JP6942779B2 - Inspection equipment and inspection program - Google Patents

Description

Embodiments of the present invention relate to inspection equipment and inspection programs.

In the inspection system for inspecting the luggage passing through customs, the operator is made to visually recognize the appearance image (appearance image) and the X-ray image (X-ray image) of the luggage supplied to the sorter (classification processing mechanism). The operator determines if the baggage needs to be checked and inputs the operation into the inspection system. The inspection system classifies the package into a package that requires confirmation and a package that does not require confirmation based on the operation input.

In this method, since the operator decides the sorting destination of the luggage, it takes time to decide the sorting destination of the luggage, the luggage to be confirmed is overlooked, and the luggage that does not need to be confirmed is classified for confirmation. There is.

Japanese Unexamined Patent Publication No. 2018-13939

An object to be solved by the present invention is to provide an inspection device and an inspection program capable of efficiently determining a cargo sorting destination by an operator.

The inspection device according to one embodiment includes a communication interface and a processor. The communication interface provides the address and name information acquired by character recognition processing from the image of the transported package, the package information including the X-ray image of the package, and the operation signal generated based on the operation by the operator. get. The processor newly uses the past history in which the inspection result information indicating whether or not the baggage generated based on the operation signal is a candidate for a regulated product is associated with the address / name information and the X-ray image. The inspection result information corresponding to the baggage information including the address and name information acquired from the transported baggage is extracted as an extraction result. Based on the operation signal, the processor adds a classification reason indicating the reason why the operator determines the baggage as a regulated product candidate to the inspection result information, and is set in advance extracted from the past history. Outputs information based on the number of packages to which the reason for classification is added.

FIG. 1 is an explanatory diagram for explaining a schematic configuration example of an inspection system according to an embodiment. FIG. 2 is an explanatory diagram for explaining a configuration example of the inspection system according to the embodiment. FIG. 3 is an explanatory diagram for explaining an example of registration information according to one embodiment. FIG. 4 is an explanatory diagram for explaining an example of the confirmation screen according to the embodiment. FIG. 5 is a flowchart for explaining an example of the operation of the inspection system according to the embodiment. FIG. 6 is an explanatory diagram for explaining an example of a confirmation screen according to another embodiment. FIG. 7 is an explanatory diagram for explaining an example of registration information according to another embodiment.

Hereinafter, embodiments will be described with reference to the drawings.
FIG. 1 is an explanatory diagram for explaining a schematic configuration example of the inspection system 1 according to the embodiment.

The inspection system 1 is a system for classifying the luggage 2 to be inspected. The inspection system 1 is used for finding regulated products at customs and the like. Controlled products are those whose domestic imports and exports to foreign countries are regulated. Controlled items include, for example, weapons, narcotics, animals and plants.

The luggage 2 is inspected by the inspection system 1. An Electrical Advance Data (EAD) tag 3 is attached to the luggage 2. The EAD tag 3 is printed with an identification code (EAD identification code) included in the EAD information which is the inspection result at the customs of another country. The EAD identification code is printed as, for example, a barcode. The EAD information is information including an EAD identification code, an address / name information, and information (content information) indicating the product name and weight of the content. The address and name information is, for example, information indicating the recipient (destination) of the package 2 (recipient information) and / or information indicating the sender (sender information).

In addition, a form 4 is attached to the luggage 2. Form 4 contains information such as address and name information indicating recipient information and / or sender information, and information indicating contents (content information).

As shown in FIG. 1, the baggage 2 includes a baggage 2 to which both the EAD tag 3 and the form 4 are attached, and a baggage 2 to which the form 4 is attached and the EAD tag 3 is not attached.

The inspection system 1 includes a sorter 11, a sorter control device 12, an EAD tag reader 13, a camera 14, an X-ray imaging device 15, a form reader 16, an operation terminal 17, a first storage device 18, a second storage device 19, and an inspection. The device 20 is provided. The sorter 11, the sorter control device 12, the EAD tag reader 13, the camera 14, the X-ray imaging device 15, the form reader 16, the operation terminal 17, the first storage device 18, the second storage device 19, and the inspection device 20 are networks. It is configured to be able to communicate with each other via 21.

The sorter 11 is a device that classifies the luggage 2 based on the control of the sorter control device 12. The sorter 11 classifies, for example, the baggage 2 into a first supplier 31 and a second supplier 32. The first supplier 31 is a supplier to which the luggage 2 that is not confirmed by the operator is transported. The second supplier 32 is a supplier to which the cargo 2 to be confirmed by the operator is transported. The package 2 transported to the second supplier 32 is opened and confirmed by, for example, an operator.

For example, the sorter 11 includes a first conveyor 41, a second conveyor 42, and a sorting mechanism 43.

The first conveyor 41 conveys the cargo 2 from the upstream side to the downstream side. The luggage 2 is supplied to the upstream end of the first conveyor 41 by a robot arm or the like. A first partition 31 is provided at the downstream end of the first conveyor 41. That is, the first conveyor 41 conveys the load 2 supplied to the upstream side to the first partition 31.

The second conveyor 42 conveys the cargo 2 from the upstream side to the downstream side. The upstream end of the second conveyor 42 is provided so as to be adjacent to the first conveyor 41. A second partition 32 is provided at the downstream end of the second conveyor 42. That is, the first conveyor 41 conveys the load 2 supplied to the upstream side to the second partition 32.

The sorting mechanism 43 is a mechanism for supplying the cargo 2 conveyed on the first conveyor 41 to the second conveyor 42. For example, the sorting mechanism 43 includes an arm that, by rotating, applies a force from the horizontal direction to the cargo 2 conveyed on the first conveyor 41 to move the cargo 2 to the second conveyor 42. ..

The sorter control device 12 controls the operations of the first conveyor 41 and the second conveyor 42 of the sorter 11 based on the control of the inspection device 20, and conveys the load 2. Further, the sorter control device 12 controls the operation of the sorting mechanism 43, and is a first partition 31 which is a downstream end of the first conveyor 41 and a second end which is a downstream end of the second conveyor 42. The supplier of baggage 2 is switched between the supplier 32 of 2 and the supplier 32 of 2.

The EAD tag reader 13 reads the EAD tag 3 from the cargo 2 conveyed by the first conveyor 41. The EAD tag reader 13 reads the EAD identification code from the EAD tag 3 and supplies the read EAD identification code to the inspection device 20. For example, when the EAD identification code is printed as a barcode on the EAD tag 3, the EAD tag reader 13 is configured as, for example, a barcode reader.

The camera 14 acquires an image (appearance image) of the appearance of the cargo 2 being conveyed by the first conveyor 41. The camera 14 supplies the acquired external image of the luggage 2 to the inspection device 20. For example, the camera 14 is configured by combining a lens and an image pickup device that converts the light imaged by the lens into an image.

The X-ray photographing apparatus 15 acquires an X-ray image of the cargo 2 conveyed by the first conveyor 41. The X-ray photographing apparatus 15 supplies the acquired X-ray image of the baggage 2 to the inspection apparatus 20. For example, the X-ray photographing apparatus 15 is configured by combining an X-ray tube that irradiates the luggage 2 with X-rays and an X-ray detector that detects the X-rays that have passed through the luggage 2.

The form reader 16 reads various information from the form 4 attached to the luggage 2 conveyed by the first conveyor 41. For example, the form reader 16 acquires recipient information, sender information, and content information from form 4. The recipient information, sender information, and content information obtained from Form 4 are hereinafter referred to as form information. The form reader 16 supplies the form information to the inspection device 20. For example, the form reader 16 has, for example, an image sensor in which pixels for converting light into an image are arranged in a line, and a lens for forming light on the pixels, and acquires an image (form image) of the form 4. It is equipped with a line image sensor. Further, the form reader 16 includes a processor that performs optical character recognition processing for recognizing form information such as recipient information, sender information, and content information from a form image read by a line image sensor.

Next, the operation terminal 17 will be described.
The operation terminal 17 generates a signal (operation signal) based on the operation input, and supplies the operation signal to the inspection device 20. In addition, the operation terminal 17 provides various information to the operator.

FIG. 2 shows a configuration example of the operation terminal 17 and the inspection device 20. The operation terminal 17 includes a communication interface 51, a control unit 52, and a touch panel 53.

The communication interface 51 is an interface for communicating with a device other than the operation terminal 17. The communication interface 51 communicates with the inspection device 20 via the network 21.

The control unit 52 is a processing unit that executes various processes. The control unit 52 includes a processor 54 and a memory 55.

The processor 54 is an arithmetic element that executes arithmetic processing. The processor 54 is configured as, for example, a CPU. The processor 54 performs various processes based on the program stored in the memory 55.

The memory 55 is a storage medium for storing programs and data. The memory 55 includes, for example, one or more of a ROM which is a read-only non-volatile memory, a RAM for temporarily storing data, and a storage for storing data.

The touch panel 53 is a device that displays a screen and generates an operation signal based on the operation. The touch panel 53 includes an integrally configured display 56 and a touch sensor 57. The operation terminal 17 may be configured to include a display that displays a screen instead of the touch panel 53, and an operation unit that generates an operation signal based on the operation. The operation unit may be any mouse, trackball, keyboard, trackpad, or the like.

The display 56 displays a screen based on display data (screen data) supplied from the control unit 52 or a graphic controller (not shown).

The touch sensor 57 generates an operation signal indicating the position touched by the operator who operates the operation terminal 17 on the screen displayed on the display 56.

The first storage device 18 stores registration information that is the result of processing in the inspection device 20. FIG. 3 is an explanatory diagram for explaining an example of registration information. The registration information is a past history in which various information regarding the past baggage 2 and inspection results are associated with each other. For example, the registration information includes the registration date and time when the registration information was registered, the operator identification code, the baggage information, and the inspection result information. The operator identification code is a code for identifying an operator. The cargo information is information about the cargo 2 being carried by the sorter 11. The baggage information includes at least one of the address / name information and the X-ray image. For example, the package information includes EAD information, form information, an external image of the package, and an X-ray image of the package. The inspection result information includes the classification destination information input by the operator and the presence / absence information of the regulated product.

The classification destination information is information indicating whether the package 2 is classified into the first supplier 31 or the second supplier 32. When the operator determines that it is not necessary to check the contents of the cargo 2, the operator operates the operation terminal 17 so as to classify the cargo 2 into the first supplier 31. Further, when the operator needs to confirm the contents of the luggage 2, that is, when the luggage 2 is determined to be a regulated product candidate, the operation terminal 17 is set so as to classify the luggage 2 into the second supplier 32. Manipulate. In other words, the classification destination information is information indicating whether or not the operator has determined that the luggage 2 is a regulated product candidate. The inspection device 20 generates classification destination information for each cargo 2 based on the operation signal generated as a result of the operator operating the operation terminal 17.

The regulated product presence / absence information is information indicating whether or not the regulated product is included in the baggage 2. The operator opens the package 2 classified in the second supplier 32. The operator operates the operation terminal 17 based on the result of opening the baggage 2 depending on whether or not the regulated item is included. As a result, the inspection device 20 generates information on the presence or absence of regulated products.

The second storage device 19 stores a database generated based on the registration information stored in the first storage device 18. In the database, for example, sender information of EAD information, recipient information of EAD information, sender information of form information, recipient information of form information, and the like are associated with past inspection result information. Specifically, the database is configured to be able to extract the number of times the contents of the package have been confirmed in the past and the number of times the package contains regulated items for each sender information or recipient information. In addition, the database may further have information such as the presence or absence of trouble for each sender information or each recipient information. The second storage device 19 may be configured as one storage device with the first storage device 18.

Next, the inspection device 20 will be described.
As shown in FIG. 2, the inspection device 20 includes a communication interface 61 and a control unit 62. The inspection device 20 is a device that provides information to the operator and discriminates between a baggage that does not require confirmation and a baggage that requires confirmation of the baggage based on an operation input by the operator. The operator logs in to the inspection device 20 by inputting the information for authentication (authentication information) into the inspection device 20. When the inspection device 20 generates the above registration information, the inspection device 20 adds the identification information (operator identification code) of the logged-in operator to the registration information.

The inspection device 20 is composed of a device such as a personal computer (PC) capable of processing data and storing data. The inspection device 20 sets the sorter control device 12 so as to transport the package 2 that does not require confirmation of the package to the first supplier 31 and the package 2 that requires confirmation of the package to the second supplier 32. Control. The inspection device 20 includes a communication interface 61 and a control unit 62.

The communication interface 61 is an interface for communicating with other devices. The communication interface 61 communicates with the EAD tag reader 13, the camera 14, the X-ray imaging device 15, the form reader 16, the operation terminal 17, the first storage device 18, and the second storage device 19 via the network 21. It is equipped with terminals and circuits that correspond to the communication standards of.

The control unit 62 is a processing unit that executes various processes. The control unit 62 includes a processor 63 and a memory 64.

The processor 63 is an arithmetic element that executes arithmetic processing. The processor 63 is configured as, for example, a CPU. The processor 63 performs various processes based on the inspection program stored in the memory 64.

The memory 64 is a storage medium for storing programs and data. The program is a program (inspection program) for causing the processor 63 to execute a process of inspecting the baggage 2 described later. The memory 64 includes, for example, a ROM which is a read-only non-volatile memory, and at least one or more storages for storing data. That is, the memory 64 is a non-temporary recording medium on which the inspection program is recorded. Further, the memory 64 may further include a RAM for temporarily storing data.

Further, the memory 64 stores a plurality of EAD information. For example, the memory 64 stores EAD information imported from a storage medium such as a USB memory. Further, the memory 64 may store the EAD information supplied via the network.

The processor 63 of the control unit 62 of the inspection device 20 controls the operations of the EAD tag reader 13, the camera 14, the X-ray imaging device 15, and the form reader 16 by executing the program stored in the memory 64. As a result, the processor 63 acquires the EAD identification code from the EAD tag reader 13. Further, the processor 63 acquires an external image of the luggage 2 from the camera 14. Further, the processor 63 acquires an X-ray image of the luggage 2 from the X-ray imaging device 15. Further, the processor 63 acquires the form information described in the form 4 attached to the baggage 2 from the form reader 16.

Further, the processor 63 acquires EAD information from the memory 64 based on the EAD identification code acquired from the EAD tag reader 13. The processor 63 extracts the EAD information including the EAD identification code acquired from the EAD tag reader 13 from the memory 64. As a result, the processor 63 acquires the EAD information.

Further, the processor 63 extracts the inspection history of the baggage having the same address and name information based on the newly acquired baggage information and the past history. For example, the processor 63 extracts the inspection history from the database stored in the second storage device 19 based on the address and name information as the baggage information. Further, for example, the processor 63 may be configured to extract the inspection history from the past history stored in the first storage device 18 based on the address and name information as the baggage information. Information extracted from the database of the second storage device 19 or the past history of the first storage device 18 based on the address and name information is referred to as extracted information.

The address and name information is at least one or more information of sender information of EAD information, recipient information of EAD information, sender information of form information, and recipient information of form information.

The extracted information is information for calling the operator's attention based on the number of packages judged to be candidates for regulated products. For example, the extracted information is information generated based on the number of packages determined to be candidates for regulated products. Further, for example, the extracted information may be information generated based on the number of packages that actually include the regulated product.

Specifically, the processor 63 extracts from the database the number of times the sender information matches and the contents of the baggage have been confirmed in the past, and the number of times the sender information matches and the baggage contains regulated items in the past. do. Further, the processor 63 extracts from the database the number of times the recipient information is matched and the contents of the package are confirmed in the past, the number of times the recipient information is matched and the package contains the regulated item in the past. .. Further, in the processor 63, both the sender information and the recipient information match, the number of times the contents of the package have been confirmed in the past, both the sender information and the recipient information match, and the package has a regulated item in the past. It may be configured to extract the number of times included from the database.

In addition, the processor 63 estimates the position of the regulated product from the X-ray image using the parameters generated by machine learning based on the past history. For example, the processor 63 superimposes the estimation result of the position of the regulated product on the X-ray image to generate a candidate image. For example, the processor 63 calculates a score for each predetermined region on the X-ray image based on the possibility that a regulated product exists. The processor 63 generates a candidate image by displaying each area differently according to the calculated score. For example, the processor 63 generates a candidate image by displaying each area in a different color according to the calculated score.

The parameters are adjusted based on a large amount of training data. The learning data has a question and an answer. The question and answer of the learning data is any data of the registration information. As described above, in the example of estimating the position of the regulated product from the X-ray image, the X-ray image corresponds to the problem, and the position where the regulated product exists in the X-ray image corresponds to the answer. That is, the parameters are machine-learned by, for example, a neural network for segmentation, based on learning data having an X-ray image (problem) and the position (answer) of the regulated product in the X-ray image. It was generated by.

The processor 63 generates a screen (confirmation screen) showing various information about the cargo 2 transported by the sorter 11, transmits it to the operation terminal 17, and displays it. The processor 63 generates a confirmation screen based on the EAD information, the form information, the appearance image, the X-ray image, the candidate image, and the extracted information extracted from the database.

FIG. 4 is an explanatory diagram for explaining an example of the confirmation screen 71.
The confirmation screen 71 is a screen displayed on the display 56 of the operation terminal 17. The confirmation screen 71 displays information used when the operator determines whether or not it is necessary to confirm that the package 2 has been opened. That is, the information displayed on the confirmation screen 71 is reference information used as a reference for the operator's judgment. The confirmation screen 71 has displays such as an external image 72, an X-ray image 73, a candidate image 74, EAD information 75, form information 76, extraction information 77, a delivery button 78, and a confirmation button 79.

The external image 72 is an external image of the luggage 2 acquired by the camera 14. The X-ray image 73 is an X-ray image of the luggage 2 acquired by the X-ray photographing apparatus 15. The candidate image 74 is an image generated by the processor 63 based on the X-ray image and the above parameters. The EAD information 75 is EAD information acquired based on the EAD identification code acquired from the baggage 2 by the EAD tag reader 13. If the EAD tag 3 is not attached to the baggage 2, the EAD identification code and the EAD information cannot be acquired, so that the EAD information 75 in the confirmation screen 71 is blank. The form information 76 is the form information read from the form 4 of the baggage 2 by the form reader 16. That is, the EAD information 75 and the form information 76 include address and name information as baggage information.

The extracted information 77 is information extracted from the past history by the processor 63 based on the address and name information as described above. For example, the extraction information 77 is the number of packages "YY" determined to be a regulated product candidate among the number of packages "ZZ" whose recipient information matches the package 2 transported by the sorter 11, and the regulated product. Includes information such as the number of packages that contained "XX". Further, the extraction information 77 is the number of packages "BB" determined to be a regulated product candidate among the number of packages "CC" whose recipient information matches the package 2 transported by the sorter 11, and the regulated product. Includes information such as the number of packages "AA" that contained. Furthermore, the extraction information 77 may include the presence or absence of past troubles of the sender or recipient of the cargo 2 carried by the sorter 11.

Further, the confirmation screen 71 may further include an alarm display 80. In FIG. 4, the character string “confirmation required” is displayed on the confirmation screen 71 as the alarm display 80. The processor 63 displays the alarm display 80 on the confirmation screen 71, for example, when the extracted information satisfies a predetermined condition. For example, the processor 63 causes the alarm display 80 to be displayed on the confirmation screen 71 when the number of regulated products in the extracted information is equal to or greater than a preset threshold value. Further, for example, when the number of regulated product candidates in the extracted information is equal to or greater than a preset threshold value, the processor 63 causes the alarm display 80 to be displayed on the confirmation screen 71. The above conditions may be set for each operator.

The delivery button 78 and the confirmation button 79 are buttons that can be selected by the touch sensor 57. The operator confirms the reference information on the confirmation screen 71 and determines whether or not it is necessary to confirm the opening of the package 2. That is, does the operator need to check the appearance image 72, the X-ray image 73, the candidate image 74, the EAD information 75, the form information 76, the extraction information 77, and the like on the confirmation screen 71 and open the package 2. It is determined whether or not, and either the delivery button 78 or the confirmation button 79 is selected.

When the delivery button 78 is selected by the operator, the inspection device 20 determines that the cargo 2 is not a regulated product candidate, and controls the sorter control device 12 so as to transport the cargo 2 to the first supplier 31. ..

When the confirmation button 79 is selected by the operator, the inspection device 20 determines that the cargo 2 is a regulated product candidate, and controls the sorter control device 12 so as to transport the cargo 2 to the second supplier 32. .. After that, the operator opens the baggage 2 transported to the second supplier 32, confirms the presence or absence of the regulated item, and inputs the confirmation result to the operation terminal 17. As a result, the inspection device 20 generates registration information as shown in FIG. The inspection device 20 stores the generated registration information in the first storage device 18. As a result, the past history is accumulated in the first storage device 18. In addition, as a result, a database capable of extracting the number of packages judged to be regulated product candidates in the past and the number of packages containing regulated products by using the sender information and the recipient information as an index is stored in the second storage. It can be generated in the device 19.

Next, the operation of the inspection system 1 will be described.
FIG. 5 is a flowchart for explaining the operation of the inspection system 1.

The EAD tag reader 13 of the inspection system 1 reads the EAD identification code from the EAD tag 3 of the cargo 2 conveyed by the first conveyor 41 (step S11). The EAD tag reader 13 supplies the EAD identification code to the inspection device 20.

The camera 14 of the inspection system 1 acquires an external image of the cargo 2 being conveyed by the first conveyor 41 (step S12). The camera 14 supplies an external image to the inspection device 20.

The X-ray imaging device 15 of the inspection system 1 acquires an X-ray image of the cargo 2 conveyed by the first conveyor 41 (step S13). The camera 14 supplies an X-ray image to the inspection device 20.

The form reader 16 of the inspection system 1 acquires form information from the form 4 of the luggage 2 conveyed by the first conveyor 41 (step S14). The form reader 16 supplies the form information to the inspection device 20.

The processor 63 of the control unit 62 of the inspection device 20 determines whether or not the EAD identification code has been acquired by the EAD tag reader 13 (step S15). As a result, the processor 63 determines whether or not the EAD identification code has been supplied from the EAD tag reader 13.

When the processor 63 determines that the EAD identification code has been acquired by the EAD tag reader 13 (step S15, YES), the processor 63 acquires the EAD information based on the EAD identification code (step S16). Through the above processing, the inspection device 20 acquires baggage information which is various information about the baggage 2.

Further, the processor 63 acquires the extracted information from the database of the second storage device 19 based on the EAD information (step S17), and proceeds to the process of step S19 described later. That is, the processor 63 stores the number of past packages that match the sender information and recipient information of the EAD information, the number of packages determined to be regulated product candidates, and the number of packages determined to include the regulated product, respectively. Get from.

When the processor 63 determines that the EAD identification code has not been acquired by the EAD tag reader 13 (step S15, NO), the processor 63 acquires the extracted information from the database of the second storage device 19 based on the form information (step S18). ), The process proceeds to step S19, which will be described later. That is, the processor 63 stores the number of past packages that match the sender information and the recipient information of the form information, the number of packages determined to be regulated product candidates, and the number of packages determined to include the regulated product, respectively. Get from.

The processor 63 uses the parameters generated by machine learning to estimate the position of the regulated product from the X-ray image and generate a candidate image (step S19).

The processor 63 generates the confirmation screen 71 shown in FIG. 4 based on the EAD information, the form information, the appearance image, the X-ray image, the candidate image, and the extracted information extracted from the database (step S20). The processor 63 outputs the confirmation screen 71 to the operation terminal 17 and displays it (step S21). As a result, the operator is made to visually recognize the confirmation screen 71 and determine whether or not the luggage 2 is a regulated product candidate. When the operator determines that the luggage 2 is a candidate for a regulated product, the operator selects the confirmation button 79 on the confirmation screen 71. When the operator determines that the package 2 is not a regulated product candidate, the operator selects the delivery button 78 on the confirmation screen 71. The operation terminal 17 supplies an operation signal to the inspection device 20 based on the selection results of the confirmation button 79 and the delivery button 78 on the confirmation screen 71.

The processor 63 of the inspection device 20 determines whether or not the luggage 2 is a regulated product candidate based on the operation signal from the operation terminal 17 (step S22). For example, when the processor 63 receives an operation signal indicating that the delivery button 78 has been selected from the operation terminal 17, the processor 63 determines that the package 2 is not a regulated product candidate. Further, for example, when the processor 63 receives an operation signal indicating that the confirmation button 79 has been selected from the operation terminal 17, the processor 63 determines that the luggage 2 is a regulated product candidate.

When the processor 63 determines that the cargo 2 is not a regulated product candidate (step S22, NO), the processor 63 inputs a control signal to the sorter control device 12 so that the cargo 2 is transported to the first supplier 31 (step S22, NO). Step S23). Further, when the processor 63 determines that the cargo 2 is a candidate for a regulated product (step S22, YES), the processor 63 inputs a control signal to the sorter control device 12 so that the cargo 2 is transported to the second supplier 32. (Step S24).

After that, the operator opens the baggage 2 transported to the second supplier 32, confirms the presence or absence of the regulated item, and inputs the confirmation result to the operation terminal 17. The processor 63 generates registration information based on the above processing result, stores the registration information in the first storage device 18 (step S25), and ends the processing. As a result, the presence / absence information of the regulated product indicating whether or not the regulated product is actually included in the baggage 2 is added to the registration information. The operator who opens the baggage 2 transported to the second partition 32 may be different from the operator who operates the operation terminal 17. In this case, the inspection system 1 may further include a device for the operator to open the package 2 transported to the second supplier 32 and input whether or not the regulated product is included. That is, the inspection system 1 may further include a device that transmits information indicating whether or not the regulated product is included to the inspection device 20 based on the input of the operator.

The registration information does not have to include the confirmation result of the presence or absence of regulated products. In this case, the item of the number of regulated products is also omitted from the extraction information 77 on the confirmation screen 71.

As described above, the processor 63 of the inspection device 20 acquires the cargo information regarding the cargo 2 carried by the sorter 11 and the operation signal generated by the operator operating the operation terminal 17 by the communication interface 61. do. The processor 63 generates inspection result information indicating whether or not the cargo 2 is a regulated product candidate based on the operation signal. The processor 63 outputs reference information based on the past history in which the baggage information of the past baggage 2 and the inspection result information are associated with each other and the newly acquired baggage information. As a result, the inspection device 20 can provide useful information for the operator to determine the sorting destination of the cargo. As a result, it is possible to provide an inspection device that allows the operator to efficiently determine the sorting destination of the cargo.

Further, the baggage information is the address / name information corresponding to the baggage 2, and the processor 63 generates reference information based on the number of baggage in which the address / name information matches and is determined to be a regulated product candidate. .. As a result, the inspection device 20 can provide useful information for the operator to determine the sorting destination of the cargo.

Further, when the processor 63 acquires an operation signal indicating that the cargo 2 contains the regulated product, the processor 63 adds information indicating that the cargo 2 includes the regulated product to the inspection result information. The processor 63 generates reference information based on the number of packages whose address and name information match and which includes regulated items. As a result, the inspection device 20 can provide useful information for the operator to determine the sorting destination of the cargo.

Further, the processor 63 acquires the EAD information that matches the EAD identification code read from the package 2 by the communication interface 61. The processor 63 uses at least one of the sender information and the recipient information included in the EAD information as the address and name information. As a result, the inspection device 20 can provide useful information for the operator to determine the sorting destination of the cargo.

Further, the processor 63 acquires the sender information and the recipient information displayed on the package 2 by the communication interface 61. The processor 63 uses at least one of the sender information and the recipient information as the address and name information. As a result, the inspection device 1 can leave a history even when the EAD tag 3 is not attached to the baggage 2.

The processor 63 of the inspection device 20 acquires an X-ray image of the luggage 2 from the X-ray photographing device 15. The processor 63 estimates the position of the regulated product from the X-ray image using the parameters generated by machine learning. The processor 63 superimposes the estimation result of the position of the regulated product on the X-ray image to generate a candidate image. The processor 63 adds the candidate image to the reference information and outputs it. As a result, the operator's attention can be drawn.

In the above embodiment, the confirmation screen 71 has been described as a screen that allows the operator to select either the delivery button 78 or the confirmation button 79, but the present invention is not limited to this configuration. The confirmation button 79 may be divided according to a plurality of classification reasons. Further, the operator may be configured to indicate the position where the regulated product exists on the X-ray image.

FIG. 6 is an explanatory diagram for explaining the confirmation screen 71A, which is another example of the confirmation screen 71 of FIG.

The processor 63 generates the confirmation screen 71A based on the EAD information, the form information, the appearance image, the X-ray image, the candidate image, and the extracted information extracted from the database.

The confirmation screen 71A is a screen displayed on the display 56 of the operation terminal 17. On the confirmation screen 71A, information used when the operator determines whether or not it is necessary to confirm the opening of the package 2 is displayed. That is, the information displayed on the confirmation screen 71A is reference information used as a reference for the operator's judgment. The confirmation screen 71A includes an external image 72, an X-ray image 73A, a candidate image 74, an EAD information 75, a form information 76, an extraction information 77, a delivery button 78, a regulated item button 81, a dangerous goods button 82, and a content mismatch button 83. It has a display of a past history button 84, another button 85, and the like.

The operator confirms the reference information on the confirmation screen 71A, determines whether or not it is necessary to confirm the opening of the package 2, the delivery button 78, the regulated item button 81, the dangerous goods button 82, the content mismatch button 83, and the past history. Select either Button 84 or Other Button 85. The regulated item button 81, the dangerous goods button 82, the content mismatch button 83, the past history button 84, and the other button 85 correspond to the confirmation button 79 on the confirmation screen 71. That is, when any of the regulated item button 81, the dangerous goods button 82, the content mismatch button 83, the past history button 84, and the other button 85 is selected by the operator, the inspection device 20 sorts the luggage 2 into the second assortment. The sorter control device 12 is controlled so as to be conveyed to the destination 32.

The regulated product button 81 is a button selected when the operator visually recognizes the confirmation screen 71A and determines that the luggage 2 may contain the regulated product.

The dangerous goods button 82 is a button selected when the operator visually recognizes the confirmation screen 71A and determines that the luggage 2 may contain a regulated item.

As a result of the operator visually recognizing the confirmation screen 71A, the content mismatch button 83 determines that the baggage 2 may contain items that do not match the baggage 2 described in the EAD information 75 or the form information 76. If the button is selected.

The past history button 84 is a button selected when the operator visually recognizes the confirmation screen 71A and determines from the extraction information 77 that it is necessary to open and confirm the baggage 2.

The other button 85 is a button that is selected when the operator determines that the baggage 2 needs to be opened and confirmed, although it does not correspond to the above reason.

The X-ray image 73A is an X-ray image of the luggage 2 acquired by the X-ray photographing apparatus 15. Further, the X-ray image 73A is configured so that the area on the image can be indicated by the touch sensor 57. That is, when the operator performs an operation of selecting (tapping, dragging, etc.) an area on the X-ray image 73A by the touch sensor 57, the operation terminal 17 indicates information indicating the area selected by the operator (X-ray image area information). To generate.

For example, the operator selects a region in the X-ray image 73A where the regulated product is presumed to be present. That is, the X-ray image area information is information indicating the position of the regulated product in the X-ray image 73A. As a result, it is possible to generate learning data in which the X-ray image 73A is a problem and the result of region selection (X-ray image region information) by the operator is the answer.

Based on the operation result of the button on the confirmation screen 71A, the operation terminal 17 supplies the operation signal to the inspection device 20.

When the processor 63 of the inspection device 20 receives the operation signal indicating that the delivery button 78 has been selected from the operation terminal 17, the processor 63 determines that the package 2 is not a regulated product candidate. In this case, the processor 63 inputs a control signal to the sorter control device 12 so that the cargo 2 is conveyed to the first partition 31. Further, the processor 63 generates registration information and stores it in the first storage device 18.

Further, the processor 63 receives an operation signal from the operation terminal 17 indicating that any of the regulated item button 81, the dangerous goods button 82, the content mismatch button 83, the past history button 84, and the other button 85 has been selected. In that case, it is determined that the opening confirmation of the baggage 2 is necessary. In this case, the processor 63 inputs a control signal to the sorter control device 12 so that the cargo 2 is conveyed to the second partition 32.

After that, the operator opens the package 2 transported to the second supplier 32, confirms the contents of the package 2, and inputs the confirmation result to the operation terminal 17. The processor 63 generates registration information based on the above processing result. As a result, registration information including the regulated product presence / absence information indicating whether or not the regulated product was actually included in the baggage 2 is generated in the inspection result information.

Further, in the processor 63, buttons such as the regulated item button 81, the dangerous goods button 82, the content mismatch button 83, the past history button 84, and the other button 85 are selected, and the luggage 2 is classified into the second supplier 32. Information indicating whether or not the processing has been performed (classification reason information) is added to the registration information. For example, when the regulated product button 81 is selected, the processor 63 adds "regulated product" to the registration information as the classification reason information. For example, when the dangerous goods button 82 is selected, the processor 63 adds "dangerous goods" to the registration information as the classification reason information. For example, when the content mismatch button 83 is selected, the processor 63 adds "content mismatch" to the registration information as the classification reason information. For example, when the past history button 84 is selected, the processor 63 adds "past history" to the registration information as the classification reason information. For example, when the other button 85 is selected, the processor 63 adds "other" as the classification reason information to the registration information.

Further, the processor 63 adds X-ray image area information indicating the position of the regulated product on the X-ray image 73A to the registration information.

As a result, the registration information shown in FIG. 7 is generated. The registration information of FIG. 7 includes the registration date and time when the registration information was registered, the operator identification code, the baggage information, and the inspection result information. The cargo information is information about the cargo 2 being carried by the sorter 11. The baggage information includes at least one of the address / name information and the X-ray image. For example, the package information includes EAD information, form information, an external image of the package, and an X-ray image of the package. The inspection result information includes classification destination information, classification reason information, restricted product presence / absence information, and X-ray image area information input by the operator.

The processor 63 stores the registration information generated by the above process in the first storage device 18 as a past history. The processor 63 generates a database based on the past history stored in the first storage device 18, and stores the database in the second storage device 19. For example, the processor 63 generates a database by associating the classification destination information, the classification reason information, and the presence / absence of regulated products for each address / name information. As a result, the processor 63 can extract the number of past packages determined to be regulated product candidates and the number of past packages including regulated products for each classification reason based on the address and name information. A database can be generated.

According to such a configuration, the processor 63 can generate reference information based on the number of packages to which a preset classification reason is added, which is extracted from the past history. That is, the processor 63 is classified according to the number of past packages for the preset classification reason, the number of past packages determined to be regulated product candidates by the preset classification reason, and the preset classification reason. Moreover, the number of packages containing regulated items can be extracted from the database. Further, the processor 63 can add the extracted information to the confirmation screen 71A as the extracted information 77.

Further, in the above embodiment, it has been described that the processor 63 acquires the extracted information from the database based on the address and name information and displays the alarm based on the extracted information, but the present invention is not limited to this configuration. The processor 63 may be configured not to display an alarm based on the classification reason information.

For example, in the above embodiment, the processor 63 extracts the extracted information from the second storage device 19 based on the EAD information or the form information. Further, it has been explained that the processor 63 displays the alarm display 80 on the confirmation screen 71 when the number of regulated product candidates in the extracted information is equal to or more than a preset threshold value. However, if the reason for classification is the past history, once it is determined that the baggage 2 is a candidate for a regulated item, the alarm display will continue to be displayed for the baggage 2 having the same address and name information. Therefore, the processor 63 may be configured not to be counted as a regulated product candidate due to the reason for classification. As a result, the processor 63 can count the number of packages classified as regulated product candidates for a preset classification reason among the number of regulated product candidates in the extracted information whose address and name information match.

Further, as described above, the registered information in FIG. 7 has an X-ray image (problem) and X-ray image area information (answer) indicating the position of the regulated product on the X-ray image. That is, learning data can be generated from the registered information. The inspection device 20 can update the parameters by performing machine learning using the learning data, for example, by using a neural network for segmentation or the like. That is, the parameters can be updated based on the X-ray image area information input by the operator.

Further, the inspection device 20 may have a configuration in which the weight of updating parameters in machine learning using learning data is set for each operator. The weight is information indicating the degree of parameter update, the upper and lower limits, the coefficient, and the like. The larger the weight, the larger the parameter changes in one learning. For example, the processor 63 acquires the weight based on the operator identification code of the registration information. For example, the processor 63 acquires the weight by referring to the operator identification code added to the registration information and the weight or a table associated with the weight. The processor 63 updates the parameters based on the acquired weights and the training data. As a result, the learning data based on the registered information of the operator having high detection accuracy of the regulated product in the X-ray image can be emphasized when updating the parameters.

Further, the inspection device 20 may have a configuration in which the weight of updating the parameter using the registration information of FIG. 7 is set for each classification reason information. As a result, when the reason for classification is "Other", "Content mismatch", or "Past history", the learning data generated based on the registration information is not emphasized when updating the parameters. Can be.

Further, the inspection device 20 may have a configuration in which the registration information of FIG. 7 is not used for updating the parameters based on the result of actually checking the baggage 2 classified into the second supplier 32. For example, as a result of actually checking the baggage 2 classified into the second supplier 32, if the regulated item is not included, the X-ray image (problem) and the X-ray image area information (answer) of the registration information are displayed. It is not preferable to use it for updating parameters. Therefore, for example, the inspection device 20 is configured so that the registration information that does not include the regulated product is not used for updating the parameter as a result of actually checking the baggage 2 classified into the second supplier 32. It may have been done.

As described above, the processor 63 acquires the X-ray image of the luggage 2 as the luggage information and outputs the confirmation screen 71A as the reference information including the X-ray image 73. The operator confirms the X-ray image 73, and inputs to the operation terminal 17 whether or not the luggage 2 is a regulated product candidate. The processor 63 generates inspection result information indicating whether or not the operator has determined that the cargo 2 is a regulated product candidate based on the operation signal. Further, the processor 63 uses the X-ray image as a problem and generates parameters by machine learning based on the learning data in which the inspection result information is the answer. According to such a configuration, the processor 63 estimates whether or not the luggage 2 is a regulated product candidate based on the newly acquired X-ray image and the parameters. The processor 63 can add the estimation result of whether or not the luggage 2 is a regulated product candidate to the confirmation screen 71A as reference information. As a result, the inspection device 20 can call attention to the operator.

Further, the processor 63 acquires the X-ray image of the luggage 2 as the luggage information and outputs the confirmation screen 71A as the reference information including the X-ray image 73. The operator confirms the X-ray image 73 and determines whether or not the luggage 2 is a regulated product candidate. When the operator determines that the luggage 2 is a regulated product candidate, the operator inputs to the operation terminal 17 the reason why the luggage 2 is presumed to be the regulated product candidate (that is, the reason for classification). The processor 63 is an inspection indicating whether or not the operator has determined the cargo 2 as a regulated product candidate based on the operation signal, and a classification reason indicating the reason why the operator has determined the cargo 2 as a regulated product candidate. Generate result information. Further, the processor 63 updates the parameters by machine learning based on the learning data in which the X-ray image is a problem and the test result information is the answer. That is, the processor 63 has a problem with the X-ray image, and generates a parameter in which the classification reason indicating the reason why the operator determines that the cargo 2 is a regulated product candidate is set as the classification destination. According to such a configuration, the processor 63 determines whether or not the luggage 2 is a regulated product candidate and the reason why the luggage 2 is a regulated product candidate based on the newly acquired X-ray image and the parameters. Can be estimated. As a result, the processor 63 can add the estimation results of whether or not the luggage 2 is a regulated product candidate and the reason why the luggage 2 is a regulated product candidate to the confirmation screen 71A as reference information. As a result, the inspection device 20 can call attention to the operator.

Further, the processor 63 acquires the X-ray image of the luggage 2 as the luggage information and outputs the confirmation screen 71A as the reference information including the X-ray image 73. The operator confirms the X-ray image 73 and determines whether or not the luggage 2 is a regulated product candidate. When the operator determines that the luggage 2 is a regulated product candidate, the operator inputs to the operation terminal 17 the reason why the luggage 2 is presumed to be the regulated product candidate (that is, the reason for classification). Further, when the operator determines that the luggage 2 is a candidate for the regulated product, the operator inputs the position of the regulated product to the operation terminal 17 on the X-ray image 73A. Based on the operation signal, the processor 63 indicates whether or not the operator has determined that the luggage 2 is a regulated product candidate, the classification reason indicating the reason why the operator has determined that the luggage 2 is a regulated product candidate, and the regulated product. X-ray image area information indicating the position and inspection result information indicating the position are generated. Further, the processor 63 updates the parameters by machine learning based on the learning data in which the X-ray image is a problem and the test result information is the answer. That is, the processor 63 considers the X-ray image as a problem, and estimates the reason for classification indicating the reason why the operator determines that the luggage 2 is a candidate for the regulated product, and the position where the regulated product exists on the X-ray image. Generate parameters for. In other words, the processor 63 has X-ray image area information generated based on the operator's operation on the X-ray image 73A, and a classification reason indicating the reason why the operator determines that the luggage 2 is a regulated product candidate. Generate parameters based on the learning data that is the answer. According to such a configuration, the processor 63 determines whether or not the luggage 2 is a regulated product candidate, the reason why the luggage 2 is a regulated product candidate, and X based on the newly acquired X-ray image and the parameters. It is possible to estimate the region where the regulated product on the line image is estimated to exist. Based on the parameters and the newly acquired X-ray image, the processor 63 generates a candidate image showing the region where the regulated product on the X-ray image is presumed to exist and the reason for classification. The processor 63 displays the generated candidate image on the confirmation screen 71A as reference information. As a result, the inspection device 20 can call attention to the operator.

Further, the processor 63 has a problem with an X-ray image, and when the parameters are updated by machine learning using the learning data obtained by answering the position of the regulated item on the X-ray image, the X-ray used for machine learning is used. You may crop the image. For example, the processor 63 crops an X-ray image based on the X-ray image area information, and uses the cropped X-ray image as a problem of training data. That is, the processor 63 uses the X-ray image of the region selected by the operator on the X-ray image as a problem of learning data. As a result, the inspection device 20 can generate parameters that make it easier to detect regulated product candidates because unnecessary information is removed from the problem of learning data.

Further, in the above embodiment, "regulated product" and "dangerous goods" and "dangerous goods" are selected based on the selection of buttons such as the regulated product button 81, the dangerous goods button 82, the content mismatch button 83, the past history button 84, and the other button 85. , "Content mismatch", "Past history", and "Other" are added to the inspection result information, but the configuration is not limited to this. The "regulated product" and the "dangerous goods" may be further divided into various items. For example, the "regulated product" may be divided into a "specific living body", a "designated drug", a "gun and gun parts", a "counterfeit product", and the like. Further, the "dangerous goods" may be divided into "explosives", "explosives", "specific chemical substances" and the like. Since the classification reason information is divided in this way, detailed classification reason information is added to the inspection result information. As a result, the inspection device 20 can add a more detailed reason for classification to the reference information.

Further, the execution of the above inspection program is not limited to the processor 63 of the inspection device 20, and any device may be used as long as it includes a communication interface and a processor. That is, the inspection program is an inspection program executed by the processor of the device including the communication interface and the processor, and the baggage information regarding the baggage carried by the sorter that sorts the baggage by the communication interface to the processor. It is executed to acquire the operation signal generated based on the operation by the operator. In addition, the inspection program has a past history in which the processor associates inspection result information indicating whether or not the package generated based on the operation signal is a regulated product candidate with the package information, and the newly acquired package. To execute the output of reference information based on the information.

Further, as described above, the inspection program may be recorded on a non-temporary recording medium in a state of being readable by a device including a processor that executes the inspection program.

Although some embodiments of the present invention have been described, these embodiments are presented as examples and are not intended to limit the scope of the invention. These novel embodiments can be implemented in various other embodiments, and various omissions, replacements, and changes can be made without departing from the gist of the invention. These embodiments and modifications thereof are included in the scope and gist of the invention, and are also included in the scope of the invention described in the claims and the equivalent scope thereof.
The scope of claims at the time of filing is added below.
[C1]
A communication interface that acquires baggage information related to the baggage carried by the sorter that classifies the baggage and an operation signal generated based on an operation by the operator.
Based on the past history in which the inspection result information indicating whether or not the baggage generated based on the operation signal is a candidate for a regulated product and the baggage information are associated with each other, and the newly acquired baggage information. And a processor that outputs reference information
Inspection device equipped with.
[C2]
The baggage information is address and name information corresponding to the baggage.
The inspection device according to C1, wherein the processor generates the reference information based on the number of packages that are candidates for regulated products extracted from the past history in which the address and name information match.
[C3]
The baggage information includes address and name information corresponding to the baggage.
The processor
When the operation signal indicating that the baggage contains a regulated item is received, information indicating that the baggage contains a regulated item is added to the inspection result information.
The inspection device according to C1 that generates the reference information based on the number of packages including regulated products extracted from the past history with which the address and name information match.
[C4]
The baggage information includes address and name information corresponding to the baggage.
The processor
Based on the operation signal, a classification reason indicating the reason why the operator determines that the baggage is a regulated product candidate is added to the inspection result information.
The inspection device according to C1 that generates the reference information based on the number of packages to which a preset classification reason is added, which is extracted from the past history.
[C5]
The baggage information includes an X-ray image of the baggage.
The processor
Whether or not the operator has confirmed the X-ray image and determined that the luggage is a regulated product candidate based on the operation signal generated based on the input of whether or not the luggage is a regulated product candidate. Generates the above-mentioned inspection result information indicating
Described in C1 for generating the reference information based on the parameters generated based on the learning data in which the X-ray image is a problem and the test result information is the answer, and the newly acquired X-ray image. Inspection equipment.
[C6]
The processor
Based on the operation signal, a classification reason indicating the reason why the operator determines that the baggage is a regulated product candidate is added to the inspection result information.
The inspection device according to C5, which adds to the reference information the reason why the baggage is presumed to be a regulated product candidate based on the parameters and the X-ray image.
[C7]
The processor
The X-ray image is a problem, and the parameter generated based on the learning data obtained by answering the position of the regulated product on the X-ray image and the X-ray image on the X-ray image. Generate a candidate image showing the area where the regulated product is presumed to exist,
The inspection device according to C5, which adds the candidate image to the reference information.
[C8]
The inspection device according to C7, wherein the learning data is data in which X-ray image area information indicating an area on the X-ray image is generated as the answer based on an operation of the operator on the X-ray image.
[C9]
The learning data includes X-ray image area information indicating an area on the X-ray image based on the operator's operation on the X-ray image, and the reason why the operator determines that the baggage is a regulated product candidate. The inspection device according to C7, wherein the classification reason shown is the data generated as the answer.
[C10]
The processor
The inspection according to C9, which generates the candidate image showing the region where the regulated product on the X-ray image is presumed to exist based on the parameter and the X-ray image, and the reason for the classification. Device.
[C11]
The inspection device according to C5, wherein the processor updates the parameters based on the learning data.
[C12]
The processor adds an operator identification code for identifying the operator to the inspection result information.
The inspection device according to C11, wherein the processor updates the parameters with different weights for each operator who generated the learning data.
[C13]
The communication interface acquires address and name information from the sender information and the recipient information displayed on the baggage.
The inspection device according to C1, wherein the baggage information includes at least one of the sender information and the recipient information of the address and name information.
[C14]
The communication interface acquires EAD information matching the EAD identification code read from the baggage from a storage medium in which a plurality of EAD information including sender information and recipient information is stored in advance, and includes the acquired EAD information. Get the address and name information
The inspection device according to C1, wherein the baggage information includes at least one of the sender information and the recipient information of the address and name information.
[C15]
An inspection program executed by an inspection device including a communication interface and a processor.
In the inspection device
By the communication interface, it is executed to acquire the cargo information about the cargo carried by the sorter for classifying the cargo and the operation signal generated based on the operation by the operator.
The past history in which the inspection result information indicating whether or not the baggage generated based on the operation signal by the processor is a regulated product candidate and the baggage information are associated with the baggage information, and the newly acquired baggage information And, based on, to execute to output reference information,
Inspection program.

1 ... Inspection system, 2 ... Luggage, 3 ... EAD tag, 4 ... Form, 11 ... Sorter, 12 ... Sorter control device, 13 ... EAD tag reader, 14 ... Camera, 15 ... X-ray imaging device, 16 ... Form reader, 17 ... operation terminal, 18 ... first storage device, 19 ... second storage device, 20 ... inspection device, 21 ... network, 31 ... first supplier, 32 ... second supplier, 41 ... first Conveyor, 42 ... second conveyor, 43 ... classification mechanism, 51 ... communication interface, 52 ... control unit, 53 ... touch panel, 54 ... processor, 55 ... memory, 56 ... display, 57 ... touch sensor, 61 ... communication interface, 62 ... Control unit, 63 ... Processor, 64 ... Memory.

Claims (15)

A communication interface that acquires address and name information acquired by character recognition processing from an image of a transported package, package information including an X-ray image of the package, and an operation signal generated based on an operation by an operator. When,
The address acquired from the newly transported baggage from the past history in which the inspection result information indicating whether or not the baggage generated based on the operation signal is a candidate for a regulated product and the baggage information are associated with each other. A processor that extracts the inspection result information corresponding to the baggage information including the name information as an extraction result, and
Equipped with
The processor
Based on the operation signal, a classification reason indicating the reason why the operator determines that the baggage is a regulated product candidate is added to the inspection result information.
Outputs information based on the number of packages to which a preset classification reason is added, which is extracted from the past history.
Inspection equipment. The inspection device according to claim 1, wherein the processor outputs the inspection result information included in the extraction result to a confirmation screen. The inspection device according to claim 1 or 2, wherein the processor outputs an X-ray image of the newly transported baggage to a confirmation screen. The inspection device according to any one of claims 1 to 3, wherein the processor outputs an image of the newly transported package as an image to a confirmation screen. The inspection device according to claim 1, wherein the processor outputs information based on the number of packages that are candidates for regulated products extracted from the past history in which the address and name information match. The processor
When the operation signal indicating that the baggage contains a regulated item is received, information indicating that the baggage contains a regulated item is added to the inspection result information.
The inspection device according to claim 1, which outputs information based on the number of packages including regulated items extracted from the past history in which the address and name information match. The processor
With the X-ray image as a problem, the regulation on the X-ray image is based on the parameters generated based on the learning data obtained by answering the position of the regulated product on the X-ray image and the X-ray image. Generate a candidate image showing the area where the product is presumed to exist,
The inspection device according to claim 1, which outputs the candidate image. The inspection according to claim 7 , wherein the training data is data in which X-ray image area information indicating an area on the X-ray image is generated as the answer based on an operation of the operator on the X-ray image. Device. The learning data includes X-ray image area information indicating an area on the X-ray image based on the operator's operation on the X-ray image, and the reason why the operator determines that the baggage is a regulated product candidate. The inspection device according to claim 7 , wherein the classification reason shown is the data generated as the answer. The processor
The ninth aspect of claim 9 is to generate the candidate image showing the region where the regulated product on the X-ray image is presumed to exist and the reason for the classification based on the parameter and the X-ray image. Inspection equipment. The inspection device according to claim 7 , wherein the processor updates the parameters based on the learning data. The processor adds an operator identification code for identifying the operator to the inspection result information.
The inspection device according to claim 11 , wherein the processor updates the parameters with different weights for each operator who generated the learning data. The communication interface acquires the sender information and the recipient information displayed on the baggage, and obtains the sender information and the recipient information.
The inspection device according to claim 1, wherein the baggage information includes at least one of the sender information and the recipient information as the address and name information. The communication interface acquires EAD information that matches the EAD identification code read from the baggage from a storage medium in which a plurality of EAD information including sender information and recipient information is stored in advance.
The inspection device according to claim 1, wherein the baggage information includes at least one of the sender information and the recipient information as the address and name information. An inspection program executed by an inspection device including a communication interface and a processor.
In the inspection device
The address and name information acquired by character recognition processing from the image of the transported package by the communication interface, the package information including the X-ray image of the package, the operation signal generated based on the operation by the operator, and the operation signal. To get the execution done,
From the past history in which the inspection result information indicating whether or not the baggage generated based on the operation signal by the processor is a candidate for a regulated product, the address and name information, and the X-ray image are associated with each other. The inspection result information corresponding to the baggage information including the address and name information acquired from the baggage carried to is extracted as an extraction result .
Based on the operation signal, the operator is made to execute the addition of the classification reason indicating the reason for determining the baggage as a regulated product candidate to the inspection result information.
It is executed to output the information based on the number of packages to which the preset classification reason is added, which is extracted from the past history.
Inspection program.

Images