JP6749655B1 - Inspection device, abnormality detection method, computer program, learning model generation method, and learning model - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an inspection device capable of accurately detecting an abnormality of an inspection object, an abnormality detection method, a computer program, a learning model generation method, and a learning model. An inspection device is based on an acquisition unit 112 that acquires an image obtained by irradiating an inspection object placed on a conveyor 61 with an electromagnetic wave, and a determination value and a threshold value calculated based on the image. When the judgment unit 114 that determines the presence or absence of abnormality in the inspection object and the image obtained by irradiating the inspection object with electromagnetic waves are input according to the judgment result of the judgment unit 114, information regarding the presence or absence of abnormality in the inspection object is input. An image obtained by irradiating an inspection object with an electromagnetic wave is input to the output learning model 152, and the learning inspection unit 117 that acquires information on the presence or absence of the abnormality and an output that outputs information on the presence or absence of the abnormality. It includes an output unit 118. [Selection diagram] Fig. 2

Description

The present invention relates to an inspection device for detecting an abnormality of an inspection object such as food, an abnormality detection method, a computer program, a learning model generation method, and a learning model.

Uses inspection waves such as X-rays, terahertz waves, infrared rays and visible light for inspection of foods such as meat and beans, inspection of packages containing contents in packaging materials, and inspection of industrial products such as electronic parts The inspection device is used. In this inspection apparatus, the inspection wave is irradiated to the inspection object, the inspection wave passing through the inspection object is detected, and the inspection image of the inspection object is acquired.

When the inspection object has a foreign substance, the inspection image has a dark portion exceeding the threshold value. Conventionally, the presence or absence of foreign matter has been determined based on the area and shape of this dark portion.
The presence or absence of abnormality in the shape of the inspection object was determined based on the area and perimeter of the blob (lump) in the inspection image. By counting the pixels forming the blob, the area of the blob is calculated, and the perimeter of the blob is calculated based on the arrangement of the background pixels forming the periphery of the blob (for example, Patent Document 1).

JP, 2005-31069, A

For the inspection image of the inspection object acquired based on a predetermined image processing algorithm, in the conventional abnormality detection method, the presence or absence of abnormality was determined by comparing with a threshold, depending on the type of inspection object and foreign matter, In some cases, the difference in pixel value between the normal portion and the foreign matter is small, and there is a problem that the detection accuracy is not good.
Similarly, when detecting an abnormality in the shape of the inspection object, the difference between the pixel values of the blob pixel and the background pixel is small, the edge is not clear, and it may not be accurately detected.

The present invention has been made in view of such circumstances, and provides an inspection apparatus, an abnormality detection method, a computer program, a learning model generation method, and a learning model that can accurately detect an abnormality of an inspection object. With the goal.

An inspection apparatus according to one aspect of the present invention, an acquisition unit that acquires an image obtained by irradiating an electromagnetic wave to an inspection object placed on a conveyor, and, based on the brightness of each pixel of the image, A specification unit that specifies a blob for determining whether there is an abnormality in the inspection object, and a judgment value is calculated based on the blob specified by the specification unit, and the calculated judgment values are the first threshold value on the normal side and the second threshold value on the abnormal side. and whether a determination unit whether the intermediate value between the threshold value, if the determination section determines that the is an intermediate value, when input images including the identified blobs, the inspected anomaly In the learning model that outputs the information regarding the presence or absence of the input, the learning inspection unit that inputs the image including the specified blob and acquires the information regarding the presence or absence of the abnormality, and the output unit that outputs the information regarding the presence or absence of the abnormality. Prepare
Here, the “image obtained by irradiation with electromagnetic waves” includes an image based on electromagnetic waves and an image obtained by performing image processing on the image.

The abnormality detection method according to one aspect of the present invention acquires an image obtained by irradiating an electromagnetic wave to an inspection object placed on a conveyor, and based on the brightness of each pixel in the image, the inspection object The blob for determining the presence or absence of abnormality is specified, and the determination value calculated based on the specified blob is determined whether it is an intermediate value between the first threshold value on the normal side and the second threshold value on the abnormal side , When it is determined that the value is the intermediate value, when the image including the specified blob is input, the learning model that outputs information regarding the presence or absence of abnormality of the inspection object is input the image including the specified blob. Then, the computer is caused to execute a process of outputting information regarding the presence or absence of the abnormality.

A computer program according to an aspect of the present invention acquires an image obtained by irradiating an electromagnetic wave to an inspection object placed on a conveyor, and based on the brightness of each pixel in the image, the inspection object. identify determining blob the presence or absence of an abnormality, the determination value calculated based on the identified blob, it is determined whether the intermediate value between the second threshold value of the first threshold and the abnormal side of the normal side, the If it is determined that the intermediate value, when entering an image including a blob identified, the learning model for outputting information about the presence or absence of abnormality of the inspected object, and enter the image containing the identified blobs , Causing the computer to execute a process of outputting information regarding the presence or absence of the abnormality.

A method for generating a learning model according to an aspect of the present invention is, in an image obtained by irradiating an electromagnetic wave to an inspection object placed on a conveyor, specified based on the brightness of each pixel, the abnormality of the inspection object. An image including a blob for determining the presence or absence of the test data is acquired by associating the recorded teacher data with information indicating the presence or absence of abnormality of the inspection object, and based on the teacher data, the judgment value calculated based on the specified blob is If it is determined that the intermediate the second threshold value of the first threshold and the abnormal side of the normal side, when the image including the blobs identified is input, outputs the information about the presence or absence of abnormality of the inspected Generate a learning model.

The learning model according to one aspect of the present invention, among the images obtained by irradiating electromagnetic waves to the inspection object placed on the conveyor, identified based on the brightness of each pixel, the presence or absence of abnormality of the inspection object When the judgment value calculated based on the judgment blob is intermediate between the first threshold value on the normal side and the second threshold value on the abnormal side, an input layer to which an image including the specified blob is input. , A parameter is learned using teacher data recorded by associating an output layer that outputs information regarding whether or not there is an abnormality in the inspection object, an image including the specified blob, and information that indicates whether or not there is an abnormality in the inspection object. When an image including a specified blob is input to the input layer, the output layer outputs information regarding the presence or absence of abnormality of the inspection object through an operation by the intermediate layer. Make the computer work.

According to the present invention, with respect to an image based on an image obtained by irradiating an electromagnetic wave to a test object, an image is input to a learning model according to whether there is an abnormality in the test object determined based on a judgment value and a threshold value. To acquire the presence or absence of abnormality. For example, when it is uncertain whether or not there is an abnormality in the inspection object that is determined based on the threshold value, by inputting an image into the learning model and acquiring the presence or absence of abnormality, it is possible to accurately detect the presence or absence of foreign matter, the abnormality in shape, etc. The presence or absence of abnormality can be detected.

FIG. 3 is a perspective view showing the configuration of the abnormality detection system according to the first embodiment. It is a block diagram which shows the structure of an abnormality detection system. It is explanatory drawing which shows the state which extracted the blob about the test object which is fish meat. It is explanatory drawing which shows the state which cut out each blob. It is explanatory drawing which shows the state which enclosed the blob by the bounding box. It is explanatory drawing of the process of resizing. It is explanatory drawing which shows the test|inspection image containing the blob with which the abnormal value was determined by the abnormality detection model to be 0.5 or more. It is explanatory drawing which shows an example of the record layout of inspection image DB. It is explanatory drawing regarding the production|generation process of an abnormality detection model. 6 is a flowchart showing an example of a processing procedure of an abnormality detection model generation processing by a control unit. 6 is a flowchart illustrating an example of a processing procedure of abnormality detection processing by a control unit. It is explanatory drawing which shows an example of the display screen in a display. It is a flow chart which shows an example of a processing procedure of re-learning processing by a control part of PC. It is a flow chart which shows an example of a processing procedure of re-learning processing of a modification by a control part of PC. 5 is a block diagram showing a configuration of an abnormality detection system according to Embodiment 2. FIG. It is explanatory drawing which shows an example of the display screen which displays the selection button of levels 1-4. It is explanatory drawing which shows an example of the display screen which displays the thresholds a and b of levels 1-4. 6 is a flowchart illustrating an example of a processing procedure of abnormality detection processing by a control unit. FIG. 9 is a block diagram showing a configuration of an abnormality detection system according to a third embodiment. 7 is a flowchart illustrating an example of a processing procedure of threshold value setting processing by a control unit. It is explanatory drawing which shows an example of the display screen for inputting the threshold value of a level.

Hereinafter, the present invention will be described in detail with reference to the drawings showing an embodiment thereof.
(Embodiment 1)
FIG. 1 is a plan view showing the configuration of an abnormality detection system 10 according to the first embodiment, and FIG. 2 is a block diagram showing the configuration of the abnormality detection system 10.
The abnormality detection system 10 includes an information processing device 1 as an inspection device and an X-ray inspection machine 2. The information processing device 1 and the X-ray inspection machine 2 may be integrated. In the present embodiment, the inspection object 4 is placed on the conveyor belt 61 of the conveyor unit (conveyor) 6 of the X-ray inspection machine 2, and the abnormality is detected on the basis of the image of the inspection object 4 captured by using X-rays. The abnormality detection system 10 for detecting the will be described. In the abnormality detection system 10, the presence/absence of abnormality of the inspection object 4 is determined based on a predetermined image processing algorithm, and the presence/absence of abnormality is estimated based on the result, based on the abnormality detection model 152. Examples of the inspection object 4 include foods such as raw meat and fish meat.

The X-ray inspector 2 includes a horizontally-long rectangular parallelepiped lower housing 23 and an upper housing 21 provided on the lower housing 23 and smaller than the lower housing 23.
The transport unit 6 is provided in the lower housing 23. The transport unit 6 is provided with an upstream roller 63 on the outer side of one end in the longitudinal direction of the lower housing 23 and a downstream roller 62 on the outer side of the other end. Between the upstream roller 63 and the downstream roller 62, two lower rollers 64 and 65 are provided below the upstream roller 63 and the downstream roller 62, and the conveyor belt 61 is provided between these rollers. Has been bridged. A driving force from a conveyance motor (not shown) is applied to one of the upstream roller 63 and the downstream roller 62, and when the inspection object is inspected, the conveyance belt 61 rotates clockwise at a constant speed V. Orbit.

An X-ray irradiation unit 3 that irradiates X-rays as electromagnetic waves is housed in the upper housing 21. The X-ray irradiation unit 3 has an X-ray tube 31. Examples of electromagnetic waves other than X-rays include terahertz waves, infrared rays and visible light.
A display 22 is provided on the front surface of the upper housing 21. The display 22 is a display device such as an LCD (Liquid Crystal Display).
The X-ray detection unit 7 is provided below the transport belt 61 that moves between the upstream roller 63 and the downstream roller 62. The X-ray detection unit 7 is called a TDI (Time Delay Integration) camera or a TDI sensor, and includes a scintillator that emits fluorescence according to the intensity of X-rays, and many photodiodes that detect the fluorescence emitted by the scintillator. have.

In the TDI camera, the photodiodes are linearly arranged in the X direction orthogonal to the Y direction, which is the moving direction of the object to be inspected, to form one detection line 71, and a plurality of detection lines are provided. The plurality of detection lines 71 are arranged in parallel in the Y direction, which is the moving direction of the inspection object. In FIG. 1, each detection line 71 is shown as an independent line sensor element, but in reality, a plurality of photodiodes are regularly arranged in the X direction and the Y direction in the housing of one TDI camera. One line of the detection line 71 is configured by the photodiodes arranged side by side and arranged in the X direction. An X-ray image is formed by the data of each line received and output by each detection line 71.

The information processing device 1 is an information processing device capable of various information processing and information transmission/reception, and is, for example, a personal computer, a server device, or the like. The X-ray inspection machine 2 is communicatively connected to the information processing apparatus 1 via a network N such as a LAN. When the information processing device 1 is a server device, the network N may be the Internet. Alternatively, a cloud computer communicatively connected via the network N such as the Internet may execute the process of the information processing device 1.
In the present embodiment, the information processing device 1 is assumed to be a personal computer, and will be read as PC1 for simplicity in the following description.

The PC 1 acquires an image of the inspection object 4 and performs a process of detecting the presence or absence of abnormality of the inspection object 4 based on the image. In the present embodiment, the PC 1 has already learned the abnormality detection model described later to detect (identify) the abnormality of the inspection object 4 from the image by machine learning according to the determination result of the abnormality determination unit (determination unit) 114 described later. The presence/absence of abnormality is detected using 152.
The PC 1 obtains an image obtained by irradiating the inspection object 4 with X-rays using the X-ray inspection machine 2 (a region surrounding the blob 41 is cut out as described later, and a predetermined image processing is performed on the region. (Inspection image) is input to the abnormality detection model 152, and an identification result indicating whether there is an abnormality in the blob 41 is acquired as an output.
Hereinafter, the type of abnormality of the inspection object 4 is presence/absence of foreign matter, the PC 1 inputs the inspection image to the abnormality detection model 152, and the probability value of normal (absence of foreign matter) and the probability value of abnormality (presence of foreign matter) by the softmax function Will be output and the probability value of abnormality will be acquired as an abnormal value.
The PC 1 sorts out whether or not to discard the inspection object 4 according to the acquired abnormal value.

The PC 1 includes a control unit 11, a main storage unit 12, a communication unit 13, a display unit 14, an auxiliary storage unit 15, and an input unit 16.
The control unit 11 has an arithmetic processing unit such as one or more CPUs (Central Processing Units), MPUs (Micro-Processing Units), and GPUs (Graphics Processing Units), and stores the programs P stored in the auxiliary storage unit 15. By reading and executing, various information processing, control processing, and the like related to the PC 1 are performed. Each functional unit in FIG. 2 is executed by the control unit 11 operating based on the program P.

The control unit 11 includes, as functional units, an image generation unit 111, an image processing unit (acquisition unit) 112, a target identification unit (identification unit) 113, an abnormality determination unit (determination unit) 114, a cutout unit 115, an image processing unit 116, and The learning inspection unit 117, a display unit (output unit) 118, a selection unit (output unit) 119, a pass/fail judgment acceptance unit (second acceptance unit) 122, and a re-learning unit 123 are included.

The image generation unit 111 transmits an inspection object 4 and generates an X-ray image based on the X-ray detected by the X-ray detection unit 7.
The image processing unit 112 performs image processing on the X-ray image using a smoothing filter, a feature extraction filter, or the like.

The target identifying unit 113 performs the blob 41 analysis and identifies the blob (area). The target specifying unit 113 extracts, for example, an emphasis line in which the brightness change of each pixel is emphasized (edge detection), and specifies a portion including the emphasis line as a blob.
The target identifying unit 113 may identify a blob whose luminance change rate is equal to or higher than a predetermined value as compared with the surrounding area.
The target specifying unit 113 may determine a standard deviation of the brightness and specify a set of pixels having a part in which the brightness changes by a predetermined level or more as a blob by comparing with the standard deviation. You may specify the part containing a as a blob.

The abnormality determination unit 114 determines whether or not the blob identified by the target identification unit 113 is a foreign substance based on the determination value and the threshold value. As the judgment value, for example, the number of images of blobs surrounded by the emphasis line, the area, and the like can be cited. By setting the threshold value on the normal side (first threshold value) a and the threshold value on the abnormal side (second threshold value) b as the judgment value. Every other time (a<b), if the calculated determination value is less than the threshold value a, it is determined that the unique portion is not a foreign substance and is normal, and if the determination value exceeds the threshold value b, the blob is determined to be a foreign substance. When the determination value is a or more and b or less, the learning inspection unit 117, which will be described later, acquires an abnormal value.
FIG. 3 shows, as an example, a state in which blobs 41 having a judgment value of a or more and b or less are extracted from the inspection object 4 which is fish meat. The blob is emphasized by image processing and is filled.

The cutout unit 115 cuts out the blob 41 having a judgment value of a or more and b or less so as to be within a rectangle of a predetermined size, surrounds it with the bounding box 81, and generates a cutout image. Here, the image surrounding the blob 41 is an image before the emphasis process. The image surrounding the blob 41 may be an image after the emphasis process.
FIG. 4 shows a state in which each blob 41 is cut out.
The image processing unit 116 performs processing such as normalization on the cut-out image and generates the inspection area (inspection image) 8 to be input to the abnormality detection model 152.

FIG. 5 is an explanatory diagram illustrating a state in which the blob identified by the target identifying unit 113 is cut out by the cutout unit 115 and the inspection image 8 in which the blob 41 is surrounded by the bounding box 81 is generated. The target specifying unit 113 of the control unit 11 detects the blob 41 as described above. In each inspection image 8, each blob 41 is surrounded by a bounding box 81 having the same size.
FIG. 6 is an explanatory diagram of the resizing process performed by the clipping unit 115.
When the inspection object 4 protrudes from the bounding box 81, the cutout portion 114 of the control unit 11 resizes the blob 41 so that the blob 41 fits inside the bounding box 81. When enlarging the blob 41, a complementary process is performed.

The learning inspection unit 117 inputs each inspection image to the abnormality detection model 152 and acquires an abnormal value.

The display unit 118 outputs the abnormal value to the display 22 and displays the presence/absence of abnormality in the inspection image on the display 22.
FIG. 7 shows an inspection image including a blob whose abnormal value is determined to be 0.5 or more by the abnormality detection model 152.

The sorting unit 119 compares the abnormal value of the inspection image 8 with a threshold value, performs sorting of whether or not to discard the inspection object 4 including the blob 41 having the abnormal value, stops the transport unit 6, and performs the inspection. Mark the object 4. For example, when the abnormal value of one blob 41 is 0.5 or more, the sorting unit 119 regards the inspection object 4 including the blob 41 as an abnormal item and sorts it if it is discarded. The threshold value of the abnormal value is not limited to 0.5. It is not limited to the case where the inspection object 4 is discarded when one blob 41 having an abnormal value equal to or greater than the threshold is included. The inspection object 4 may be discarded when a predetermined number or more of the blobs 41 show an abnormal value equal to or more than a threshold value.

The acceptability determination accepting unit 122 accepts the acceptability determination of the abnormality detection model 152 input by the operator through the input unit 16.
The re-learning unit 123 re-learns the abnormality detection model 152 by using the teacher data in which the inspection image 8 is associated with the quality determination input by the operator.

The main storage unit 12 is a temporary storage area such as an SRAM (Static Random Access Memory), a DRAM (Dynamic Random Access Memory), a flash memory, etc., and temporarily stores data necessary for the control unit 11 to execute arithmetic processing. Remember. The communication unit 13 is a communication module for performing processing relating to communication, and transmits/receives information to/from the outside.

The auxiliary storage unit 15 is a large-capacity memory, a hard disk, or the like, and stores the program P and other data necessary for the control unit 11 to execute processing. The auxiliary storage unit 15 also stores an inspection image DB 151 and the abnormality detection model 152. The inspection image DB 151 is a database that stores information about the inspection object 4 that is an abnormality detection target.

The program P stored in the auxiliary storage unit 15 may be provided by the recording medium 156 in which the program P is readably recorded. The recording medium 156 is, for example, a portable memory such as a USB (Universal Serial Bus) memory, an SD (Secure Digital) card, a micro SD card, or a compact flash (registered trademark). The program P recorded in the recording medium 156 is read from the recording medium 156 using a reading device (not shown) and installed in the auxiliary storage unit 15. When the information processing device 1 includes a communication unit that can communicate with an external communication device, the program P stored in the auxiliary storage unit 15 may be provided by communication via the communication unit.
The auxiliary storage unit 15 may be an external storage device connected to the PC 1. Further, the PC 1 may be a multi-computer including a plurality of computers, or may be a virtual machine virtually constructed by software.

The input unit 16 receives an input by a worker such as a threshold value of a determination value described later.

FIG. 8 is an explanatory diagram showing an example of the record layout of the inspection image DB 151. The inspection image DB 151 includes an inspection image ID sequence, a coordinate information sequence, an abnormal value sequence, and a judgment value sequence. The inspection image ID column stores inspection image IDs for identifying each inspection image 8. The coordinate information sequence stores coordinate information indicating the position of the blob 41 in the X-ray irradiation area on the conveyor belt 61 in association with the inspection image ID. The coordinate information is the X coordinate and the Y coordinate at the four corners of the bounding box 81 when the origins are the centers in the X direction and the Y direction between the upstream roller 63 and the downstream roller 62 in FIG. The abnormal value is the probability value of abnormality as described above. The judgment value is a judgment value calculated by the abnormality judgment unit 114.

FIG. 9 is an explanatory diagram related to generation processing of the abnormality detection model 152. FIG. 9 conceptually shows a process of performing the machine learning to generate the abnormality detection model 152.

The control unit 11 of the PC 1 inputs the inspection image 8 as an abnormality detection model 152 by learning the image feature amount of the abnormality in the inspection image 8 of the blob 41, and outputs information indicating the presence or absence of abnormality in the blob 41. Constructs (generates) a neural network. For example, the neural network is a CNN (Convolution Neural Network), and has an input layer that receives an input of the inspection image 8, an output layer that outputs a discrimination result of presence/absence of an abnormality, and an intermediate layer that extracts an image feature amount of the inspection image 8. Have.

The input layer has a plurality of neurons that receive the input of the pixel value of each pixel included in the inspection image 8, and transfers the input pixels to the intermediate layer. The intermediate layer has a plurality of neurons for extracting the image feature amount of the inspection image, and transfers the extracted image feature amount to the output layer. In FIG. 9, the number of intermediate layers is three, but the number is not limited to this. For example, when the abnormality detection model 152 is CNN, the convolution layer that convolves the pixel value of each pixel input from the input layer and the pooling layer that maps the pixel value that is convoluted by the convolution layer alternate in the intermediate layer. The feature amount of the image is finally extracted while the pixel information of the inspection image 8 is compressed. In FIG. 9, the description of the convolution layer and the pooling layer is omitted. The output layer has two neurons that output an identification result that identifies an abnormality in the blob 41, and identifies the presence or absence of abnormality in the blob 41 based on the image feature amount output from the intermediate layer. One neuron outputs a probability value of normal (absence of foreign substance) in the blob 41, and another neuron outputs a probability value of abnormality (existence of foreign substance) in the blob 41.

In the present embodiment, the abnormality detection model 152 is described as being CNN, but the abnormality detection model 152 is not limited to CNN, and a neural network other than CNN, SVM (Support Vector Machine), Bayesian network, regression tree It may be a trained model constructed by another learning algorithm.

The control unit 11 associates the plurality of inspection images 8 with the information indicating whether or not the blob 41 in each inspection image 8 is abnormal (whether or not the foreign object is present, or is normal without the foreign object) ( Learning is performed using abnormal data and normal data. For example, as shown in FIG. 9, the teaching data is labeled with respect to the inspection image 8 of the blob 41, the ID of the inspection image 8, the type of abnormality (presence/absence of foreign matter), and the rank (1: abnormal, 0: normal). It is the attached data.

The control unit 11 inputs the inspection image 8 that is the teacher data to the input layer, obtains the identification result indicating the presence or absence of abnormality in the blob 41 from the output layer through the arithmetic processing in the intermediate layer. The identification result output from the output layer is a continuous probability value (for example, a value in the range from “0” to “1”) when the softmax function is used. The identification result may be a value (for example, a value of “0” or “1”) that discretely indicates the presence or absence of abnormality.

The control unit 11 compares the identification result output from the output layer with the information labeled on the inspection image 8 in the teacher data, that is, the correct value, so that the output value from the output layer approaches the correct value. The parameters used for the arithmetic processing in the intermediate layer are optimized. The parameters are, for example, weights between neurons (coupling coefficient), coefficients of an activation function used in each neuron, and the like. The method of optimizing the parameters is not particularly limited, but for example, the control unit 11 optimizes various parameters using the error back propagation method.

The control unit 11 performs the above process on each inspection image 8 included in the teacher data to generate the abnormality detection model 152. When the control unit 11 acquires the inspection image 8 of the blob 41, the control unit 11 uses the abnormality detection model 152 to detect the presence/absence of foreign matter in the blob 41.

FIG. 10 is a flowchart illustrating an example of a processing procedure of generation processing of the abnormality detection model 152 by the control unit 11.
The control unit 11 acquires the teacher data in which the inspection images 8 of the plurality of blobs 41 are associated with the information indicating the abnormality of the blobs 41 in each inspection image 8 (S11).

The control unit 11 uses the teacher data to generate the abnormality detection model 152 (learned model) that outputs the abnormality presence/absence information of the blob 41 when the inspection image 8 of the blob 41 is input (S12). Specifically, the control unit 11 inputs the inspection image 8 which is the teacher data to the input layer of the neural network, and acquires the identification result for identifying the presence or absence of the foreign matter in the inspection image 8 from the output layer. The control unit 11 compares the acquired identification result with the correct value of the teacher data (information labeled on the inspection image 8), and in the intermediate layer so that the identification result output from the output layer approaches the correct value. The parameters (weight, etc.) used in the calculation processing of are optimized. The control unit 11 stores the generated abnormality detection model 152 in the auxiliary storage unit 15 and ends the series of processes.

FIG. 11 is a flowchart showing an example of a processing procedure of abnormality detection processing by the control unit 11.
The control unit 11 of the PC 1 transmits an inspection object 4 and generates an X-ray image based on the X-ray detected by the X-ray detection unit 7 (S21).
The control unit 11 performs image processing on the X-ray image using a smoothing filter, a feature extraction filter, etc. (S22).
The control unit 11 performs the blob analysis as described above and identifies the blob (S23). The control unit 11 extracts an emphasis line in which the luminance change of each pixel is emphasized (edge detection), and identifies a portion including the emphasis line as a blob.

The control unit 11 determines whether or not the blob is a foreign substance (S24). The control unit 11 calculates a judgment value such as the number of blob images surrounded by an emphasis line, an area, and the like, and makes a judgment based on the calculated judgment value and a threshold value. The control unit 11 determines whether it is abnormal (NG: judgment value>threshold value b), normal (OK: judgment value<threshold value a), or in between (a≦judgment value≦b).

The control unit 11 determines whether or not it is determined to be NG (abnormal) or OK (normal) in S24 (S25). When the control unit 11 determines that it is determined to be NG or OK (S25: YES), it determines whether it is OK (S26).
When the control unit 11 determines that the result is OK (S26: YES), the learning inspection is skipped and the process ends.
When it is determined that the control unit 11 is not OK, that is, NG (S26: NO), the learning inspection is skipped and the process proceeds to S33.

When the control unit 11 determines that it is not NG or OK (S25: NO), the blob 41 determined to be not NG or OK is cut out so as to be in a rectangle of a predetermined size, and a cutout image is generated. (S27). The cut-out image is an image before the blob 41 is filled.
The control unit 11 performs processing such as normalization on the cutout image to generate an inspection image (S28).

The control unit 11 inputs the inspection image 8 to the abnormality detection model 152, outputs the probability value of normality and the probability value of abnormality, and acquires the abnormality value (S29).
The control unit 11 determines whether or not the abnormal value is greater than or equal to the threshold value (S30). When the control unit 11 determines that the abnormal value is equal to or larger than the threshold value (S30: YES), the abnormal value is output to the display 22 and the inspection image 8 is displayed in a state of being surrounded by the red bounding box (BB) 82 ( S31). When the control unit 11 determines that the abnormal value is not greater than or equal to the threshold value (S30: NO), the inspection image 8 is displayed in a state where it is not surrounded by the red bounding box (BB) 82 (S32).
The control unit 11 specifies the inspection object 4 based on the abnormal value of the inspection image 8 and the inspection image ID, and when it is determined to be NG in S26, selects whether or not to discard the inspection object 4 ( S33), the process ends.

As described above, according to the present embodiment, even in the case where the difference in pixel value between the foreign matter and the other portion is small in the inspection image 8 and it is difficult to determine the presence or absence of the foreign matter by the conventional image processing method. By using the abnormality detection model 152, it is possible to accurately determine the foreign matter.
According to the present embodiment, with respect to the blob 41 acquired based on a predetermined image processing algorithm, the determination result whether the blob 41 is a foreign matter acquired based on the determination value and the threshold values a and b is accurate. If not, by inputting the inspection image 8 to the abnormality detection model 152 and acquiring the abnormal value, it is possible to detect the presence or absence of a foreign object with high accuracy.
It is not necessary for the operator to visually detect the abnormality, and the efficiency of the abnormality detection work can be improved.
In the above-described embodiment, the presence/absence of foreign matter is mentioned as the type of abnormality, but the present invention is not limited to this, and other abnormality such as a shape abnormality may be detected.

When the judgment value is represented by a numerical value from 0 to 1, for example, 0 may be the threshold value a and 1 may be the threshold value b. In this case, when the judgment value is 0, it is normal, and when it is 1, It is abnormal.
Further, the learning inspection is not limited to the case where the judgment value is a or more and b or less, and the learning inspection may be performed when the judgment value is a or more. When the judgment value is b or less, the learning inspection is performed. May be performed.
The image processing in S28 can be omitted. The inspection image 8 may be an image after the blob 41 is emphasized.

FIG. 12 is an explanatory diagram showing an example of the display screen 5 on the display 22.
The display screen has a first screen 51 on the left side and a second screen 52 on the right side in FIG.
The control unit 11 displays the inspection image 8 on the first screen 51 in association with the inspection object 4 placed on the conveyor belt 61 to the position of the conveyor belt 61. The control unit 11 displays the inspection images 8 corresponding to the blobs 41 on the second screen 52 side by side in the order of the inspection image IDs, and the abnormal inspection image 8 is surrounded by the red bounding box 82. In the case of FIG. 12, when the inspection image 8 related to the blob 41 is surrounded by the bounding box 82, the blob 41 is a foreign substance, and the inspection object 4 including the blob 41 is an abnormal product.

The operator can select one of the inspection images 8 on the second screen 52 by the selection button 56 including “>” and “<” for moving in the row direction of the inspection image 8 in order. The inspection image 8 may be selected by a pointing device such as a mouse.
On the second screen 52, an OK button 53 that is pressed when the operator determines that the estimation result (judgment) of the abnormality detection model 152 is correct, an NG button 54 that is pressed when it is determined that the judgment is not correct, And an Other button 55 which is used when inputting an instruction for selecting a level of presence or absence of abnormality described later.

For the inspection image 8 selected by the selection button 56, when the operator accepts the input of the judgment of the abnormality detection model 152 by the OK button 53 or the NG button 54 and the data of a predetermined number or more is collected, the control unit 11 The abnormality detection model 152 can be re-learned using the inspection image 8 and the quality of the judgment input by the operator as teacher data.

FIG. 13 is a flowchart showing an example of the processing procedure of the above-mentioned re-learning processing by the control unit 11 of the PC 1.

The control unit 11 accepts whether the abnormality detection model 152 is good or bad for the selected inspection image 8 based on the operator pressing the OK button 53 or the NG button 54 on the second screen 52 (S41). When the control unit 11 determines that the determination of acceptability is not accepted (S41: NO), the determination process is repeated.
When it is determined that the control unit 11 has accepted the predetermined number of pass/fail determinations (S41: YES), the inspection image 8 of the blob 41 in which the abnormality is detected is associated with the pass/fail determination of the control unit 11 input by the operator. The attached teacher data is acquired (S42).
The control unit 11 uses the teacher data to generate the abnormality detection model 152 that outputs the abnormality presence/absence information of the blob 41 when the inspection image of the blob 41 is input (S43).
By the above, it is relearned. As a result, the abnormality detection model 152 is updated as the abnormality detection processing is continued, and more accurate abnormality detection can be performed.
In addition, the quality of the determination of the abnormality detection model 152 by the operator may be determined based on the X-ray image instead of the inspection image 8.

FIG. 14 is a flowchart showing an example of a processing procedure of a re-learning process of a modified example by the control unit 11 of the PC 1.

The control unit 11 accepts whether the abnormality detection model 152 is good or bad for the selected inspection image 8 based on the operator pressing the OK button 53 or the NG button 54 on the second screen 52 (S51). When the control unit 11 determines that the determination of acceptability is not accepted (S51: NO), the determination process is repeated.
When the control unit 11 determines that the acceptance/rejection determination is accepted (S51: YES), it determines whether or not the acceptance is accepted (S52). When it is determined that the control unit 11 has not accepted the request (S52: NO), the process returns to S51.
When the control unit 11 determines that the answer is NO (S52: YES), when a predetermined number of the results of the NO determination are acquired, the teacher associates the inspection image 8 of the blob 41 that has detected an abnormality with the determination result. Data is acquired (S53).
The control unit 11 uses the teacher data to generate the abnormality detection model 152 that outputs the abnormality presence/absence information of the blob 41 when the inspection image of the blob 41 is input (S54).
By the above, re-learning is performed. As a result, it becomes possible to perform accurate abnormality detection with a smaller amount of teacher data.

(Embodiment 2)
In the PC 1 of the abnormality detection system 10 according to the second embodiment, the control unit 11 detects the presence or absence of foreign matter based on the four image processing algorithms, and stores the program P corresponding to each image processing algorithm in the auxiliary storage unit 15. Then, the same processing as that of the PC 1 according to the first embodiment is performed except that a plurality of abnormality detection models 152 to 155 are stored. The types of foreign matter detected based on the four image processing algorithms may be all the same, some of them may be the same, or all of them may be different. Examples of the type of foreign matter include bone, metal, wood, and synthetic resin. When the type of foreign matter is the same, the calculation method of the judgment value is different.
Further, different types of abnormalities in the inspection object 4 may be detected by the image processing algorithm. The types of abnormality include presence/absence of foreign matter, shape abnormality, and the like.
The number of image processing algorithms is not limited to four.

For example, in the image processing algorithm of level 1, as in the first embodiment, the target identifying unit 113 identifies, as the blob 41, the portion including the enhancement line extracted by performing the enhancement processing (edge detection) on the luminance change of each pixel. .. The abnormality determination unit 114 makes the determination based on the determination values such as the number of images and the area of the blob 41 surrounded by the emphasis line, for example. Threshold values a and b are set as the judgment value, and if the judgment value is less than the threshold value, it is determined that the unique portion is not a foreign substance.
In the level 2 image processing algorithm, the target identifying unit 113 extracts, as the blob 41, a peculiar portion in which the rate of change in luminance is equal to or higher than a predetermined value as compared with the surrounding area. The abnormality determination unit 114 makes the determination based on, for example, a determination value such as the number of pixels whose change rate is equal to or greater than a predetermined value.
In the level 3 image processing algorithm, the standard deviation of the luminance is calculated, and compared with this standard deviation, the unique portion where the luminance changes by a predetermined level or more is specified as the blob 41. In the level 4 image processing algorithm, a peculiar portion having different brightness is specified as the blob 41. The abnormality determination unit 114 performs image matching processing to determine whether or not the shape of these unique portions may be, for example, the shape of bone.

FIG. 15 is a block diagram showing the configuration of the abnormality detection system 10. The same parts as those in FIG. 2 are denoted by the same reference numerals and detailed description thereof will be omitted.
The control unit 11 includes a level receiving unit (first receiving unit) 120. The level corresponds to an image processing algorithm for detecting the presence/absence of foreign matter, and it is assumed here that there are four types of levels 1 to 4.
The level receiving unit 120 receives the selection of the level by the worker. When the operator presses the Other button 55 as shown in FIG. 16, the control unit 11 displays the buttons 57a to 57d of "level 1" to "level 4" on the first screen 51. The operator selects the button of the desired level.
The target identifying unit 113 identifies the blob 41 based on the image processing algorithm corresponding to the selected level, and the abnormality determining unit 114 determines whether the blob 41 is abnormal or normal. It should be noted that depending on the image processing algorithm, not only the processing of the target specifying unit 113 and the abnormality determining unit 114 may be different, but the processing of the image processing unit 112 and the image processing unit 116 may be different.

The abnormality determination unit 114 calculates a determination value for the blob 41 identified by the target identification unit 113 based on an image processing algorithm. In level 1, for example, when the target specifying unit 113 specifies the portion including the emphasis line extracted by performing the brightness change of each pixel as the blob 41, the abnormality determination unit 114 determines that the blob 41 surrounded by the emphasis line is A judgment value such as the number of images and the area is calculated, and the judgment is performed based on the calculated judgment value. The threshold value a on the normal side and the threshold value b on the abnormal side are set for the judgment value (a<b), and if the calculated judgment value is less than the threshold value a, the singular part is judged to be normal, not a foreign substance, and the judgment value Is greater than or equal to the threshold value b, it is determined that the unique portion is a foreign matter and the blob 41 is abnormal.

When the operator presses the Other button 55 twice in succession as shown in FIG. 17, the control unit 11 displays the threshold values a and b of “level 1” to “level 4” on the first screen 51. indicate. The judgment value at each level is calculated based on each image processing algorithm, and the contents differ depending on the level, the thresholds a and b are different, and the normal and abnormal ranges are also different.

Similarly to the above, the control unit 11 learns, as the abnormality detection model 152, the image feature amount of the abnormality in the inspection image 8 related to the blob 41 generated by the image processing algorithm corresponding to the level 1, thereby performing the inspection. A neural network is generated which receives the image 8 as an input and outputs information indicating the presence or absence of foreign matter. The neural network is a CNN, and has an input layer that receives the input of the inspection image 8, an output layer that outputs the identification result of the presence or absence of abnormality, and an intermediate layer that extracts the image feature amount of the inspection image 8.

The control unit 11 performs learning using teacher data in which a plurality of inspection images 8 are associated with information indicating whether the blob 41 in each inspection image 8 is abnormal (whether a foreign object is present or normal). The teacher data is data in which the inspection image 8 of the blob 41 is labeled with the ID of the inspection image 8, the type of abnormality (presence/absence of foreign matter), and the rank (1: abnormal, 0: normal).
The control unit 11 inputs the inspection image 8 which is the teacher data to the input layer, and through the arithmetic processing in the intermediate layer, the identification result indicating the presence or absence of abnormality of the blob 41 from the output layer (probability value of normality and abnormality (foreign matter (Probability value of Yes) is output and the probability value of abnormality is acquired as an abnormal value. The identification result output from the output layer is a continuous probability value (for example, a value in the range from “0” to “1”) when the softmax function is used.

The control unit 11 compares the identification result output from the output layer with the information labeled on the inspection image 8 in the teacher data, that is, the correct value, so that the output value from the output layer approaches the correct value. The parameters used for the arithmetic processing in the intermediate layer are optimized. The parameters are, for example, weights between neurons (coupling coefficient), coefficients of an activation function used in each neuron, and the like. The method of optimizing the parameters is not particularly limited, but for example, the control unit 11 optimizes various parameters using the error back propagation method.

The control unit 11 performs the above process on each inspection image 8 included in the teacher data to generate the abnormality detection model 152. When the control unit 11 acquires the inspection image 8 of the blob 41, the control unit 11 uses the abnormality detection model 152 to detect the presence or absence of abnormality of the blob 41.
Similarly, the abnormality detection models 153-155 input the inspection image 8 by learning the image feature amount of the abnormality in the inspection image 8 related to the blob 41 generated by the image processing algorithm corresponding to the levels 2-4. Then, a neural network that outputs information indicating the presence or absence of foreign matter is generated.

The learning inspection unit 117 inputs the inspection image 8 into any of the abnormality detection models 152 to 155 in accordance with the selected level, outputs the above two probability values, and outputs the probability value (abnormal value) of having a foreign substance. get.

FIG. 18 is a flowchart showing an example of a processing procedure of abnormality detection processing by the control unit 11.
The control unit 11 of the PC 1 transmits an inspection object 4 and generates an X-ray image based on the X-ray detected by the X-ray detection unit 7 (S61).
The control unit 11 determines whether or not the level has been accepted by the worker as described above (S62). When the control unit 11 determines that the level has not been received (S62: NO), the determination is repeated.
When the control unit 11 determines that the level is accepted (S62: YES), the program P of the corresponding level is read from the auxiliary storage unit 15 (S63).
The control unit 11 performs image processing on the X-ray image with a smoothing filter, a feature extraction filter, or the like based on the image processing algorithm corresponding to the received level (S64).
The control unit 11 performs the blob analysis based on the image processing algorithm corresponding to the received level, and identifies the blob 41 (S65).

The control unit 11 calculates a judgment value (S66).
The control unit 11 determines whether or not a≦determination value≦b (S67). When the control unit 11 determines that a≦judgment value≦b is not satisfied (S67: NO), the judgment value is less than a or exceeds b, so it is determined whether or not the judgment value <a (S68).
When the control unit 11 determines that the determination value is <a (S68: YES), the control unit 11 skips the learning test and ends the process because it is normal and has no foreign matter.
When the determination value <a, that is, the determination value>b, the control unit 11 determines that there is a foreign substance and is definitely abnormal (S68: NO), the learning test is skipped and the process proceeds to S76.

When the control unit 11 determines that a≦judgment value≦b (S67: YES), the blob 41 with a≦judgment value≦b is cut out so as to be within a rectangle of a predetermined size, and a cutout image is generated. Yes (S69).
The control unit 11 performs processing such as normalization on the cutout image to generate an inspection image (S70).
The control unit 11 selects the abnormality detection model 152 corresponding to the received level (S71).
The control unit 11 inputs the inspection image 8 to the abnormality detection model 152, outputs the probability value of normality and the probability value of abnormality, and acquires the abnormality value (S72).
The control unit 11 determines whether or not the abnormal value is greater than or equal to the threshold value (S73). When the control unit 11 determines that the abnormal value is equal to or greater than the threshold value (S73: YES), the abnormal value is output to the display 22, and the inspection image 8 is displayed in a state of being surrounded by the red bounding box (BB) 82 ( S74). When the control unit 11 determines that the abnormal value is not greater than or equal to the threshold value (S73: NO), the inspection image 8 is displayed without being surrounded by the red bounding box (BB) 82 (S75).
Based on the abnormal value of the blob 41 and the inspection image ID, the control unit 11 identifies the inspection object 4 when it is determined to be NG in S68, and selects whether to discard the inspection object 4 (S76). ), the processing ends.
In the present embodiment, it is possible to select the image processing algorithm and detect the presence or absence of abnormality for a desired type of abnormality at a desired level.
Furthermore, when the target identification in S65 is performed based on the X-ray image, the image processing in S64 can be omitted. The image processing of S70 can also be omitted.

(Embodiment 3)
The PC 1 according to the third embodiment performs the same processing as the PC according to the first embodiment, except that the control unit 11 has the threshold setting unit (setting unit) 121.
FIG. 19 is a block diagram showing the configuration of the abnormality detection system 10. The same parts as those in FIG. 15 are designated by the same reference numerals, and detailed description thereof will be omitted.
The control unit 11 has a threshold value setting unit 121. The threshold value setting unit 121 sets (updates) the threshold value when the operator accepts the setting of the threshold value a and the threshold value b for any level by the input unit 16.

FIG. 20 is a flowchart showing an example of a processing procedure of threshold setting processing by the control unit 11.
The control unit 11 determines whether or not the operator has accepted the selection of the level (S81). When the worker presses the Other button 55 once, the control unit 11 displays the above-described first screen 51 shown in FIG. 16, and the worker selects the level. When the control unit 11 determines that the level is not accepted (S81: NO), this determination process is repeated.
When determining that the selection of the level has been accepted (S81: YES), the control unit 11 determines whether or not the threshold a and the threshold b have been accepted for this level (S82). As shown in FIG. 21, when the level is selected, the control unit 11 displays the level threshold value a and the threshold value b on the first screen 51. In FIG. 21, the screen when level 1 is selected is displayed. The operator inputs the threshold value a and the threshold value b of the selected level. When the control unit 11 determines that the threshold value a and the threshold value b are not accepted (S82: NO), this determination process is repeated.
When the control unit 11 receives the thresholds a and b (S82: YES), the thresholds a and b are set and the thresholds a and b of levels 1 to 4 in FIG. 17 are displayed (S83), and the process ends. To do.

According to the present embodiment, when the detection result by the abnormality determination unit 114 is estimated to be uncertain based on the set threshold value, it is possible to properly determine the presence or absence of abnormality using the learning model.

As described above, the inspection device according to one aspect of the present invention is an acquisition unit that acquires an image obtained by irradiating an electromagnetic wave to the inspection object placed on the conveyor, and the brightness of each pixel in the image. On the basis of the specification part for specifying a blob for determining whether there is an abnormality in the inspection object, and a judgment value is calculated based on the blob specified by the specification part, and the calculated judgment value is the normal first threshold value. an intermediate of whether the determination unit is a value of the second threshold value of the abnormal side, if the determination section determines that the is an intermediate value, when input images including the identified blobs , A learning inspection unit for inputting an image including the specified blob to a learning model for outputting information regarding presence/absence of abnormality of the inspection object, and outputting information regarding presence/absence of abnormality of the learning inspection unit And an output unit for

According to the above configuration, with respect to the image obtained by irradiating the inspection object with the electromagnetic wave, the image is input to the learning model according to the presence or absence of the abnormality of the inspection object determined based on the determination value and the threshold, and Get presence or absence. For example, when it is uncertain whether or not there is an abnormality in the inspection object that is determined based on the threshold value, by inputting an image into the learning model and acquiring the presence or absence of abnormality, it is possible to accurately detect the presence or absence of foreign matter, the abnormality in shape, etc. The presence or absence of abnormality can be detected.
According to the above configuration, it is possible to accurately detect the presence or absence of abnormality when the result detected based on the determination value calculated by applying the predetermined image processing algorithm and the threshold value is estimated to be uncertain.
According to the above configuration, the blob for determining the presence/absence of abnormality is identified, the presence/absence of abnormality is determined for the identified blob, and the learning model is applied according to the determination result, so that the presence/absence of abnormality can be detected efficiently.

In the above inspection device, the specifying unit applies the images processing algorithms, identifying the blob.

The above-described inspection device may include a first reception unit that receives the selection of the image processing algorithm.

According to the above configuration, the image processing algorithm can be selected according to the content of the abnormality, and the information regarding the presence or absence of the abnormality of the inspection object can be acquired with high accuracy.

The above-described inspection device may include a setting unit that sets the first threshold value and the second threshold value .

According to the above configuration, whether or not to apply the learning model is determined based on the threshold value. Therefore, when the determination result by the determination unit is estimated to be uncertain, the learning model is used to determine whether or not there is an abnormality. You can judge.

In the above-described inspection apparatus, the output unit may include a second reception unit that displays the image and receives the quality of the output result of the output unit with respect to the displayed image.

According to the above configuration, the operator can acquire the quality of the determination result of the learning model and re-learn the image.

The above-described inspection device may include a re-learning unit that re-learns the learning model based on the image and the information regarding the quality.

According to the above configuration, it is possible to re-learn the quality of the determination of the output result received by the operator as teacher data, and generate a learning model with higher accuracy.

In the above-described inspection apparatus, the re-learning unit may re-learn when the second receiving unit accepts no.

According to the above configuration, when the second accepting unit accepts no, re-learning is performed with the teacher data in which the image and the quality of the determination of the output result are associated with each other. A learning model that is good can be generated.

The abnormality detection method according to one aspect of the present invention acquires an image obtained by irradiating an electromagnetic wave to an inspection object placed on a conveyor, and based on the brightness of each pixel in the image, the inspection object The blob for determining the presence or absence of abnormality is specified, and the determination value calculated based on the specified blob is determined whether it is an intermediate value between the first threshold value on the normal side and the second threshold value on the abnormal side , When it is determined that the value is the intermediate value, when the image including the specified blob is input, the learning model that outputs information regarding the presence or absence of abnormality of the inspection object is input the image including the specified blob. Te, before outputting the information as to whether or not Kikoto normal to execute the process to the computer.

According to the above configuration, the learning model is used to acquire information regarding the presence/absence of abnormality according to the determination result. For example, when the determination result is estimated to be uncertain, an image is input to the learning model to determine whether there is abnormality. By acquiring, it is possible to accurately detect the presence/absence of a foreign matter, the presence/absence of an abnormality such as a shape abnormality.

A computer program according to an aspect of the present invention acquires an image obtained by irradiating an electromagnetic wave to an inspection object placed on a conveyor, and based on the brightness of each pixel in the image, the inspection object. identify determining blob the presence or absence of an abnormality, the determination value calculated based on the identified blob, it is determined whether the intermediate value between the second threshold value of the first threshold and the abnormal side of the normal side, the If it is determined that the intermediate value, when entering an image including a blob identified, the learning model for outputting information about the presence or absence of abnormality of the inspected object, and enter the image containing the identified blobs , before outputting the information as to whether or not Kikoto normal to execute the process to the computer.

According to the above configuration, the learning model is used to acquire information regarding the presence/absence of abnormality according to the determination result. For example, when the determination result is estimated to be uncertain, an image is input to the learning model to determine whether there is abnormality. By acquiring, it is possible to accurately detect the presence/absence of a foreign matter, the presence/absence of an abnormality such as a shape abnormality.

A method for generating a learning model according to an aspect of the present invention is, in an image obtained by irradiating an electromagnetic wave to an inspection object placed on a conveyor, specified based on the brightness of each pixel, the abnormality of the inspection object. An image including a blob for determining the presence or absence of the test data is acquired by associating the recorded teacher data with information indicating the presence or absence of abnormality of the inspection object, and based on the teacher data, the judgment value calculated based on the specified blob is If it is determined that the intermediate the second threshold value of the first threshold and the abnormal side of the normal side, when the image including the blobs identified is input, outputs the information about the presence or absence of abnormality of the inspected Generate a learning model.

According to the above configuration, the presence or absence of the abnormality can be accurately obtained when an image obtained by applying a predetermined image processing algorithm corresponding to the abnormality content of the inspection object is input.

The learning model according to one aspect of the present invention, among the images obtained by irradiating electromagnetic waves to the inspection object placed on the conveyor, identified based on the brightness of each pixel, the presence or absence of abnormality of the inspection object When the judgment value calculated based on the judgment blob is intermediate between the first threshold value on the normal side and the second threshold value on the abnormal side, an input layer to which an image including the specified blob is input. , A parameter is learned using teacher data recorded by associating an output layer that outputs information regarding whether or not there is an abnormality in the inspection object, an image including the specified blob, and information that indicates whether or not there is an abnormality in the inspection object. When an image including a specified blob is input to the input layer, the output layer outputs information regarding the presence or absence of abnormality of the inspection object through an operation by the intermediate layer. Make the computer work.

According to the above configuration, the presence or absence of the abnormality can be accurately obtained when an image obtained by applying a predetermined image processing algorithm corresponding to the abnormality content of the inspection object is input.

The embodiments disclosed this time are to be considered as illustrative in all points and not restrictive. The scope of the present invention is shown not by the above meaning but by the scope of the claims, and is intended to include meanings equivalent to the scope of the claims and all modifications within the scope.
For example, the inspection item is not limited to food, and may be a package, an industrial product, or the like.
The electromagnetic waves with which the inspection object 4 is irradiated are not limited to X-rays, and may be terahertz waves, infrared rays, visible light, or the like.
Further, the target specifying unit 113 may extract the blobs 41 for which a≦judgment value≦b.

1 PC (inspection device)
10 Abnormality Detection System 11 Control Section 12 Main Storage Section 13 Communication Section 14 Display Section 15 Auxiliary Storage Section P Program 151 Inspection Image DB
152 to 155 Abnormality detection model 156 Recording medium 2 X-ray inspection machine 22 Display 3 X-ray irradiation part 4 Inspection object 8 Inspection image 81, 82 Bounding box 5 Display screen 6 Conveying part 61 Conveying belt 7 X-ray detecting part 71 Detection line

Claims (11)

An acquisition unit that acquires an image obtained by irradiating an electromagnetic wave to the inspection object placed on the conveyor,
Of the image, based on the brightness of each pixel, a specifying unit for specifying a blob for determining the presence or absence of abnormality of the inspection object,
A judgment unit for calculating a judgment value based on the blob specified by the specifying unit, and for judging whether the calculated judgment value is an intermediate value between the first threshold value on the normal side and the second threshold value on the abnormal side ; ,
If the determination unit determines that the value is the intermediate value, when the image including the specified blob is input, the learning model that outputs information regarding the presence or absence of abnormality of the inspection object, the image including the specified blob. And a learning inspection unit that acquires information regarding the presence or absence of the abnormality,
And an output unit that outputs information regarding the presence or absence of the abnormality. The specifying unit applies the images processing algorithms, identifying the blob, inspection apparatus according to claim 1. The inspection apparatus according to claim 2 , further comprising a first reception unit that receives selection of an image processing algorithm. The inspection apparatus according to any one of claims 1 to 3, further comprising a setting unit that sets the first threshold value and the second threshold value . The output unit displays the image,
The inspection apparatus according to any one of claims 1 to 4, further comprising a second reception unit that receives the quality of the output result of the output unit with respect to the displayed image. The inspection apparatus according to claim 5 , further comprising a re-learning unit that re-learns the learning model based on the image and the information regarding the quality. 7. The inspection apparatus according to claim 6 , wherein the re-learning unit re-learns when the second receiving unit receives “no”. Acquire the image obtained by irradiation of electromagnetic waves to the inspection object placed on the conveyor,
Of the image, based on the brightness of each pixel, to identify the blob to determine the presence or absence of abnormality of the inspection object,
It is determined whether the determination value calculated based on the specified blob is an intermediate value between the first threshold value on the normal side and the second threshold value on the abnormal side ,
When it is determined that the value is the intermediate value, when the image including the specified blob is input, the learning model that outputs information regarding the presence or absence of abnormality of the inspection object is input the image including the specified blob. Te, before outputting the information about the presence or absence of Kikoto normal,
An anomaly detection method that causes a computer to perform processing. Acquire the image obtained by irradiation of electromagnetic waves to the inspection object placed on the conveyor,
Of the image, based on the brightness of each pixel, to identify the blob to determine the presence or absence of abnormality of the inspection object,
It is determined whether the determination value calculated based on the specified blob is an intermediate value between the first threshold value on the normal side and the second threshold value on the abnormal side ,
When it is determined that the value is the intermediate value, when the image including the specified blob is input, the learning model that outputs information regarding the presence or absence of abnormality of the inspection object is input the image including the specified blob. Te, before outputting the information about the presence or absence of Kikoto normal,
A computer program that causes a computer to perform processing. Among the images obtained by irradiating the inspection object placed on the conveyor with electromagnetic waves, the image including the blob for determining the presence or absence of abnormality of the inspection object specified based on the brightness of each pixel, and the inspection object The teacher data recorded in association with the information indicating the presence or absence of
When it is determined that the judgment value calculated based on the specified blob based on the teacher data is between the first threshold value on the normal side and the second threshold value on the abnormal side, an image including the specified blob is input. when it is, the method of generating the learning model for generating a learning model that outputs information about the presence or absence of abnormality of the inspected object. Of the image obtained by irradiation of electromagnetic waves to the inspection object placed on the conveyor, specified based on the brightness of each pixel, the judgment value calculated based on the blob to determine the presence or absence of abnormality of the inspection object, An input layer to which an image including the identified blob is input when it is determined that the first threshold value on the normal side and the second threshold value on the abnormal side are intermediate .
An output layer that outputs information regarding the presence or absence of abnormality of the inspection object,
An image including the identified blob and an intermediate layer in which parameters are learned using teacher data recorded by associating information indicating whether or not there is an abnormality in the inspection object,
A learning model that causes a computer to output, when an image including the identified blob is input to the input layer, information from the output layer regarding whether or not there is an abnormality in the inspection object through an operation by the intermediate layer.