priority based on the length of the detection area in the longitudinal direction based on the feature amount.
  • the priority may be determined based on the distance between the first detection area and the second detection area, which is the closest neighbor of the first detection area, from the feature amount.
  • the “distance” is the shortest of the distance between the centroids (or centers) of the two regions or the line segment connecting the points on the boundary line of each region.
  • the priority may be determined by focusing on the shape of the detection target area or the regularity of distribution from the feature amount. For example, a detection area having characteristics such as a so-called elongated shape with a large ratio of the long axis to the short axis of the defect area, a shape that is far from a circle, such as a concave shape, or a shape that includes a pointed shape. , stress is likely to be applied to the relevant portion in the inspection object, and the priority may be determined to be higher.
  • the detection target areas consisting of two or more, the detection target areas for which the distribution information is obtained that they occur at equal intervals or occur linearly, cracks and cracks are likely to occur.
  • it may be determined to have a higher priority than non-linear detection target regions.
  • the priority of the detection target areas that make up the visual field area may be increased according to the number of detection areas (so-called density) included in a certain test visual field based on the feature amount.
  • density the number of detection areas included in a certain test visual field based on the feature amount.
  • the range of the test field of view which should be the reference for inspection, is defined as an area with fixed vertical and horizontal lengths, or an area with a specific length along the weld line, out of the entire target image.
  • the priority may be determined from the feature amount only in the range including the area where the detection target exists.
  • part of the test field of view includes the background that is reflected in the image, that is, a subject other than the subject to be inspected
  • priority is given to the feature amount only in the area that includes the subject in the inspection field of view. degree can be determined.
  • the inspection target is a meandering weld line or a cast component with a curved outer shape
  • the inspection field is deformed or divided along the contour of the component, and the priority is determined based on the feature value only in that range. You can judge.
  • the priority may be determined using the pixel value of the detection target area as a feature amount.
  • the density information of the defect if it is a foreign substance, its characteristics
  • the pixel value is higher than the surroundings, it can be determined that there is a high-density foreign object, and if it is lower than the surroundings, it can be determined that there is a possibility that it is a cavity or a low-density foreign object.
  • type identification may be performed by automatic determination (AI), but the results obtained by determining only the presence or absence of defects may be classified by type in post-processing based on pixel values.
  • the pixel value of the extracted detection target region and the pixel value of the peripheral region of the extracted detection target region can be included, and the pixel value of the detection target region and the pixel value of the peripheral region can be
  • the priority may be determined from the contrast of . A determination can be made to give higher priority to a detection target area having a stronger contrast with its surroundings.
  • a numerical value and a feature value calculated by combining any of the major axis, area, shape, pixel value of the detection target region, and pixel values in the peripheral region of the detection target region, and detection based on this feature value may be determined.
  • the priority of the detection target area may be determined to be high.
  • a distance standard (hereinafter referred to as a threshold) determined from quality standards such as ASTM standards, characteristics of the inspection object, past inspection history, inspection acceptance criteria, etc. information is stored, the corresponding threshold is acquired based on the individual information of the detection target area, or the threshold is acquired by setting it individually before the inspection, and the acquired threshold and the acquired By comparing with the feature amount obtained, it may be determined to increase the priority only when the former value exceeds the former value.
  • the display control unit 125 controls the display content of the detection target area displayed on the display screen based on the priority determined by the priority determination unit 124, and causes the display device 20 to display it.
  • the display control unit 125 Based on the determined priority, the display control unit 125 gives the display of the detection target area a step-by-step display priority, that is, a difference in the degree of conspicuousness.
  • the display content may be controlled and displayed on the display device 20 by encircling with an emphasized frame, blinking, or the like.
  • the display control unit 125 displays a frame that is emphasized by color and thickness so that the area that is determined to have a particularly high priority should be noted, and a message indicating that the area is of high priority. may be additionally displayed in the form of balloons and/or text to control the display content and cause the display device 20 to display.
  • the display control unit 125 may have a function of switching between display and non-display on the display device 20.
  • the display control unit 125 can, for example, display only the priority detection areas within a certain range of the inspection object, and not display the other detection areas.
  • the display control unit 125 may combine information other than the priority, and selectively display, for example, only unconfirmed detection areas with a certain level of priority or more on the display device 20 .
  • the display control unit 125 may display the detection areas on the display device 20 in order from highest priority to lowest priority while performing narrowing down according to the priority.
  • the detection target area with the highest priority is highlighted first.
  • the periphery of the detection target area with the highest priority is first displayed in an enlarged manner.
  • the detection target area with the second priority may be displayed, and then the areas may be displayed in order according to the priority.
  • the detection target areas with the lowest priority and the detection target areas with the highest priority may be displayed in order.
  • the priority of the detection target area can be communicated to the operator.
  • FIG. 5 is a block diagram schematically showing a configuration example of the imaging system 100.
  • the imaging system 100 is for imaging an object OBJ placed in an imaging room 114, and includes an imaging control unit 102, an imaging operation unit 104, an image storage unit 106, a camera 108, and a radiation source 112. .
  • the imaging control unit 102 includes a CPU that controls the operation of each unit of the imaging system 100 .
  • the imaging control unit 102 receives an operation input from the operator via the imaging operation unit 104, transmits a control signal corresponding to the operation input to each unit of the imaging system 100, and controls the operation of each unit.
  • the imaging operation unit 104 includes an input device that receives operation inputs from the operator. Via the imaging operation unit 104, the operator inputs information about the object OBJ, inputs imaging condition instructions and imaging execution instructions to the camera 108, inputs radiation irradiation conditions instructions to the radiation source 112, and acquires by imaging. It is possible to input an instruction to store the obtained transparent image in the image storage unit 106 .
  • the shooting parameters include, for example, shooting conditions such as exposure time and focal length, shooting angle, shooting location, and the like.
  • Radiation irradiation conditions include irradiation start time, irradiation duration, irradiation angle, irradiation intensity, and the like.
  • conditions related to irradiation intensity include tube current and tube voltage.
  • the imaging system 100 when capturing a transmitted image using radiation, for example, radiation in the X-ray wavelength region (about 1 pm to 10 nm) is emitted under conditions of a tube voltage of 60 kV to 450 kV and a tube current of about 1 mA to 10 mA.
  • the object OBJ is irradiated for 1 second or more and several minutes or less, for example, 5 minutes or less.
  • the image storage unit 106 stores a transparent image of the object OBJ photographed by the camera 108 .
  • Information for specifying the target object OBJ is stored in the image storage unit 106 in association with the captured image data.
  • the camera 108 for example, is a method (DDA: Digital Detector Array, digital detector) or IP (Imaging Plate) to store signals and digitize them with an external reader (CR: computed radiography).
  • the camera 108 and the radiation source 112 are arranged inside the imaging room 114 .
  • the radiation source 112 is, for example, an X-ray source, and X-ray protection is applied to partition walls and doorways between the imaging room 114 and the outside with X-ray protection materials (eg, lead, concrete, etc.).
  • the radiation source 112 irradiates the object OBJ placed in the imaging room 114 with radiation according to the instruction from the imaging control unit 102 .
  • the camera 108 receives the radiation emitted from the radiation source 112 to the object OBJ and transmitted through the object OBJ in accordance with the instruction to execute imaging from the imaging control unit 102, and images the object OBJ.
  • the object OBJ is held in the imaging room 114 by a holding member (for example, a manipulator, a mounting table, or a movable mounting table) (not shown).
  • the angle is adjustable. The operator can control the relative positions of the object OBJ, the camera 108, and the radiation source 112 via the imaging control unit 102, and can photograph a desired portion of the object OBJ.
  • the radiation source 112 finishes irradiating the object OBJ with radiation in synchronization with the end of execution of imaging by the camera 108 .
  • one camera 108 and one radiation source 112 are provided, but the number of cameras and radiation sources is not limited to this. For example, there may be multiple cameras and radiation sources, or there may be one each.
  • the imaging control unit 102, the imaging operation unit 104, and the image storage unit 106 can be implemented using a combination of computer hardware and software.
  • the required image quality level (image quality) of the transmitted image of radiation is, for example, preferably Class B or higher as defined by JISZ3104 (radiographic test method for steel welded joints).
  • JISZ3104 radiographic test method for steel welded joints.
  • the sensitivity of detecting "flaws" can be increased by class B.
  • the image quality level is, for example, a gradation meter, a transmittance meter, or an indicator called IQI (Image Quality Indicator) is photographed at the same time as the subject, and if it is determined that the image quality is above a certain level specified by JIS, it is grade B. If it does not meet the requirements, it will be classified as Class A. Furthermore, in Table 4 of JISZ3104, standards for observation of transmission photographs (transmission images) are set.
  • the information processing method includes, for example, a step of acquiring a transparent image (step S1), a step of extracting a detection target area (step S2), and a step of acquiring a feature amount (step S3). , a step of determining the priority (step S4), and a step of controlling the display content of the detection target area (step S5).
  • step S1 a radiographic transmission image of the object to be inspected is obtained.
  • a transparent image acquisition unit 121 of the control unit 12 acquires a transparent image from an external device such as the storage unit 14, the imaging system 100, or an image management server (not shown) through the I/F 16.
  • FIG. The transmission image may be included in the object photographed data D100.
  • step S2 a detection target area is extracted from the transparent image.
  • a detection target extraction unit 122 of the control unit 12 extracts a detection target region from the transparent image.
  • the feature amount for the extracted detection target area is obtained.
  • the feature quantity acquisition unit 123 of the control unit 12 acquires the feature quantity for the extracted detection target region.
  • the feature values include the long diameter, area, shape, position, pixel value and depth of the detection regions, the distance between the detection regions, the number of detection regions within a certain test field of view, and the thickness of the detection object. elements can be mentioned.
  • the distance between the detection areas can be, for example, the distance between the centers of gravity (or centers) of the two areas, or the shortest distance among line segments connecting points on the boundary line of each area.
  • step S4 the priority of the detection target area is determined based on the feature amount acquired in step S3.
  • the priority determination unit 124 of the control unit 12 determines the priority of the detection target region based on the feature amount acquired by the feature amount acquisition unit 123 . As described above, the priority determination unit 124 determines the priority of diagnosing a defect, which is a detection target area, and the importance of the necessity of visual recognition by the operator.
  • FIG. 7 is a diagram illustrating an example of processing related to priority determination according to the embodiment.
  • the detection target extraction unit 122 uses the segmentation model to perform region extraction processing from the transparent image, and extracts the detection target region. As shown in 700A, the detection target extraction unit 122 extracts two elliptical defect regions 302 and 304, which are detection target regions, on the segmentation image 300.
  • the feature amount acquisition unit 123 can acquire, as feature amounts, the major axis, area, shape, or distance between the detection areas of the defect areas 302 and 304, which are detection target areas, for example.
  • the two defect regions 302, 304 are arranged continuously in the longitudinal direction.
  • the priority determination unit 124 determines a high priority based on the acquired feature amount.
  • FIG. 8 is a diagram showing another example of processing related to priority determination according to the embodiment.
  • the detection target extracting unit 122 executes region extraction processing using the segmentation model from the transparent image to extract the detection target region.
  • a plurality of defect regions 308 are extracted from the segmentation image 300 by the detection target extraction unit 122 as indicated by 800A.
  • the defect region 308 is, for example, a chain porosity in which welding gas defects occur at regular intervals.
  • the feature quantity acquisition unit 123 acquires, for example, the areas, shapes, distances and positions between the detection regions of the plurality of defect regions 308, which are detection target regions, as feature quantities.
  • the priority determination unit 124 determines a high priority based on the acquired feature amount.
  • FIG. 9 is a diagram showing another example of processing related to priority determination according to the embodiment.
  • the detection target extracting unit 122 extracts a detection target region from the transparent image by executing region extraction processing using the segmentation model.
  • two L-shaped areas indicate subjects 312 and 314 to be inspected, and dotted circles indicate a plurality of defect areas 316 .
  • a frame 318 indicates the range defined as the inspection field of view.
  • the defective areas 316 that are included in the frame 318 but include the subject 314 are not counted as the feature amount related to the subject 312, and the priority determination unit 124 Priority is determined with respect to the object 312 as two defect areas 316 .
  • FIG. 10 is a diagram showing another example of processing related to priority determination according to the embodiment.
  • FIG. 10 shows a graph showing a one-dimensional cross-sectional view of the increase and decrease in the pixel value, which is the feature amount in the detection target area.
  • the vertical axis indicates the pixel value
  • the horizontal axis indicates the distance (mm) of the measured cross section (straight line on the transmission image).
  • the range from 20 mm to 120 mm is the range AR1 in which the part to be inspected exists.
  • a range AR2 and a range AR3 positioned on both sides of the range AR1 are ranges not subject to inspection.
  • the part to be inspected consists of ranges AR4 and AR6 (two locations) with pixel values of approximately 30,000 and range AR5 with approximately 23,000 pixel values.
  • the thicknesses of the areas AR4 and AR6 are greater than the thickness of the area AR5.
  • a defect area A is extracted at a position of 40 mm on the horizontal axis, and a defect area B is extracted at a position of 50 mm as a detection area, and pixel values are obtained for each. If it can be determined that the defect area A and the defect area B exist in the range AR4, the priority determination unit 124 compares only the pixel values of the defect area A and the defect area B based on this information to determine the priority. can.
  • a defect area C is extracted as a detection area at a position of 80 mm on the horizontal axis, and pixel values are obtained. If it can be determined from the characteristics of the background that the defect area C exists in the thin range AR5, which is thinner than the range AR4, as a component, the contrast in the defect areas A and B is the same as that in the defect area C. Even if there is, the thinner portion may adversely affect the quality of the part. Therefore, it is not appropriate to simply compare the defect area A and the defect area C, and the defect area B and the defect area C by contrast. Based on this information, the priority determination unit 124 may determine the priority of the detection target region for each thickness (layer) of a specific part.
  • the priority determination unit 124 determines the priority of the defect area A and the defect area B by limiting the pixel value of 30000 (range AR4), and determines the priority of the defect area C by limiting the pixel value of 23000 (range AR5). priority can be determined.
  • the priority determination unit 124 determines whether the defect area A and the defect area C, and the defect area B and the defect area C are superior or inferior. may be added and converted to priority on the same scale to determine the priority of the defect area A, the defect area B, and the defect area C.
  • the determination of the ranges AR4, AR5, and AR6 may be divided into a plurality of layers in consideration of the continuity and steps of pixel values, as described above. Moreover, when the product data D200 includes thickness (or depth) information, this information may be used to separate the layers into a plurality of layers.
  • step S5 the display content of the detection target area displayed on the display screen is controlled based on the determined priority.
  • the display control unit 125 of the control unit 12 controls the display content of the detection target area displayed on the display screen based on the priority determined by the priority determination unit 124, and causes the display device 20 to display the content.
  • FIG. 11 is a diagram showing an example of the display content of the detection target area obtained through processing by the information processing device.
  • the defect that is the detection target area, the priority of the determined defect, and the color of the frame are displayed in correspondence with each other on a density scale.
  • the display device 20 displays the metal part 50 to be inspected.
  • Three defective areas 51, 52 and 53 are extracted by the information processing apparatus 10, and priority is determined for the defective areas 51, 52 and 53.
  • FIG. A defect area 51 , a defect area 52 and a defect area 53 are displayed on the metal part 50 .
  • the information processing apparatus 10 determines the priority in the order of defective area 53 ⁇ defective area 52 ⁇ defective area 51.
  • the display device 20 displays a density scale 26 indicating priority levels other than the metal component 50, the defect area 51, the defect area 52, and the defect area 53, and a needle 28 for displaying and selecting the priority level. It is On the density scale 26, a color closer to black indicates a lower priority, and a color closer to white indicates a higher priority. When the needle 28 points to the density scale 26 , the display device 20 displays a display that allows recognition of defects having a higher priority than the color (level) of the pointed density scale 26 .
  • the density scale 26 may be displayed not only in black and white, but also in color.
  • the needle 28 is positioned at the lower end of the density scale 26.
  • Defect area 51 , defect area 52 and defect area 53 which are the priority of the color (level or higher) indicated by needle 28 , are highlighted by frames 61 , 62 and 63 .
  • the frames 61 , 62 and 63 are displayed in colors corresponding to the colors (levels) of the density scale 26 .
  • a frame 61 is the lightest color and indicates a high priority
  • a frame 63 is the darkest color and indicates a high priority
  • a frame 62 is a medium color and indicates a medium priority.
  • the operator can know that the priority corresponding to the density scale 26 is determined, and also recognizes the colors of the frames 61, 62 and 63. Thus, the priority levels of the defective area 51, the defective area 52, and the defective area 53 can be recognized.
  • the needle 28 is positioned above the density scale 26, as shown at 1100B. Only a frame 61 showing the color corresponding to the density scale 26 is displayed for the defect area 51 having a priority higher than the level indicated by the needle 28 .
  • the position of the needle 28 is freely selectable by the operator, and the level of priority desired to be displayed can be selected.
  • An example of the display content of the detection target area obtained through processing by the information processing device 10 is shown, but the display content is not limited to this.
  • the priority is determined based on the detailed information of the shape and distribution of the detection target area, and by highlighting, partially displaying, or displaying in descending order, the operator can Detected points that should be particularly noted are clarified, and inspection work becomes more efficient.
  • the hardware structure of the processing unit that executes various processes of the control unit 12 is various processors as shown below.
  • the circuit configuration can be changed after manufacturing such as CPU (Central Processing Unit), which is a general-purpose processor that executes software (program) and functions as various processing units, FPGA (Field Programmable Gate Array), etc.
  • Programmable Logic Device PLD
  • ASIC Application Specific Integrated Circuit
  • One processing unit may be composed of one of these various processors, or may be composed of two or more processors of the same type or different types (eg, multiple FPGAs, or combinations of CPUs and FPGAs).
  • a plurality of processing units can be configured by one processor.
  • a processor functions as multiple processing units.
  • SoC System On Chip
  • SoC System On Chip
  • the various processing units are configured using one or more of the above various processors as a hardware structure.
  • the hardware structure of these various processors is, more specifically, an electrical circuit that combines circuit elements such as semiconductor elements.
  • Each configuration and function described above can be appropriately realized by arbitrary hardware, software, or a combination of both.
  • a program that causes a computer to execute the processing steps (procedures) described above a computer-readable storage medium (non-temporary storage medium) storing such a program, or a computer capable of installing such a program
  • the present invention can be applied.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Multimedia (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • General Health & Medical Sciences (AREA)
  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Chemical & Material Sciences (AREA)
  • Pathology (AREA)
  • Immunology (AREA)
  • Biochemistry (AREA)
  • Analytical Chemistry (AREA)
  • Artificial Intelligence (AREA)
  • Quality & Reliability (AREA)
  • Software Systems (AREA)
  • Medical Informatics (AREA)
  • Computing Systems (AREA)
  • Evolutionary Computation (AREA)
  • Databases & Information Systems (AREA)
  • Analysing Materials By The Use Of Radiation (AREA)
PCT/JP2022/039845 2021-12-13 2022-10-26 情報処理装置、情報処理方法、プログラム及び記録媒体 Ceased WO2023112497A1 (ja)

Priority Applications (4)

Application Number Priority Date Filing Date Title
EP22907031.3A EP4450959A4 (en) 2021-12-13 2022-10-26 Information processing device, information processing method, program, and recording medium
CN202280081559.4A CN118382799A (zh) 2021-12-13 2022-10-26 信息处理装置、信息处理方法、程序及记录介质
JP2023567583A JPWO2023112497A1 (https=) 2021-12-13 2022-10-26
US18/738,416 US20240331339A1 (en) 2021-12-13 2024-06-10 Information processing apparatus, information processing method, and program

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
JP2021201740 2021-12-13
JP2021-201740 2021-12-13

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US18/738,416 Continuation US20240331339A1 (en) 2021-12-13 2024-06-10 Information processing apparatus, information processing method, and program

Publications (1)

Publication Number Publication Date
WO2023112497A1 true WO2023112497A1 (ja) 2023-06-22

Family

ID=86774461

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2022/039845 Ceased WO2023112497A1 (ja) 2021-12-13 2022-10-26 情報処理装置、情報処理方法、プログラム及び記録媒体

Country Status (5)

Country Link
US (1) US20240331339A1 (https=)
EP (1) EP4450959A4 (https=)
JP (1) JPWO2023112497A1 (https=)
CN (1) CN118382799A (https=)
WO (1) WO2023112497A1 (https=)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2025115178A1 (ja) * 2023-11-30 2025-06-05 富士通株式会社 機械学習プログラム、機械学習方法、および情報処理装置
EP4586187A1 (en) * 2024-01-12 2025-07-16 Valeo Systemes Thermiques Method and system for automated characterisation of material defects through an image of the material

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2017116483A (ja) * 2015-12-25 2017-06-29 株式会社イシダ 物品検査システム
JP2018173374A (ja) * 2017-03-31 2018-11-08 松定プレシジョン株式会社 Ｘ線検査装置
WO2020003917A1 (ja) * 2018-06-29 2020-01-02 富士フイルム株式会社 欠陥表示装置及び方法
JP2021120639A (ja) * 2020-01-30 2021-08-19 株式会社日立製作所 アラート出力タイミング制御装置、アラート出力タイミング制御方法、アラート出力タイミング制御プログラム
JP2021156757A (ja) 2020-03-27 2021-10-07 日鉄ステンレス株式会社 疵検査装置、検査結果表示装置、疵検査方法及び検査結果表示方法

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8284247B2 (en) * 2008-02-15 2012-10-09 Enerize Corporation Method and apparatus for detecting and inspecting through-penetrating defects in foils and films
JP6197659B2 (ja) * 2014-01-20 2017-09-20 富士ゼロックス株式会社 検出制御装置、プログラム及び検出システム
CN108158597B (zh) * 2016-12-07 2021-08-06 北京东软医疗设备有限公司 确定原始x射线能量数据的方法、装置及ct设备
DE102021107640A1 (de) * 2020-03-30 2021-09-30 Sintokogio, Ltd. Inspektionsergebnisanzeigevorrichtung und Speichermedium
CN111596179A (zh) * 2020-05-25 2020-08-28 国网湖南省电力有限公司 基于数字成像技术的电缆缓冲层缺陷带电检测方法、系统、介质及设备

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2017116483A (ja) * 2015-12-25 2017-06-29 株式会社イシダ 物品検査システム
JP2018173374A (ja) * 2017-03-31 2018-11-08 松定プレシジョン株式会社 Ｘ線検査装置
WO2020003917A1 (ja) * 2018-06-29 2020-01-02 富士フイルム株式会社 欠陥表示装置及び方法
JP2021120639A (ja) * 2020-01-30 2021-08-19 株式会社日立製作所 アラート出力タイミング制御装置、アラート出力タイミング制御方法、アラート出力タイミング制御プログラム
JP2021156757A (ja) 2020-03-27 2021-10-07 日鉄ステンレス株式会社 疵検査装置、検査結果表示装置、疵検査方法及び検査結果表示方法

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP4450959A4

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2025115178A1 (ja) * 2023-11-30 2025-06-05 富士通株式会社 機械学習プログラム、機械学習方法、および情報処理装置
EP4586187A1 (en) * 2024-01-12 2025-07-16 Valeo Systemes Thermiques Method and system for automated characterisation of material defects through an image of the material

Also Published As

Publication number Publication date
EP4450959A4 (en) 2025-03-26
JPWO2023112497A1 (https=) 2023-06-22
EP4450959A1 (en) 2024-10-23
US20240331339A1 (en) 2024-10-03
CN118382799A (zh) 2024-07-23

Similar Documents

Publication Publication Date Title
Gobert et al. Application of supervised machine learning for defect detection during metallic powder bed fusion additive manufacturing using high resolution imaging.
JP6810505B2 (ja) 教師データ作成方法及び装置並びに欠陥検査方法及び装置
US11016039B2 (en) Measurement processing device, measurement processing method, measurement processing program, and method for manufacturing structure
US10760902B2 (en) Measurement processing device, x-ray inspection apparatus, method for manufacturing structure, measurement processing method, x-ray inspection method, measurement processing program, and x-ray inspection program
JP6767490B2 (ja) 欠陥検査装置、欠陥検査方法、およびプログラム
JP6556266B2 (ja) 欠陥検査装置、方法およびプログラム
CN117120951A (zh) 生产线的异常预兆检测装置、方法及程序以及制造装置及检点装置
US12051185B2 (en) Method and apparatus for the determination of defects during a surface modification method
JP6930926B2 (ja) 欠陥検査装置、方法およびプログラム
US20240331339A1 (en) Information processing apparatus, information processing method, and program
WO2020071162A1 (ja) 教師データ生成装置及び教師データ生成プログラム
US20090097729A1 (en) Method and system for identifying defects in a radiographic image of a scanned object
WO2020003917A1 (ja) 欠陥表示装置及び方法
US20240221343A1 (en) Display processing device, display processing method, and display processing program
US20240362888A1 (en) Information processing apparatus, information processing method, and information processing program
JP2011232302A (ja) 画像検査方法及び画像検査装置
US20240370148A1 (en) Information processing apparatus, information processing method, and program
Johnstone Comparison of Porosity Detection Methods in Infrared Imaging for Electron Beam Powder Bed Fusion
WO2026018584A1 (ja) 外観検査システム、制御装置および制御方法
CN118043651A (zh) 图像处理装置、处理系统、图像显示方法及程序

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 22907031

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 2023567583

Country of ref document: JP

Kind code of ref document: A

WWE Wipo information: entry into national phase

Ref document number: 202280081559.4

Country of ref document: CN

WWE Wipo information: entry into national phase

Ref document number: 2022907031

Country of ref document: EP

NENP Non-entry into the national phase

Ref country code: DE

ENP Entry into the national phase

Ref document number: 2022907031

Country of ref document: EP

Effective date: 20240715