WO2023203993A1

WO2023203993A1 - Program and server for pest insect inspection

Info

Publication number: WO2023203993A1
Application number: PCT/JP2023/013535
Authority: WO
Inventors: 敦岡田; 忠谷元; 英男白井; 勇太笠原
Original assignee: 日本農薬株式会社; 株式会社アグリマート
Priority date: 2022-04-22
Filing date: 2023-03-31
Publication date: 2023-10-26
Also published as: JP2023160299A

Abstract

This program for pest insect inspection causes a computer to: receive an operation by a user to register an execution range of identification in an inspection model that is configured to identify at least the type of pest insects stuck to an insect-trapping sheet on the basis of an image of the insect-trapping sheet and count the identified pest insects for each type; receive an operation by the user to register information on a section which is of sections set in advance to the insect-trapping sheet and corresponds to the execution range of the identification; input the image of the insect-trapping sheet to the inspection model to cause the inspection model to execute the identification and the count; calculate, on the basis of the information on the section, a second count result, which has the section as a unit, from a first count result in the execution range of the identification by the inspection model; and present at least the second count result to the user.

Description

Programs and servers for pest inspection

The present disclosure relates to a program and server for pest inspection.

Food factories, restaurants, etc. are required to inspect for pests as part of hygiene management. In pest inspection, pests that have stuck to sticky traps or the like that have been installed in advance at predetermined locations such as food factories are identified and counted. Countermeasures are taken depending on the results of identification and counting of these pests. Here, the work of identifying and counting pests and creating a report including the results is often performed by a PCO (Pest Control Operator), which is an external company commissioned by the user. Conventionally, PCO visually identifies and counts pests stuck to sticky traps and the like. In addition, PCO summarizes the results of pest identification and counting in a report and submits it to business partners. Such visual identification and counting of pests requires a great deal of effort. Attempts have also been made to identify and count pests using artificial intelligence (AI). However, identification and counting of pests using such AI is performed for the range shown in the image. Therefore, in order to create a report, it may be necessary to repeatedly perform the same work, which may not necessarily lead to a reduction in the PCO's workload.

An object of the present disclosure is to provide a program and a server for pest inspection that reduce the burden of pest inspection.

In one embodiment, a pest inspection program includes an inspection model configured to identify at least the types of pests stuck to the insect paper based on an image of the insect paper, and to count each type of pest identified. Accepting a user's registration operation of the identification implementation range in , and accepting a user's registration operation of information on a section corresponding to the identification implementation range among the sections set in advance on the insect trapping paper; The image of the insect trapping paper is input into the inspection model to carry out identification and counting, and based on the section information, the section is determined as a unit from the first counting result which is the result of counting in the identification implementation range by the inspection model. The computer is caused to calculate a second counting result that is a result of the counting, and to present at least the second counting result to the user.

According to the present disclosure, a program and a server for pest inspection are provided that reduce the burden associated with pest inspection.

FIG. 1 is a diagram showing the configuration of a system according to an embodiment. FIG. 2 is a block diagram showing an example of the hardware configuration of the terminal. FIG. 3 is a block diagram showing an example of the hardware configuration of the server. FIG. 4 is a flowchart showing processing after the pest inspection program is started. FIG. 5 is a flowchart showing image registration processing. FIG. 6 is a diagram showing an example of an image registration screen. FIG. 7 is a flowchart showing the section registration process. FIG. 8 is a diagram showing an example of a section registration screen. FIG. 9 is a flowchart showing the analysis process. FIG. 10 is a diagram showing an example of an analysis result confirmation screen. FIG. 11 is a flowchart showing the confirmation/correction process. FIG. 12A is a diagram showing an example of a confirmation/correction screen. FIG. 12B is a diagram showing an example of a confirmation/correction screen. FIG. 13 is a flowchart showing the identification/count correction process. FIG. 14 is a diagram showing an example of an identification/count correction screen.

Hereinafter, embodiments will be described with reference to the drawings. FIG. 1 is a diagram showing the configuration of a system according to an embodiment. The system 1 includes

terminals

2a and 2b and a server 3. The

terminals

2a and 2b and the server 3 are connected via a network NW. The

terminals

2a and 2b can be connected to the network NW by, for example, wireless communication.

The terminal 2a is a terminal device that can be carried by the user, such as a smartphone or a tablet terminal. Terminal 2a was carried by a user specializing in pest inspection, such as a PCO, to places where insect trap IT was installed, such as factories and restaurants, and took an image of the insect trap paper attached to the insect trap IT. It is used for processing to inspect pests, such as identifying and counting pests stuck to the insect trap IT based on images. Here, the terminal 2a may be a terminal device such as a personal computer that cannot be carried by the user. In this case, the terminal 2a receives an image of the insect trapping paper taken with a smartphone or the like and performs the inspection process. Furthermore, the terminal 2b is a terminal that can be operated by, for example, a person in charge at a business partner of the user of the terminal 2a who has requested the user of the terminal 2a to inspect for pests, and the results of the inspection for pests at the terminal 2a are sent in the form of a report. Used when receiving. The terminal 2b may be a terminal device that the user can carry, such as a smartphone or a tablet terminal, or a terminal device that cannot be carried by the user, such as a personal computer. In FIG. 1, there is one terminal 2a and one terminal 2b. There may be two or

more terminals

2a and 2b.

FIG. 2 is a block diagram showing an example of the hardware configuration of the

terminals

2a and 2b. Here, in order to simplify the explanation, it will be assumed that the terminal 2a and the terminal 2b have the same configuration. The

terminals

2a and 2b include a processor 21, a ROM 22, a RAM 23, a storage 24, an input interface 25, a display 26, a camera 27, and a communication module 28. Note that the

terminals

2a and 2b do not necessarily have to have the same configuration.

The processor 21 is a processor configured to control the operations of the

terminals

2a and 2b. The processor 21 executes various processes according to various programs stored in the storage 24. The processor 21 is, for example, a CPU. The processor 21 may be an MPU, a GPU, or the like instead of a CPU. Further, the processor 21 does not need to be composed of one CPU or the like, and may be composed of a plurality of CPUs or the like.

The ROM 22 is, for example, a nonvolatile memory, and stores startup programs for the

terminals

2a and 2b. The RAM 23 is, for example, a volatile memory. The RAM 23 is used, for example, as a working memory during processing in the processor 21.

The storage 24 is, for example, a flash memory. The storage 24 stores various programs and data used in the

terminals

2a and 2b. In the embodiment, the storage 24 stores an OS (operating system) 241. The storage 24 may store programs and data other than the OS 241. For example, the storage 24 may store a browser application program or the like. The OS 241 is a program for realizing the basic functions of the

terminals

2a and 2b. Various programs stored in the storage 24 are executed under the control of the OS.

The input interface 25 includes an input device such as a touch panel. When the input interface 25 is operated, a signal corresponding to the content of the operation is input to the processor 21. The processor 21 performs various processes in response to this signal.

The display 26 is a display device such as a liquid crystal display or an organic EL display. The display 26 displays various images.

The camera 27 is a camera included in at least the terminal 2a, and captures an image of a subject. In the embodiment, the object is, for example, insect paper attached to the insect trap IT. The camera 27 includes an image sensor such as a CCD image sensor or a CMOS image sensor, and a lens. The configuration of camera 27 is not limited to a specific configuration. Note that the terminal 2b does not need to have the camera 27.

The communication module 28 is a module that includes an interface configured to perform processing when the

terminals

2a and 2b communicate with the server 3. The communication module 28 is configured to connect to the network NW using a mobile phone line, a wireless LAN line, a wired LAN line, or the like.

FIG. 3 is a block diagram showing an example of the hardware configuration of the server 3. The server 3 includes a processor 31, a ROM 32, a RAM 33, a storage 34, an input interface 35, a display 36, and a communication module 37. Here, the server 3 does not have to be a single server. Furthermore, the server 3 may be configured as a cloud server, for example.

The processor 31 is a processor configured to control the operation of the server 3. The processor 31 is, for example, a CPU. The processor 31 may be an MPU, a GPU, or the like instead of a CPU. Further, the processor 31 does not need to be composed of one CPU or the like, and may be composed of a plurality of CPUs or the like.

The ROM 32 is, for example, a nonvolatile memory, and stores a startup program for the server 3 and the like. The RAM 33 is, for example, a volatile memory. The RAM 33 is used, for example, as a working memory during processing in the processor 21.

The storage 34 is, for example, a hard disk drive or a solid state drive. The storage 34 stores an OS 341, a pest inspection program 342, an inspection model 343, customer information 344, user information 345, photographed images 346, and a report 347. The storage 34 may store programs and data other than the OS 341, pest inspection program 342, inspection model 343, customer information 344, user information 345, photographed image 346, and report 347.

The OS 341 is a program for realizing the basic functions of the server 3. Various programs stored in the storage 34 are executed under the control of the OS.

The pest inspection program 342 is a program for causing a computer to execute a series of processes for inspecting pests in accordance with a request from the user of the terminal 2a. The pest inspection program 342 may be a program executed by the processor 31 of the server 3. Alternatively, the pest inspection program 342 may be a program that is downloaded to the terminal 2a as a web application and executed on the web browser of the terminal 2a. Furthermore, the pest inspection program 342 may be a program installed on the terminal 2a and executed on the terminal 2a. Pest inspection program 342 will be described in detail later.

The inspection model 343 is a trained AI model that performs analysis for inspection and is equipped with a discriminator that identifies the pest in the selected image. The inspection model 343 predicts what type the pest in the input image will be classified into, and analyzes the input image based on the result with the highest predicted probability. This analysis includes pest species identification and enumeration. The test model 343 is configured by, for example, a convolutional neural network (CNN). Images of various pests, such as images of insect paper with pests stuck to them, are input to the inspection model 343 as training data along with labels indicating the type of each pest, thereby learning feature quantities for each pest. It will be done. The inspection model 343 may be configured to perform learning and analysis including feature quantities other than feature quantities obtained from images. For example, the feature amounts other than the feature amounts obtained from the image may include feature amounts such as the location where the image was taken and the time when the image was taken.

The business partner information 344 is information about a business partner of the user of the terminal 2a, such as a PCO. The business partner information may include, for example, information such as the business partner's company name or person's name and address. The company name may be recorded in multiple hierarchies such as company name and branch name. Similarly, addresses may also be recorded in multiple hierarchies.

The user information 345 is information about the user of the terminal 2a. The user's information may include information such as the user's name, email address, address, and login password.

The photographed image 346 is image data uploaded by the user of the terminal 2a. The captured image 346 is an image that can be analyzed using the inspection model 343. As will be explained in detail later, the captured image 346 is associated with insect trap information and section information. The information on the insect trap is information on the insect trap that is the subject of the photographed image, and may include information such as the installation location of the insect trap, the survey method, and the date of trapping. Further, the section information is information on the number of sections set on the insect trapping paper. When humans visually identify and count pests stuck to insect paper, they may divide the insect paper into multiple sections and identify and count the pests in each section, or identify and count pests in one section. By multiplying the result according to the number of sections, the total counting result of the insect paper may be obtained. In embodiments, the information on the plots is used to calculate pest identification and counting results on a plot-by-plot basis. The sections may be divided by the user or by a frame line printed on the insect trapping paper in advance. Here, the photographed image 346 may include other images, such as an image used for learning the inspection model 343, in addition to the image uploaded by the user of the terminal 2a.

The report 347 is data of a report created based on the test results including the results of the analysis by the test model 343. The report may be prepared according to a format requested by the business partner. For this purpose, a report template may be prepared for each business partner, and data of such a report template may also be included in the report 347.

The input interface 35 includes input devices such as a keyboard, a mouse, and a touch panel. When the input interface 35 is operated, a signal corresponding to the content of the operation is input to the processor 31. The processor 31 performs various processes in response to this signal.

The display 36 is a display device such as a liquid crystal display or an organic EL display. The display 36 displays various images.

The communication module 37 is a module that includes an interface configured to perform processing when the server 3 communicates with the

terminals

2a and 2b. The communication module 37 is configured to connect to the network NW using a mobile phone line, a wired LAN line, a wireless LAN line, or the like.

Next, the operation of the system 1 will be explained. FIG. 4 is a flowchart showing processing after the pest inspection program 342 is activated. Here, the pest inspection program 342 will be explained as being downloaded from the server 3 to the terminal 2a as a web application and executed on the web browser of the terminal 2a. In this case, the processor 21 executes the process according to the pest inspection program 342. In order to execute the pest inspection program 342, authentication using an ID and password may be performed. Further, as described above, the pest inspection program 342 may be executed on the server 3. In this case, the processing described below is mainly executed by the processor 31 of the server 3. By executing such a pest inspection program 342, the

processor

21 or 31 can operate as a first reception section, a second reception section, an inspection section, a calculation section, and a presentation section.

The process of step S1 is started after the user selects a business partner. Selection of a business partner can be performed, for example, by selecting from a list of business partners displayed based on the business partner information 344.

In step S1, the processor 21 executes period registration processing. The period registration process is a process of registering the date on which the insect trap to be inspected was caught using the insect trap, based on the input from the user. Here, the capture implementation date is, for example, a period from the date when the user installs insect paper in the insect trap and starts capturing pests to the date when the user collects the insect paper from the insect trap and finishes capturing pests. In the period registration process, the processor 21 displays a period registration screen on the screen of the pest inspection program 342 of the web browser, for example. The user registers, for example, the capture start date and capture end date on the period registration screen. In response to this, the processor 21 registers the capture start date and capture end date. After the capture date is registered, the process moves to step S2.

In step S2, the processor 21 executes insect trap registration processing. The insect trap registration process is a process of registering information regarding the insect trap to be inspected based on input from the user. In the insect trap registration process, the processor 21 displays an insect trap registration screen on the screen of the pest inspection program 342 of the web browser, for example. The user registers information regarding the insect trap on the insect trap registration screen. After registering the information regarding the insect trap, the process moves to step S3. Here, the information regarding the insect trap may include, for example, information such as a management number, installation location, and investigation method.

The management number is a number for managing the image of the insect paper. A control number can be assigned by the user for each insect trapping paper. Some insect traps can have multiple insect traps attached to them. In order to distinguish between images obtained for each of these plurality of insect-trapping papers, a management number can be assigned to each insect-trapping paper.

The installation location is the location where the insect trap is installed. The installation location is registered as information that can specify the location where the insect trap is installed in food factories, restaurants, etc. The installation location may be arbitrarily registered by the user of the terminal 2a, or may be registered by the user of the terminal 2a selecting it from among a plurality of installation location candidates registered in advance. However, it is desirable that the installation location be input in a format that can be recognized by both the user of the terminal 2a and the user of the terminal 2b. Therefore, it is desirable that the installation locations be registered in multiple layers to reduce variations in the registered contents for each user.

The survey method refers to the method of survey using installed insect traps. For example, if the installed insect trap uses a light trap, "light trap" is entered in the survey method. In addition, in the embodiment, the method of investigating the insect trap is not limited to a specific method.

In step S3, the processor 21 performs image registration processing. The image registration process is a process for registering an analysis target image to be analyzed by the inspection model 343. Here, in the image registration process in the embodiment, the scope of analysis in the inspection model 343 is registered by the user. Furthermore, in the image registration process in the embodiment, if information on sections is registered on the insect trapping paper, the user also associates the scope of analysis with the sections. The image registration process will be explained in detail later.

In step S4, the processor 21 performs analysis processing. When the user instructs to execute the analysis, the processor 21 requests the server 3 to perform the analysis. Based on this request, the server 3 inputs the analysis target image registered by the user into the inspection model 343 and performs analysis for inspection. The server 3 then returns the analysis results to the terminal 2a. The analysis process will be explained in detail later, but in the analysis process in the embodiment, pests are identified and counted within the scope of the analysis. Furthermore, based on the correspondence information between the scope of analysis and the divisions, the results of identification and counting of pests in units of divisions are also calculated. When the analysis is completed, the results of the analysis are presented to the user. The user checks the analysis results and makes corrections as necessary. The user then selects, for example, a button indicating that the confirmation has been completed. Upon receiving this, the process moves to step S5.

In step S5, the processor 21 executes a report creation process. After the report creation process, the process moves to step S6. For example, if a format is defined for each business partner, the report can be created according to that format. The report includes information such as supplier name, capture date, and inspection results. Further, the report may further include comments made by the user. Here, the test results include the identification results and counting results of pests stuck to the insect trap paper identified by the test model 343. As described above, when a person visually identifies and counts pests stuck to insect paper, the insect paper may be divided into a plurality of sections and the pests are identified and counted for each section. In embodiments, pest identification and enumeration results on a per-plot basis are also calculated. Therefore, a report equivalent to a report based on a visual inspection performed by a person can be created. After the report creation is completed, the user checks the created report and corrects it as necessary. The user then selects, for example, a button indicating that the confirmation has been completed. Upon receiving this, the process moves to step S6. In the report creation process, instead of creating a report, only the pest identification and counting result data is converted into data in various formats such as CSV (Comma Separated Value) and output. It's okay. In this case, the process may proceed to step S6 without waiting for the user's confirmation.

In step S6, the processor 21 executes countermeasure content registration processing. The countermeasure content registration process is a process in which the user registers the content of the countermeasures implemented after the inspection in association with a report or the like. In the countermeasure content registration process, several typical countermeasures are illustrated to the user, and the user registers whether or not these countermeasures are to be implemented and their contents. Files related to countermeasure content may also be registered as necessary. After the registration of the countermeasure contents is completed, the user selects, for example, a button indicating that the registration is completed. In response to this, the processor 21 transmits the report and data on the contents of the countermeasure to the server 3. After the report is stored in the storage 34 of the server 3, the process in FIG. 4 ends. The report 347 stored in the storage 34 of the server 3 can be viewed by the user of the terminal 2b who is a business partner of the user of the terminal 2a. Further, the report data may be directly transmitted from the terminal 2a to the terminal 2b.

Next, image registration processing will be explained. FIG. 5 is a flowchart showing image registration processing. In step S101, the processor 21 displays an image registration screen on the screen of the pest inspection program 342 of a web browser, for example. FIG. 6 is a diagram showing an example of an image registration screen.

As shown in FIG. 6, the image registration screen 400 includes an information display field 401, an image registration field 402, and an analysis target image registration field 403. The image registration screen 400 also includes a return button 404 and an analysis execution button 405.

The information display column 401 is a display column in which various information related to the examination is displayed. Various information related to the inspection includes, for example, information on business partners and information on the date of capture. The customer information is the name of the selected customer. The capture date is the capture date registered in the period registration process. The information display column 401 may display information other than the information on the business partner and the capture date.

The image registration field 402 is a display field for registering images to be uploaded to the server 3. The image registration field 402 displays, for example, a file selection button 4021, a drag and drop area 4022, a clear button 4023, and an upload button 4024.

The file selection button 4021 is a button operated by the user to select an image that the user wants to upload from a list of images of insect paper. When the file selection button 4021 is selected, for example, a list of insect paper images recorded in the storage 24 is displayed. The user can select an image that he/she wishes to upload to the server 3 from this list. If an image is selected, the file name of the selected image is added to the list of file names to be uploaded. Here, the list of insect paper images displayed when the file selection button 4021 is selected may be a list of all images recorded in the storage 24, or a list of images that have not been uploaded. It may be a list. Furthermore, the list of images may be a list of all images taken on the day of capture, or a list of images taken on the day of capture that have not been uploaded.

The drag and drop area 4022 is an area for selecting files to be uploaded by drag and drop operations. A user may drag and drop an image file icon into drag and drop area 4022. The file name of the image dragged and dropped into the drag and drop area 4022 is also added to the list of file names to be uploaded.

The clear button 4023 is a button that is operated when excluding a file displayed in the list of file names to be uploaded from being uploaded. If the clear button 4023 is selected while at least one file name displayed in the list of file names to be uploaded is selected, that file name is deleted from the list.

The upload button 4024 is a button that is operated when uploading a file displayed in the list of file names to be uploaded to the server 3. When the upload button 4024 is selected, uploading is performed.

The analysis target image registration field 403 is a display field for registering images to be analyzed by the inspection model 343 among the files uploaded to the server 3. The analysis target image registration field 403 displays, for example, a section image tab 4031 and a panoramic image tab 4032.

The partition image tab 4031 is a tab selected when switching the image displayed in the analysis target image registration field 403 to a partition image. The section image is an image taken separately for each section of the insect trapping paper. For example, an image obtained by cutting a sheet of insect paper into multiple pieces, arranging them, and photographing them at the same time is classified as a segmented image.

The panoramic image tab 4032 is a tab selected when switching the image displayed in the analysis target image registration field 403 to a panoramic image. The panoramic image is an image taken without dividing the insect paper into sections. For example, an image in which a sheet of insect paper is photographed as it is is classified as a panoramic image.

Further, in the analysis target image registration column 403, an image 4033 belonging to the selected one of the section image tab 4031 and the panoramic image tab 4032 is displayed. The image displayed in the analysis target image registration column 403 is, for example, an image photographed on the registered capture date among the images recorded in the file already uploaded to the server 3. Further, each image 4033 also displays a check box 4033a, a status display field 4033b, and a delete button 4033c. The check box 4033a is a check box for indicating whether the corresponding image 4033 is being selected. When the corresponding image 4033 is being selected, the check box 4033a is checked. The status display column 4033b is a display column for displaying the status of whether or not the section registration process, which will be described later, has been performed for the corresponding image 4033. When the partition registration process has been completed, the status display field 4033b displays, for example, "Input completed", and when the division registration process has not been completed, the status display field 4033b displays, for example, "Not input". The delete button 4033c is a button selected when excluding the currently selected image 4033 from the analysis target.

The return button 404 is a button that is selected when returning the process to the insect trap registration process. Although the description of FIG. 5 will be omitted below, if the return button 404 is selected during the process of FIG. 5, the image registration process of FIG. Return to insect trap registration process. The analysis execution button 405 is a button that is selected when proceeding to analysis processing.

Here, we return to the explanation of FIG. 5. In step S102 after displaying the image registration screen 400 in step S101, the processor 21 determines whether a file to be uploaded has been selected or deselected. If a file is selected from the file selection button 4021, a file icon is dragged and dropped to the drag and drop area 4022, or the clear button 4023 is selected, it is determined that the file to be uploaded has been selected or deselected. be done. If it is determined in step S102 that a file to be uploaded has been selected or deselected, the process moves to step S103. If it is determined in step S102 that the file to be uploaded has not been selected or deselected, the process moves to step S104.

In step S103, the processor 21 adds or deletes the name of the file to be uploaded to the list depending on the operation. For example, when a file is selected from the file selection button 4021 or a file icon is dragged and dropped into the drag and drop area 4022, the processor 21 adds the file name of the corresponding file to the list. When the clear button 4023 is selected, the file name of the corresponding file is deleted from the list.

In step S104, the processor 21 determines whether to upload the file. If the upload button 4024 is selected, it is determined that the file is to be uploaded. If it is determined in step S104 that the file should be uploaded, the process moves to step S105. If it is determined in step S104 that the file is not to be uploaded, the process moves to step S106.

In step S105, the processor 21 uploads the files included in the upload target list to the server 3. The uploaded file is stored in the storage 34 of the server 3 as a photographed image 346.

In step S106, the processor 21 determines whether the analysis target image has been selected or deselected. When an unchecked checkbox 4033a is selected, it is determined that an image to be analyzed has been selected. On the other hand, when the checked checkbox 4033a is selected, it is determined that the selection has been canceled. If it is determined in step S106 that an image has been selected or deselected, the process moves to step S107. If it is determined in step S106 that no image has been selected or deselected, the process moves to step S108.

In step S107, the processor 21 adds the selected image to the list of images to be analyzed. Alternatively, the processor 21 deletes the deselected image from the list of images to be analyzed. When an image to be analyzed is selected, the processor 21 checks the check box 4033a. On the other hand, when the selection of the analysis target image is canceled, the processor 21 removes the check from the check box 4033a.

In step S108, the processor 21 determines whether to delete the image. It is determined that the image is to be deleted when the delete button 4033c is selected. If it is determined in step S108 that the image is to be deleted, the process moves to step S109. If it is determined in step S108 that the image is not to be deleted, the process moves to step S110.

In step S109, the processor 21 deletes the image 4033 corresponding to the selected delete button 4033c from the analysis target image registration field 403. In addition to the image 4033, image files uploaded to the server 3 may also be deleted. On the other hand, the image file uploaded to the server 3 may remain while the image 4033 is deleted from the analysis target image registration field 403.

In step S110, the processor 21 determines whether to perform partition registration processing. For example, when any image 4033 is selected, it is determined that the section registration process is to be performed. Alternatively, it may be determined that the section registration process is to be performed when a dedicated button for the section registration process is selected. In this case, when a dedicated button for section registration processing is selected while a plurality of images 4033 are selected, it may be determined that section registration processing is to be performed for these plurality of images 4033. If it is determined in step S110 that the block registration process is to be performed, the process moves to step S111. If it is determined in step S110 that the partition registration process is not to be performed, the process moves to step S112.

In step S111, the processor 21 performs partition registration processing. After performing the section registration process, the process moves to step S112. The partition registration process will be explained in detail later.

In step S112, the processor 21 determines whether or not to perform analysis processing. For example, when the analysis execution button 405 is selected, it is determined that analysis processing is to be performed. If it is determined in step S112 that the analysis process is to be performed, the process moves to step S113. If it is determined in step S112 that the analysis process is not to be performed, the process returns to step S102.

In step S113, the processor 21 determines whether the partitions have been registered for the image to be analyzed, that is, whether the partition registration process for the image to be analyzed has been performed. If it is determined in step S113 that the section has been registered, the image registration process ends. In this case, the process moves to the analysis process of step S4 in FIG. If it is determined in step S113 that the partition has not been registered, the process moves to step S114. As will be explained later, in the section registration process, the management number and section information are associated with each analysis target image. In the determination in step S113, if the management numbers have been associated, it may be determined that the registration has been completed even if there is no partition information.

In step S114, the processor 21 notifies the user of an error indicating that the partition is not registered. Error notification may be performed, for example, by displaying an error message. After that, the process returns to step S102.

Next, the section registration process will be explained. FIG. 7 is a flowchart showing the section registration process. In step S201, the processor 21 displays a section registration screen. FIG. 8 is a diagram showing an example of a section registration screen.

As shown in FIG. 8, the section registration screen 500 includes an information display field 501, an analysis implementation range registration field 502, and a section registration field 503. Furthermore, the section registration screen 500 includes a return button 504 and an analysis execution button 505.

The information display field 501 is a display field in which various information related to the examination is displayed. For example, the same information as the information displayed in the information display field 401 may be displayed in the information display field 501. If necessary, information different from that in the information display field 401 may be displayed in the information display field 501.

The analysis implementation range registration field 502 is a display field for registering the implementation range of analysis of the analysis target image by the inspection model 343. The analysis execution range registration field 502 displays an analysis target image 5021, an enlarge button 5022, and a select all button 5024. Furthermore, an implementation range selection frame 5023 may be displayed on the analysis target image 5021 by the user's operation.

The analysis target image 5021 is an image that is the target of the division registration process among the images that are set as the analysis target images in the image registration process. When a plurality of images are simultaneously targeted for section registration processing, a plurality of analysis target images 5021 may be displayed simultaneously or in a switched manner. In this case, the enlarge button 5022 and the entire selection button 5024 may be provided for each analysis target image 5021.

The enlarge button 5022 is a button selected to enlarge/reduce the analysis target image 5021. Every time the enlarge button 5022 is selected, the magnification of the analysis target image 5021 displayed in the analysis range registration field 502 is changed. This helps the user to visually recognize pests and the like stuck to the insect paper in the analysis target image 5021. Here, a portion that cannot be displayed due to enlargement may be displayed by scrolling, for example. Instead of the enlargement button 5022, a UI (User Interface) for inputting a numerical value of the enlargement ratio may be provided, or another UI may be provided.

The implementation range selection frame 5023 may be displayed in response to a user's drag operation on the analysis target image 5021. The range of the implementation range selection frame 5023 becomes the implementation range of the analysis by the inspection model 343.

The entire selection button 5024 is a button for setting the entire range of the analysis target image 5021 as the range of analysis. When the entire selection button 5024 is selected, an implementation range selection frame 5023 is displayed so as to include the entire range of the analysis target image 5021.

The section registration column 503 is a display column for registering sections for the analysis target image 5021. In the section registration column 503, a number (No.) assigned to the information of each section regarding the analysis target image 5021 is displayed. Furthermore, the section registration field 503 is provided with input fields 5031 for management number, installation location, total number of sections, number of selected sections, and status as section information for the analysis target image 5021.

The management number is the number assigned when registering the insect trap. As described above, the management number indicates which insect trap and which insect paper the analysis target image 5021 is an image of.

The installation location is information on the installation location of the insect trap with the corresponding management number, and is the information on the installation location registered at the time of insect trap registration.

The total number of sections is the total number of sections of insect paper in the analysis target image 5021. For example, if one sheet of insect trapping paper is divided into 10 sections, "10" may be registered as the total number of sections.

The number of selected sections is the number of sections included in the implementation range selection frame 5023. In other words, the number of selected sections indicates how many sections the range of analysis by the inspection model 343 corresponds to. Here, if the boundary of the division and the boundary of the implementation range selection frame 5023 match, the number of selected divisions takes an integer value. For example, in FIG. 8, "1" is registered as the number of selected sections. In this case, the implementation range selection frame 5023 corresponds to one section set on the insect trapping paper. On the other hand, if the boundaries of the sections and the boundaries of the implementation range selection frame 5023 do not match, the number of selected sections can take a decimal value.

The status indicates whether or not the analysis target image 5021 has been analyzed by the inspection model 343. When the analysis has been completed, for example, "Analysis completed" is input as the status, and when the analysis has not been performed, for example, "Not performed" is input as the status. Input of information to the status may be automatically performed by the processor 21 without user operation.

When no partition information is associated with the analysis target image 5021, the input fields 5031 for management number, installation location information, total number of partitions, number of selected partitions, and status are blank. Information can be registered in each input field 5031 by subsequent editing by the user. When performing the partition registration process after partition information has been associated with the analysis target image 5021, input fields 5031 for management number, installation location information, total number of partitions, number of selected partitions, and status are filled out in advance. Associated information will be displayed.

Additionally, an edit button 5032 is provided in the section registration field 503. The edit button 5032 is a button selected to edit the management number, total number of sections, number of selected sections, and status. After selecting the edit button 5032, the user can edit the contents of the management number, total number of sections, number of selected sections, and status.

Additionally, a deletion checkbox 5033 is provided in the section registration column 503. Furthermore, the section registration column 503 is provided with a delete button 5034. When the delete button 5034 is selected, the information of the sections whose delete checkbox 5033 is checked is deleted all at once.

The return button 504 is a button that is selected when returning the process to the image registration process. The analysis execution button 505 is a button that is selected when proceeding to analysis processing. That is, in the embodiment, the process can proceed to the analysis process even in the partition registration process.

Here, we return to the explanation of FIG. 7. In step S202 after displaying the section registration screen 500 in step S201, the processor 21 determines whether to display the analysis target image 5021 in an enlarged or reduced size. For example, when the enlarge button 5022 is selected, it is determined that the display will be enlarged or reduced. When it is determined in step S202 that the analysis target image 5021 should be displayed in an enlarged or reduced size, the process moves to step S203. If it is determined in step S202 that the analysis target image 5021 should not be displayed in an enlarged or reduced size, the process moves to step S204.

In step S203, the processor 21 changes the magnification of the analysis target image 5021 displayed in the section registration field 503 to display it in an enlarged or reduced size.

In step S204, the processor 21 determines whether the scope of analysis has been specified. For example, it is determined that the range of analysis has been specified when a drag operation is performed on the analysis target image 5021 or when the select all button 5024 is selected. When the scope of analysis is specified in step S204, the process moves to step S205. In step S204, if the scope of analysis has not been specified, the process moves to step S206.

In step S205, the processor 21 displays an implementation range selection frame 5023 in the range specified by the user. Further, the processor 21 registers information on the coordinates of the implementation range selection frame 5023 in the analysis target image 5021 as information on the implementation range of the analysis.

In step S206, the processor 21 determines whether to edit the partition information. For example, when the edit button 5032 is selected, it is determined that the section information is to be edited. If it is determined in step S206 that the partition information is to be edited, the process moves to step S207. If it is determined in step S206 that the partition information is not to be edited, the process moves to step S208.

In step S207, the processor 21 updates the partition information according to the user's editing operation. Along with this, the processor 21 also updates the display of each input field 5031.

In step S208, the processor 21 determines whether a return operation has been performed. For example, when the return button 504 is selected, it is determined that a return operation has been performed. In step S208, when it is determined that a return operation has been performed, the section registration process ends. In this case, the process returns to the image registration process. In step S208, if the return operation has not been performed, the process moves to step S209.

In step S209, the processor 21 determines whether or not to perform analysis processing. For example, when the analysis execution button 505 is selected, it is determined that analysis processing is to be performed. If it is determined in step S209 that the analysis process is to be performed, the process moves to step S210. If it is determined in step S209 that the analysis process is not to be performed, the process returns to step S202.

In step S210, the processor 21 determines whether the partitions have been registered for the image to be analyzed, that is, whether the partition registration process for the image to be analyzed has been performed. If it is determined in step S210 that the section has been registered, the image registration process ends. In this case, the process moves to the analysis process of step S4 in FIG. If it is determined in step S210 that the partition has not been registered, the process moves to step S211. In the determination in step S210, if the management numbers have been associated, it may be determined that the registration has been completed even if there is no partition information.

In step S211, the processor 21 notifies the user of an error indicating that the partition is not registered. After that, the process returns to step S202.

Next, the analysis process will be explained. FIG. 9 is a flowchart showing the analysis process. In step S301, the processor 21 displays an analysis result confirmation screen on the screen of the pest inspection program 342 of a web browser, for example. FIG. 10 is a diagram showing an example of an analysis result confirmation screen.

As shown in FIG. 10, the analysis result confirmation screen 600 includes an information display field 601, an analysis execution button 602, a status display field 603, a back button 604, and a report creation button 605.

The information display column 601 is a display column in which various information related to the examination is displayed. For example, the same information as the information displayed in the

information display columns

401 and 501 can be displayed in the information display column 601. If necessary, information different from the

information display columns

401 and 501 may be displayed in the information display column 601.

The analysis execution button 602 is a button that is selected when performing analysis using the test model 343. When the analysis execution button 602 is selected, an analysis execution request is transmitted from the terminal 2a to the server 3.

The status display column 603 is a display column for displaying the results of analysis by the inspection model 343. The status display field 603 includes a target display field 6031, a status display field 6032, an image display field 6033, and a modification status display field 6034. Further, a scroll bar 6035 for scrolling display is also displayed in the status display field 603.

The target display field 6031 is a display field in which a UI for performing various operations on each analysis target image is arranged. A confirmation button 6031a may be displayed in the target display field 6031. Further, a check box 6031b may be displayed in the target display field 6031.

One confirmation button 6031a may be displayed for images to be analyzed that have the same management number. The confirmation button 6031a is a button selected to confirm or edit the analysis result. If the confirmation button 6031a is selected, confirmation/correction processing, which will be described later, is started. Here, the confirmation button 6031a is configured so that it cannot be selected when the status of the image to be analyzed with the corresponding management number is "not yet performed."

One check box 6031b may be displayed for analysis target images having the same management number. The check box 6031b is displayed only when the status of the corresponding management number is "not implemented". The check box 6031b is a check box for selecting whether or not to perform analysis using the inspection model 343 on the analysis target image with the corresponding management number. When the check box 6031b is checked, the analysis target image with the corresponding management number becomes the target of analysis by the inspection model 343.

The status display column 6032 is a display column for displaying the status of each analysis target image. When the analysis of the corresponding image to be analyzed has been completed, for example, "Analysis completed" is displayed in the status display field 6032, and when the analysis has not been completed, for example, "Not performed" is displayed in the status display field 6032.

The image display column 6033 is a display column for the analysis target image registered in the image registration process. In the image display field 6033, images to be analyzed are displayed for each management number. The analysis target image displayed in the image display field 6033 also displays the implementation range selection frame registered in the section registration process.

The modification status display field 6034 is a display field for displaying the modification status indicating whether or not the user has manually modified the analysis result for each image to be analyzed. For example, "corrected" is displayed when the analysis result of the corresponding analysis target image has been corrected. On the other hand, if the analysis results of the relevant image to be analyzed have not been corrected, nothing is displayed.

The return button 604 is a button that is selected when returning the process to the image registration process. Although the description of FIG. 9 will be omitted below, if the return button 604 is selected during the process of FIG. 9, the analysis process of FIG. Return to registration process. The report creation button 605 is a button that is selected when proceeding to report creation processing.

Here, we return to the explanation of FIG. 9. In step S302 after displaying the analysis result confirmation screen 600 in step S301, the processor 21 determines whether or not an analysis target image has been selected or deselected. When an unchecked checkbox 6031b is selected, it is determined that an image to be analyzed has been selected. On the other hand, when the checked checkbox 6031b is selected, it is determined that the selection has been canceled. If it is determined in step S302 that the image to be analyzed has been selected or deselected, the process moves to step S303. If it is determined in step S302 that the analysis target image has not been selected or deselected, the process moves to step S304.

In step S303, the processor 21 adds the selected image to be analyzed to the list of targets for analysis. Alternatively, the processor 21 deletes the deselected analysis target image from the list of analysis targets. When an image to be analyzed is selected, the processor 21 checks the check box 6031b. On the other hand, when the selection of the analysis target image is canceled, the processor 21 removes the check from the check box 6031b.

In step S304, the processor 21 determines whether to perform analysis using the test model 343. For example, when the analysis execution button 602 is selected, it is determined that the analysis is to be executed. If it is determined in step S304 that analysis is to be performed, the process moves to step S305. If it is determined in step S304 that analysis is not to be performed, the process moves to step S306.

In step S305, the processor 21 requests the server 3 to perform analysis. At this time, the processor 21 sends the list of targets for analysis to the server 3. This list includes the file names of the images to be analyzed, information on the range of analysis associated with each image to be analyzed, and information on sections. The processor 31 of the server 3 receives the request from the terminal 2a, refers to the list of targets for analysis, and reads the corresponding file from the captured image 346 in the storage 34. Then, the processor 31 inputs the image of the read file to the inspection model 343 as an analysis target image, and performs analysis using the inspection model 343. The processor 31 then returns the analysis result based on the test model 343 to the terminal 2a that requested the analysis. Upon receiving this, the process moves to step S306.

In step S306, the processor 21 updates the display in the status display column 603 of the analysis result confirmation screen 600. For example, the processor 21 updates the status of the analysis target image for which the analysis result has been received to "analysis complete." Furthermore, the processor 21 makes it possible to select the confirmation button 6031a for the relevant image to be analyzed, and hides the checkbox 6031b for the relevant image to be analyzed.

In step S307, the processor 21 determines whether to confirm or modify the analysis results. For example, if the confirmation button 6031a is selected, it is determined that the analysis results should be confirmed or modified. If it is determined in step S307 that the analysis result should be confirmed or modified, the process moves to step S308. If it is determined in step S307 that the analysis result is not to be confirmed or modified, the process moves to step S309.

In step S308, the processor 21 performs confirmation/correction processing. After the confirmation/correction process, the process moves to step S309. The confirmation/correction process will be explained in detail later.

In step S309, the processor 21 determines whether or not to perform report creation processing. For example, when the report creation button 605 is selected, it is determined that the report creation process is to be performed. When it is determined in step S309 that the report creation process is to be performed, the analysis process ends. In this case, the process moves to the report creation process in step S5 of FIG. If it is determined in step S309 that the report creation process is not to be performed, the process returns to step S302.

Next, the confirmation/correction process will be explained. FIG. 11 is a flowchart showing the confirmation/correction process. In step S401, the processor 21 displays a confirmation/correction screen on the screen of the pest inspection program 342 of a web browser, for example. 12A and 12B are diagrams showing examples of confirmation/correction screens. Here, FIG. 12A shows a confirmation/correction screen displayed immediately after the confirmation/correction process. On the other hand, FIG. 12B shows a confirmation/correction screen displayed by scrolling the confirmation/correction screen of FIG. 12A downward.

As shown in FIG. 12A, the upper screen of the confirmation/correction screen 700 includes an information display field 701, an analysis result display field 702, a back button 703, an identification/counting correction button 704, and an update button 705. .

The information display column 701 is a display column in which various information related to the examination is displayed. The information display column 701 may display information such as a customer name, installation location, management number, investigation method, and capture date, for example. Information other than these may be displayed in the information display field 701, or some information may not be displayed in the information display field 701.

The analysis result display column 702 is a display column for displaying the analysis result for the analysis target image corresponding to the selected confirmation button 6031a. The analysis result display field 702 includes a type display field 7021, an image total display field 7022, a number of partitions display field 7023, a coefficient check box display field 7024, an overall total display field 7025, and an overall change display field 7026.

The type display column 7021 is a display column for displaying a list of types of pests identified in the image to be analyzed. Here, the types of pests may be color-coded or the like. For example, similar types of pests may be divided into the same color, or the color of highly dangerous pests may be distinguished from the color of other pests, etc. Moreover, "unclassified" shown in FIG. 12A is a type that is classified when it cannot be identified as a specific type of pest, for example, because the reliability of prediction was below a threshold value in the analysis of the image to be analyzed. “Unclassified” does not necessarily have to be provided.

The image total display column 7022 is a display column for displaying the image total for each pest type. The image total is the value before and after the change of the total of each pest identified in the image to be analyzed. Note that when analysis is performed on two or more images to be analyzed, the image total is the sum of the totals for each of the images to be analyzed. Moreover, the "before change" is the total before being modified by the user. If no modification has been made by the user, the image before modification is the sum of images obtained as a result of analysis by the inspection model 343. On the other hand, after modification is the total after being modified by the user. Further, for example, a frame 7022a is displayed for the image total that can have a large effect on the total of the entire insect trapping paper before and after the change, thereby distinguishing it from the others. The image sum, which has a large effect on the total of the entire insect trapping paper before and after the change, is a small number such as 1, for example. In other words, if there are a small number of sources, even a small change will have a large effect on the overall total. Such a distinguishing display is provided to allow the user to confirm whether such a change may be implemented.

The number of sections display column 7023 is a display column for displaying information about the number of sections of the image to be analyzed. The measurement section and the total section can be displayed in the section number display field 7023. The measurement section is the number of sections included in the range of analysis of the image to be analyzed, and corresponds to the number of selected sections registered during the section registration process. The total number of sections is the number of sections set on the target insect trapping paper of the image to be analyzed, and corresponds to the total number of sections registered during the section registration process. Note that when analysis is performed on two or more images to be analyzed, the measurement section is the total number of selected sections for each image to be analyzed. On the other hand, the total number of sections is the total number of sections.

The coefficient check box display field 7024 is a coefficient check box display field for selecting whether or not to employ the partition method in calculating the overall total displayed in the overall total display field 7025. The division method is a method of calculating the total number of pests in all sections of insect paper based on the counting results in some sections of insect paper. If the coefficient checkbox is checked, the overall total is calculated using the parcel method. Specifically, the overall total is calculated by calculating (overall total)=(image total)×(overall section)/(measurement section). In this way, in the embodiment, calculations based on the partition method using partition information are used to calculate the results of counting in units of partitions from the first counting result, which is the result of counting in the range of analysis performed by the inspection model 343. A second count result may be calculated. Note that when the calculation result is a decimal number, the overall total may be a rounded number. On the other hand, if the coefficient checkbox is not checked, the overall total is the image total.

The overall total display column 7025 is a display column for displaying the overall total. The overall total of pests with a check in the coefficient checkbox is the total of the corresponding type of pests in the entire insect trapping paper calculated by the division method. The overall total of pests whose coefficient checkboxes are not checked is the same as the image total.

The overall change display field 7026 is a display field that includes an input field for accepting the user's input of a revised value for the overall total. In the overall change display field 7026, the overall total value based on the analysis result by the inspection model 343 is displayed as an initial value. The user can input the total value of the pests to be modified in the total change display field 7026 directly or by selecting an increase/decrease icon. After the analysis results are reflected, the total number modified by the user is displayed in the total change display column 7026. Here, the counting result entered in the report may be the overall total registered in the overall change display field 7026.

The return button 703 is a button that is selected when returning the process to the analysis process. The identification/counting correction button 704 is a button that is selected when proceeding to identification/counting correction processing. The update button 705 is a button that is selected when returning the process to the analysis process after reflecting the corrections made by the user.

As shown in FIG. 12B, the lower screen of the confirmation/correction screen 700 includes an analysis result editing field 706. The analysis result editing field 706 is a display field for editing the analysis results. The analysis result editing field 706 displays an analysis target image 7061, a type display field 7062, a count result n display field 7063, a change result n display field 7064, a change result n input field 7065, and a delete n check box 7066. Furthermore, the analysis result editing field 706 displays a measurement section number display field 7067, a totaling target checkbox 7068, an add button 7069, and a delete button 7070.

The analysis target image 7061 is an analysis target image on which analysis processing has been performed. When analysis processing is being performed simultaneously on a plurality of images to be analyzed, a plurality of images to be analyzed 7061 may be displayed.

The type display field 7062 is a display field for displaying a list of types of pests identified in the analysis target image of the pests identified in the analysis target image 7061. A type display field 7062 is provided for each analysis target image 7061.

The count result n display column 7063 is a display column for the count results for each pest type based on the inspection model 343 for the n-th (n is a positive integer) image to be analyzed 7061. The count result n display column 7063 is provided for each analysis target image 7061.

The change result n display column 7064 is a display column for the count results for each type of pest that have been changed by the user for the n-th analysis target image 7061. When the counting results have not been changed by the user, the changed result n display column 7064 displays the counting results for each pest type by the inspection model 343. A change result n display field 7064 is provided for each analysis target image 7061.

The change result n input field 7065 is an input field for receiving the user's input of a change value of the counting result for the n-th image to be analyzed 7061. The change result n input field 7065 displays the value of the counting result displayed in the change result n display field 7064 as an initial value. The user can input the value of the pest count result to be changed in the change result n input field 7065 directly or by selecting an increase/decrease icon. According to the change result in the change result n input field 7065, the values of the image total and overall total described above can also be changed. A change result n input field 7065 is provided for each analysis target image 7061.

The delete n check box 7066 is a display column for a delete check box for selecting whether or not the count result for each pest regarding the n-th image to be analyzed 7061 is to be deleted. A delete n checkbox 7066 is provided for each analysis target image 7061.

The measurement section number display column 7067 is a display column for the number of measurement sections in the n-th analysis target image 7061, that is, the number of sections included in the analysis implementation range.

The aggregation target check box 7068 is a check box for selecting whether or not the analysis result for the n-th analysis target image 7061 is to be aggregation target. If the aggregation target check box 7068 is checked, the analysis result for the n-th analysis target image 7061 is subject to the aggregation of the image total and overall total when displayed in the analysis result display field 702. If the aggregation target check box 7068 is not checked, the analysis results for the n-th analysis target image 7061 are excluded from the aggregation targets displayed in the analysis result display field 702.

The add button 7069 is a button that is selected when adding a type of pest other than the pest displayed in the type display column 7062 for the n-th image to be analyzed 7061. When the add button 7069 is selected, a dialog for selecting an additional pest type is displayed, for example. The user selects the type of pest to be added from this dialog. As a result, the item of pest type is added to the n-th analysis target image 7061. The add button 7069 may be provided for each image to be analyzed 7061, or only one button may be provided.

The delete button 7070 is a button selected when deleting the analysis result for the n-th image to be analyzed 7061. When the delete button 7070 is selected, information on sections whose deletion checkboxes are checked is deleted all at once. The delete button 7070 may be provided for each analysis target image 7061, or only one delete button 7070 may be provided.

Here, we return to the explanation of FIG. 11. In step S402 after displaying the confirmation/correction screen 700 in step S401, the processor 21 determines whether there is a correction operation by the user. For example, if there is an input in the overall change display field 7026 or the change result n input field 7065, and if the delete button 7070 is selected, it is determined that the user has made a correction operation. If it is determined in step S402 that there is a correction operation by the user, the process moves to step S403. If it is determined in step S402 that there is no correction operation by the user, the process moves to step S404.

In step S403, the processor 21 updates the confirmation/correction screen 700 according to the content of the correction operation by the user. For example, when the counting result is changed in the change result n input field 7065, the processor 21 updates the changed image total value in the image total display field 7022, and also updates the image total value after the change in the image total display field 7022. Recalculate the overall total value based on Then, the processor 21 displays the overall total value after recalculation in the overall change display column 7026. Further, for example, if the coefficient checkbox in the coefficient checkbox display column 7024 is checked or unchecked, the processor 21 calculates the overall total using or not using the partition method. Recalculate the value. Then, the processor 21 displays the overall total value after recalculation in the overall change display column 7026. Note that here, only the display is updated, and the actual values such as the overall total are not updated.

In step S404, the processor 21 determines whether to perform identification/count correction processing. For example, when the identification/count correction button 704 is selected, it is determined that the identification/count correction process is to be performed. If it is determined in step S404 that the identification/count correction process is to be performed, the process moves to step S405. If it is determined in step S404 that the identification/count correction process is not performed, the process moves to step S406.

In step S405, the processor 21 performs identification/count correction processing. After the identification/counting process, the process moves to step S406. Identification/count correction processing will be explained in detail later.

In step S406, the processor 21 determines whether or not to reflect the modification made by the user. For example, when the update button 705 is selected, it is determined that the corrections made by the user are reflected. If it is determined in step S406 that the modification by the user is to be reflected, the process moves to step S407. If it is determined in step S407 that the modification by the user is not reflected, the process moves to step S408.

In step S407, the processor 21 reflects the corrections made by the user. For example, when the overall total has been changed, the processor 21 overwrites the previous overall total with the changed overall total. Thereafter, the confirmation/correction process ends. In this case, the process moves to step S309 in FIG.

In step S408, the processor 21 determines whether a return operation has been performed. For example, when the return button 703 is selected, it is determined that a return operation has been performed. If it is determined in step S408 that a return operation has been performed, the confirmation/correction process ends. In this case, the confirmation/correction process ends without the user's corrections being reflected. The process then moves to step S309 in FIG. If it is determined in step S408 that a return operation has not been performed, the process returns to step S402.

Next, the identification/count correction process will be explained. FIG. 13 is a flowchart showing the identification/count correction process. In step S501, the processor 21 displays an identification/count correction screen on the screen of the pest inspection program 342 of a web browser, for example. FIG. 14 is a diagram showing an example of an identification/count correction screen.

As shown in FIG. 14, the identification/counting correction screen 800 includes a correction target image display field 801, an enlargement/reduction UI 802, a diagnosis result list display field 803, an image selection field 804, a return button 805, and an update button. 806.

The correction target image display column 801 is a display column of the correction target image 8011 that is the target of the identification/count correction process among the analysis target images. The correction target image 8011 is the analysis target image selected in the image selection field 804. Furthermore, an identification range identification frame 8012 is displayed in the correction target image display field 801. The identification range identification frame 8012 is a frame indicating a location identified in the inspection model 343 of the image to be analyzed. Further, a scroll bar 8013 is provided in the correction target image display field 801. The scroll bar 8013 can be selected when scrolling the correction target image 8011.

The enlargement/reduction UI 802 is a UI for enlarging/reducing the correction target image 8011. In FIG. 13, a slider 8021 is provided as the enlargement/reduction UI 802. Any magnification ratio can be specified by the user's operation of the slider. Instead of or together with the slider 8021, an enlargement button, a UI for inputting a numerical value of an enlargement ratio, etc. may be provided.

The diagnosis result list display field 803 is a display field in which the analysis results of the correction target image 8011 using the inspection model 343 are displayed. In the diagnosis result list display field 803, a list 8031 of pest types identified in each identification range identification frame 8012 is displayed. The list includes numbers associated with identification range identification frames 8012 and names of pest types as identification results in the corresponding identification range identification frames 8012. Radio buttons 8031a for selection are provided near each identification result. Furthermore, the display of the name of the pest type as each identification result is configured as a drop-down list 8031b. When drop-down list 8031b is selected, a list of pests is displayed. The user can select the type of pest he/she desires from this list of pests. Further, a delete button 8031c is also provided near the name of the pest type as each identification result. When the delete button 8031c is selected, the identification result in the corresponding row is deleted.

Additionally, the diagnosis result list display column 803 displays an add button 8032, a delete button 8033, and an add pest type button 8034. The add button 8032 is a button that is selected when adding information about a pest that is not displayed in the list 8031 of pest types. When the add button 8032 is selected, for example, a line with a new number is added to the end. The delete button 8033 is a button selected when deleting the identification results for the correction target image 8011 all at once. The pest type addition button 8034 is a button that is selected when adding information about a pest that is not on the pest list. When the pest type addition button 8034 is selected, a dialog for adding a pest type is displayed.

The image selection field 804 is a display field for selecting an image to be corrected. The image selection column 804 displays a preview 8051 of the image that is the target of the identification/count correction process among the analysis target images. The selected image among the image previews 8051 is displayed with a thick frame, for example. Further, a forward button 8042 and a backward button 8043 are also displayed in the image selection field 804. The forward button 8042 is a button selected to advance the currently selected image by one. The return button 8043 is a button selected to return the currently selected image by one. The user can switch images using a forward button 8042 and a backward button 8043. The image may be forwarded or reversed in response to a flick operation or the like.

The return button 805 is a button that is selected when returning the process to the confirmation/correction process. The update button 806 is a button that is selected when returning the process to the confirmation/correction process after reflecting the corrections made by the user.

Here, we return to the explanation of FIG. 13. In step S502 after displaying the identification/counting correction screen 800 in step S501, the processor 21 determines whether to display the correction target image 8011 in an enlarged or reduced size. For example, when the slider 8021 is operated, it is determined that the display will be enlarged or reduced. If it is determined in step S502 that the correction target image 8011 should be displayed in an enlarged or reduced size, the process moves to step S503. If it is determined in step S502 that the correction target image 8011 should not be displayed in an enlarged or reduced size, the process moves to step S504.

In step S503, the processor 21 changes the magnification of the correction target image 8011 displayed in the correction target image display field 801 to enlarge or reduce the display.

In step S504, the processor 21 determines whether any radio button 8031a in the pest type list 8031 or any identification range identification frame 8012 has been selected. In step S504, when any radio button 8031a in the pest type list 8031 or any identification range identification frame 8012 is selected, the process moves to step S505. In step S504, if the radio button 8031a or the identification range identification frame 8012 is not selected, the process moves to step S506.

In step S505, the processor 21 displays the selection on the identification/counting correction screen 800 according to the selection. For example, when the radio button 8031a is selected, the processor 21 displays the identification result corresponding to the selected radio button 8031a as being selected. Further, the processor 21 displays the identification range identification frame 8012 corresponding to the selected radio button 8031a as being selected. Further, for example, when the identification range identification frame 8012 is selected, the processor 21 displays the selected identification range identification frame 8012 as being selected. Further, the processor 21 displays the identification result corresponding to the selected identification range identification frame 8012 as being selected.

In step S506, the processor 21 determines whether the user has performed a modification operation. For example, if the drop-down list 8031b is selected, it is determined that the user has performed a modification operation. If it is determined in step S506 that the user has performed a correction operation, the process moves to step S507. If it is determined in step S506 that no correction operation has been performed by the user, the process moves to step S508.

In step S507, the processor 21 updates the identification/counting correction screen 800 according to the corrections made by the user. For example, when the type of pest is changed using the drop-down list 8031b, the processor 21 updates the display of the name of the pest type. Note that here, only the display is updated, and the actual values such as the overall total are not updated.

In step S508, the processor 21 determines whether or not to add a pest type. For example, when the pest type addition button 8034 is selected, it is determined that a pest type is to be added. If it is determined in step S508 that a pest type is to be added, the process moves to step S509. If it is determined in step S508 that the pest type is not to be added, the process moves to step S510.

In step S509, the processor 21 adds the name of the pest type to the pest list according to the user's operation.

In step S510, the processor 21 determines whether an operation to change the correction target image has been performed. For example, when the forward button 8042 and the backward button 8043 are selected, it is determined that the image to be corrected has been changed. If it is determined in step S510 that an operation to change the correction target image has been performed, the process moves to step S511. If it is determined in step S510 that no operation has been performed to change the correction target image, the process moves to step S512.

In step S511, the processor 21 changes the correction target image according to the user's operation. Along with this, the processor 21 updates the identification/count correction screen 800. For example, the processor 21 displays the modified image to be corrected 8011 in the image to be corrected display field 801 . Further, the processor 21 changes the identification result displayed in the diagnosis result list display field 803 to the identification result for the modified correction target image 8011.

In step S512, the processor 21 determines whether or not to reflect the corrections made by the user. For example, when the update button 806 is selected, it is determined that the corrections made by the user are reflected. If it is determined in step S512 that the modification by the user is to be reflected, the process moves to step S513. If it is determined in step S512 that the modification by the user is not reflected, the process moves to step S514.

In step S513, the processor 21 reflects the corrections made by the user. For example, when the type of pest has been changed, the processor 21 recalculates the image total and overall total based on the changed pest type, and uses the recalculated image total and overall total to sum up the image before the change. and overwrite the overall total. Thereafter, the identification/count correction process ends. In this case, the process moves to step S406 in FIG. 11.

In step S514, the processor 21 determines whether a return operation has been performed. For example, when the return button 805 is selected, it is determined that a return operation has been performed. In step S514, when it is determined that a return operation has been performed, the identification/count correction process ends. In this case, the identification/counting correction process ends without any corrections made by the user being reflected. The process then moves to step S406 in FIG. 11. If it is determined in step S514 that a return operation has not been performed, the process returns to step S502.

As described above, according to the embodiment, the processing from inspection based on the image of insect paper to creation of a report is performed as processing on one application. Therefore, the PCO can perform a series of operations from inspection to report creation according to the processing flow. This greatly reduces the workload associated with inspection of pests by PCO.

Further, according to the embodiment, identification and counting of pests based on images of insect paper is performed by an inspection model. This also reduces the burden of pest inspection by the PCO. Furthermore, the identification and counting results by the test model can be modified by the user. A test model using AI performs analysis using a statistical method, so it is theoretically impossible to derive 100% analysis results. By allowing the user to modify the results of the inspection model, a report can be created based on more accurate analysis results. In addition, the PCO only needs to confirm errors, and the workload is small.

Furthermore, in the embodiment, by having the PCO register the partition information, the counting results based on the partition method can be presented. If the division method is used, there is no need to inspect the entire area of insect paper. However, if only a part of the insect paper is inspected, it is difficult for the inspection model to recognize how many sections the inspection has been carried out. In the embodiment, the correspondence information between the scope of analysis and the zone is input by inputting zone information by PCO, so that the scope of analysis and zone are associated with a simple process, and the zone law is followed. based counting results can be calculated correctly.

[Modified example]
A modified example will be explained below. In the embodiment described above, various screens are displayed, but these screens are merely examples and may be changed as appropriate. For example, a display may be displayed on each screen indicating which process among the overall processes shown in steps S1 to S6 the current process is. Further, in this display, a link may be set for each process, and the process may be changed by selecting the link. Further, when selecting a business partner, the progress of processing for each business partner may be displayed. In this case, if a business partner in process is selected, the process may be directly transferred to the corresponding process regardless of the order of steps S1 to S6.

Furthermore, in the embodiment described above, the pest inspection program is a program executed on the web browser of the terminal 2a. As mentioned above, the pest inspection program may be a program executed on a server.

Further, each process according to the embodiment described above can also be stored as a program that can be executed by the

processor

21 or 31 that is a computer. In addition, it can be stored and distributed in a storage medium of an external storage device such as a magnetic disk, an optical disk, or a semiconductor memory. The

processor

21 or 31 reads a program stored in the storage medium of this external storage device, and its operations are controlled by the read program, thereby being able to execute the above-described processes.

Furthermore, the present invention is not limited to the above-described embodiments, and can be variously modified at the implementation stage without departing from the spirit thereof. Moreover, each embodiment may be implemented in combination as appropriate, and in that case, the combined effect can be obtained. Furthermore, the embodiments described above include various inventions, and various inventions can be extracted by combinations selected from the plurality of constituent features disclosed. For example, if a problem can be solved and an effect can be obtained even if some constituent features are deleted from all the constituent features shown in the embodiment, the configuration from which these constituent features are deleted can be extracted as an invention.

1 system, 2a, 2b terminal, 3 server, 21 processor, 22 ROM, 23 RAM, 24 storage, 25 input interface, 26 display, 27 camera, 28 communication module, 31 processor, 32 ROM, 33 RAM, 34 storage, 35 Input interface, 36 Display, 37 Communication module, 241 OS, 341 OS, 342 Pest inspection program, 343 Inspection model, 344 Business partner information, 345 User information, 346 Photographed image, 347 Report.

Claims

Registration by the user of the scope of implementation of the identification in an inspection model configured to identify at least the type of pest stuck to the insect paper based on an image of the insect paper and count each type of identified pest. Accepting operations;
accepting an operation by the user to register information on a section that corresponds to the identification implementation range among sections set in advance on the insect trapping paper;
Inputting an image of the insect trapping paper into the inspection model to perform the identification and the counting;
Calculating a second counting result, which is a result of counting using the partition as a unit, from a first counting result, which is a counting result in the implementation range of the identification using the inspection model, based on information on the partition. ,
Presenting at least the second counting result to a user;
A pest inspection program that runs on a computer.
The program for inspecting pests according to claim 1, wherein the information on the sections includes the number of sections included in the range of implementation of the identification with respect to the total number of sections set on the insect trapping paper.
3. The program for pest inspection according to claim 2, further causing the computer to accept an operation for modifying at least one of the identified pest type and the first counting result by a user.
The pest according to claim 3, further comprising causing the computer to recalculate the second counting result in accordance with at least one of the identified pest type and the first counting result. Program for inspection of.
For inspection of pests according to any one of claims 1 to 4, further comprising causing the computer to create a report including the identified pest type and the second counting result. program.
Registration by the user of the scope of implementation of the identification in an inspection model configured to identify at least the type of pest stuck to the insect paper based on an image of the insect paper and count each type of identified pest. a first reception section that receives operations;
a second reception unit that receives an operation by the user to register information on a section that corresponds to the identification implementation range among the sections set in advance on the insect trapping paper;
an inspection unit that inputs the image of the insect trapping paper into the inspection model to perform the identification and the counting;
a calculation unit that calculates a second counting result, which is a result of counting using the section as a unit, from a first counting result, which is the result of counting in the range of implementation of the identification using the inspection model, based on information on the section; and,
a presentation unit that presents at least the second counting result to a user;
A server equipped with.