JP2023068234A - Excrement image display system and toilet bowl - Google Patents

Abstract

To provide an excrement image display system, wherein the system enables accurate detection of dirt by distinguishing between the dirt left in a toilet bowl and actual excrement, especially a small amount of excrement, that was difficult with conventional image recognition algorithms for excretion, and in addition, the system also enables a caregiver to make an accurate determination of medication for a care-receiver by misdetecting and notifying that there was a bowel movement when the toilet bowl was dirty.SOLUTION: An excrement image display system of the present disclosure can accurately determine a nature of excrement by using a determination algorithm that takes into consideration toilet bowl stains and lens cover stains. Alternatively, the system encourages cleaning of a toilet bowl and enables accurate detection of excrement properties by notifying a user (e.g. care-receiver) stool viewer (e.g. caregiver) that the toilet bowl and imaging device lens are dirty.SELECTED DRAWING: Figure 10

Description

The present disclosure relates to an excrement image display system and a toilet bowl.

As disclosed in Patent Literature 1, by inputting the image information into a model learned by machine learning using a neural network for a target image of the inside of a toilet bowl during excretion, an imaging device or an imaging environment causes an image. Determination devices exist that determine whether dirt is being imaged.

JP 2020-190181 A

In conventional image recognition algorithms for excretion, it is difficult to accurately detect dirt by distinguishing between the dirt remaining in the toilet bowl and actual bowel movements, especially a small amount of bowel movements. Further, in a care facility or the like, erroneously detecting that a toilet bowl has been soiled and notifying that there has been defecation may cause a caregiver to make a mistake in determining medication for a care-receiver.

The excrement image display system according to the present disclosure includes:
An excrement image display system for displaying the state of excrement on a display unit,
The excrement image display system includes:
toilet bowl,
a server, and a user-side information processing device comprising the display unit,
here,
The toilet comprises a toilet seat, an imaging device, and a toilet-side information processing device,
The imaging device comprises a light receiving unit, a lens, and a lens cover that protects the lens,
The toilet-side information processing device comprises a toilet-side data transmission unit and a toilet-side control unit,
the server comprises a server-side receiving unit, a server-side transmitting unit, a server-side storage unit, and a server-side control unit;
The server-side storage unit
(I) an excrement learning model; and (II) at least one dirt learning model selected from the group consisting of a toilet bowl dirt learning model and a lens cover dirt learning model;
and the user-side information processing device further comprises a user-side input unit, a user-side transmission unit, a user-side reception unit, a user-side storage unit, and a user-side control unit,
In the excrement image display system,
The toilet bowl-side control unit controls the imaging element so that the imaging element images the inside of the toilet bowl,
the toilet-side control unit controls the toilet-side data transmission unit so as to transmit the captured image captured by the imaging device to the server as captured image information;
The server-side control unit controls the server-side reception unit so as to receive the captured image information,
Based on the excrement learning model, the server-side control unit estimates the excrement information as excrement estimation information from at least one captured image included in the received captured image information,
Based on the at least one dirt learning model, the server-side control unit selects from a group consisting of information on toilet bowl dirt adhering to the inner wall surface of the toilet bowl and information on lens cover dirt adhering to the outer peripheral surface of the lens cover. estimating at least one information to be obtained from at least one captured image included in the received captured image information as dirt estimation information;
the server-side control unit controls the server-side transmission unit so as to transmit at least one data selected from the group consisting of first data and second data to the user-side reception unit;
here,
the first data includes the excrement estimation information and the at least one captured image;
the second data includes the dirt estimation information and the at least one captured image;
the user-side controller controls the user-side receiver to receive the at least one data;
The user-side control unit controls the display unit so as to display the at least one data received by the user-side reception unit on the display unit.

According to the excrement image display system of the present disclosure, it is possible to accurately determine the nature of excrement by using a determination algorithm that considers at least one stain selected from the group consisting of toilet bowl stains and lens cover stains. . Alternatively, by notifying the viewer (e.g. caregiver) of the stool of the user (e.g. care recipient) that the toilet bowl is dirty, it is possible to encourage cleaning inside the toilet bowl and more accurately detect the nature of the excrement. make it

Diagram of excretion device according to Embodiment 1 Sectional view of the toilet seat in Embodiment 1 Block diagram including toilet seat 2 and server 101 included in excrement image display system according to Embodiment 1 Block diagram including toilet seat 2 and server 101 included in excrement image display system according to Embodiment 1 Block diagram of the system according to Embodiment 1 Diagram of user-side information processing device 201 according to Embodiment 1 Diagram of user-side information processing device 201 according to Embodiment 1 Diagram of user-side information processing device 201 according to Embodiment 1 Diagram of user-side information processing device 201 according to Embodiment 1 Diagram of user-side information processing device 201 according to Embodiment 1 Diagram of user-side information processing device 201 according to Embodiment 1 Flowchart in Embodiment 1 Flowchart in Embodiment 1 Schematic diagram of mask image in Embodiment 1 Schematic diagram of a captured image from which dirt has been removed by image processing according to Embodiment 1. FIG. Schematic diagram of captured image information in Embodiment 1 Schematic diagram of captured image information in Embodiment 1 Schematic diagram of captured image information in Embodiment 1

(Embodiment 1)
1A is a diagram of an excretion device according to Embodiment 1. FIG.

FIG. 2A is a block diagram including toilet seat 2 and server 101 included in the excrement image display system (hereinafter simply referred to as "system") according to Embodiment 1. FIG.

(Description of overview)
As shown in FIGS. 1A and 2A, a toilet bowl 1, which is a device for excreting, has a toilet bowl, a toilet seat 2, an imaging device 4, and a toilet bowl-side information processing device 5. FIG. A device-specific ID (that is, an ID specific to the toilet bowl 1) is assigned to the toilet bowl-side information processing device 5. FIG. Also, the toilet seat 2 is provided with a seating sensor 3 for detecting that the user is seated on the toilet seat 2 . The toilet seat 2 has an imaging element 4 for imaging the inside of the toilet bowl. An example of the imaging element 4 is a camera. An example of the toilet-side information processing device 5 is a semiconductor element (that is, an electric/electronic circuit) such as an IC or LSI.

As shown in FIG. 1B, the imaging element 4 is arranged inside the toilet seat 2 . The imaging device 4 includes a light receiving portion (not shown) made up of an image sensor, a lens 42 and a lens cover 43 . A lens cover 43 is provided to protect the lens 42 . The imaging element 4 is provided inside the toilet seat 2 . The lens cover 43 forms part of the lower surface of the toilet seat 2 . Light that has passed through the lens cover 43 and reaches the lens 42 is collected by the lens 42 into a light receiving section (not shown) and converted into electronic image information. In this way, the inside of the toilet bowl (strictly speaking, the inner peripheral surface of the toilet bowl) is photographed, and as a result, the toilet bowl-side information processing device 5 forms a captured image 502 of the inside of the toilet bowl.

In the system according to Embodiment 1, the image of the user's (eg, care recipient's) excrement and the excrement information are displayed on the display unit 202, which will be described later. Alternatively, in the system according to Embodiment 1, (a) an image of excrement of a user (eg, a care recipient) and (b) an image of at least one dirt selected from the group consisting of toilet bowl dirt and lens cover dirt. , is displayed on the display unit 202 .

As shown in FIG. 3, the system according to Embodiment 1 comprises a toilet bowl 1, a server 101, and a user-side information processing device 201. As shown in FIG.

As described above, the toilet bowl 1 includes the toilet seat 2 , the imaging device 4 , the seating sensor 3 , and the toilet bowl side information processing device 5 . The toilet-side information processing device 5 includes a toilet-side data transmission section and a toilet-side control section. The toilet bowl-side information processing device 5 may include a toilet bowl-side storage unit.

The server 101 includes a server-side reception section, a server-side transmission section, a server-side storage section, and a server-side control section, like a server that is normally used. As will be described later, the server-side storage unit stores excrement learning models.

Examples of the server 101 are general servers on the Internet, servers in nursing homes, and servers in one's home.

As shown in FIG. 3, the user-side information processing device 201 has a display section 202 . The user-side information processing device 201 further includes a user-side input unit, a user-side transmission unit, a user-side reception unit, a user-side storage unit, and a user-side control unit. Examples of the user-side information processing device 201 are a personal computer, tablet, and smart phone having a display functioning as the display unit 202 .

The storage unit is a so-called memory (especially a non-volatile memory among them). The control unit is a so-called central processing unit.

(Description of operation of toilet bowl 1)
As indicated by step 101 (that is, S101) in FIG. 10, when it is detected that the user is seated on the toilet seat 2, the imaging device 4 starts imaging the interior of the toilet. In this manner, the toilet bowl-side control unit controls the imaging device 4 so that the imaging device 4 captures an image of the interior of the toilet bowl 1 (that is, the interior of the toilet bowl).

Upon receiving a seating signal from the seating sensor 3, the toilet-side information processing device 5 placed inside the toilet seat activates the imaging element 4, and continues imaging while the user is seated on the toilet seat. The imaging interval is set to every designated interval, here, every second, but it may be longer or shorter. The captured image is stored in the toilet-side information processing device 5 (strictly speaking, if a toilet-side storage unit is provided, the toilet-side storage unit), and the image name includes the device ID associated with the toilet seat where the image was captured. The shooting time is recorded. Also, an image that is saved first after the start of imaging is saved with an image name to which a text is added so that the fact can be determined.

When it is detected that the user has left the toilet seat 2, as indicated by step 102 (that is, S102) in FIG. , if a toilet-side storage unit is provided, the data is uploaded to the server 101 from the toilet-side storage unit). In this manner, the toilet-side control unit controls the toilet-side data transmission unit so that the captured image captured by the image sensor 4 is transmitted to the server 101 as captured image information.

As described above, in the system according to Embodiment 1, first, as indicated by S101 in FIG. The toilet-side controller controls the imaging element 4 so as to start imaging.

Next, a plurality of captured images 502 (see FIGS. 13A to 13C) captured by the imaging element 4 at predetermined time intervals are stored as captured image information 501 (see FIGS. 13A to 13C) in the toilet storage unit. Secondly, the toilet bowl-side controller controls the toilet bowl-side storage unit. In other words, captured image information 501 is formed in the toilet storage unit by sequentially accumulating captured images 502 from the imaging device 4 in the toilet storage unit over time. As shown in FIGS. 13A-13C, the captured image information 501 includes a plurality of captured images 502 and text 503 for each of the plurality of captured images 502 . The text 503 includes the device ID and the imaging time.

Next, as indicated by S102 in FIG. 10, the toilet-side control unit controls the toilet-side data transmission unit so as to transmit the captured image information 501 to the server 101. FIG. More specifically, in S102, the toilet-side control unit controls the toilet-side data transmission unit so that the captured image 502 captured by the image sensor 4 is transmitted to the server 101 as the captured image information 501, and the captured image information 501 is transmitted to the server 101. The server-side control unit controls the server-side receiving unit so as to receive the .

Finally, when the seating sensor 3 detects that the user leaves the toilet seat 2 , the toilet bowl side controller controls the imaging device 4 so that the imaging device 4 stops imaging the inside of the toilet bowl 1 .

Note that the toilet bowl-side information processing device 5 does not have to include the toilet bowl-side storage unit. That is, while the imaging element 4 is imaging the inside of the toilet bowl 1 (strictly speaking, the inside of the toilet bowl), the toilet bowl side information processing device 5
Alternatively, the captured image 502 captured by the image sensor 4 may be immediately transmitted to the server 101 . In this case, the server-side control unit generates captured image information 501 and stores the generated captured image information 501 in the server-side storage unit.

In the present embodiment, each captured image 502 is a soil attached to the inner wall surface of the toilet bowl (more precisely, the toilet bowl) (especially feces attached to the inner peripheral surface of the toilet bowl, especially the inner peripheral surface of the toilet bowl). An image of stool that adheres to the surface and cannot be removed by flushing water) and an image of dirt adhering to the outer peripheral surface of the lens cover 43 may also be included. Hereinafter, in the present specification, dirt adhering to the inner wall surface of the toilet bowl (to be more precise, the toilet bowl) is simply referred to as "toilet dirt." Similarly, dirt adhering to the outer peripheral surface of the lens cover 43 is simply referred to as "lens cover dirt". The term "dirt" as used herein includes at least one selected from the group consisting of toilet bowl dirt and lens cover dirt.

Since each captured image 502 can also include an image of at least one stain selected from the group consisting of toilet bowl stains and lens cover stains, naturally the captured image information 501 is also selected from the group consisting of toilet bowl stains and lens cover stains. It may also include an image of at least one stain that is painted. This is because the captured image information 501 includes a plurality of captured images 502, as is clear from the above description.

(Description of Image Recognition in Server 101)
FIG. 3 is a block diagram of the system according to Embodiment 1. FIG.

In FIG. 3, the server 101 has an input image, a learning section 102, a recognition section 103, and a database 110. FIG.

The user-side information processing device 201 has a display unit 202 and an erroneous determination correcting unit 203 .

The learning unit 102 takes in a large number of excrement images in advance, extracts features from these images, and generates an excrement learning model that can respond to the features. The generated excrement learning model is stored in the server-side storage unit as an excrement learning model.

Similarly, the learning unit 102 preliminarily detects dirt adhering to the inner wall surface of the toilet bowl (more precisely, the toilet bowl) (especially feces adhering to the inner peripheral surface of the toilet bowl, especially A large number of images of stool that cannot be removed by pouring water on the toilet are captured, features are extracted from those images, and a toilet bowl dirt learning model that can respond to the features is generated. The generated toilet soiling learning model is stored in the server-side storage unit as a toilet soiling learning model.

Similarly, the learning unit 102 preliminarily acquires a large number of images of dirt adhering to the outer peripheral surface of the lens cover 43 (that is, images captured by the imaging element 4 when the outer peripheral surface of the lens cover 43 is stained). , performs feature extraction from those images and generates a lens cover dirt learning model that can react to the features. The generated lens cover dirt learning model is stored in the server-side storage unit as a lens cover dirt learning model.

Hereinafter, the term "at least one dirt learning model" means at least one selected from the group consisting of a toilet bowl learning model and a lens cover dirt learning model.

The recognition unit 103 has a recognition algorithm equipped with an excrement learning model and at least one dirt learning model. The recognition unit 103 recognizes excrement and dirt from the input image of excrement (that is, captured image information 501, more specifically, a plurality of captured images 502 included in the captured image information 501), Output the recognition result.

The database 110 stores recognition results (eg presence/absence of excrement, shape, amount, color, number of drops, toilet stains, lens cover stains, etc.). A database 110 is included in the server-side storage.

The captured image information 501 received by the server-side receiving unit is temporarily stored in the server-side storage unit. Next, the recognizing unit 103 determines the excrement properties (for example, based on the captured image information 501 (specifically, at least one captured image 502 included in the captured image information 501) (for example, from 7 classifications of the Bristol scale index to be described later). Any one stool property selected from the group of stools) is recognized, and the result is output from the recognition unit 103 . The output results are stored in the database 110 (that is, the server-side storage unit) as excrement estimation information.

Similarly, the recognition unit 103 recognizes at least one dirt selected from the group consisting of toilet bowl dirt and lens cover dirt based on the captured image information 501, and the recognition unit 103 outputs the result. The output result is stored in the database 110 (that is, the server-side storage unit) as dirt estimation information.

As is clear from the above description, based on the excrement learning model, the server-side control unit converts excrement information into at least one image included in the received captured image information 501 as excrement estimation information. Data estimated from the captured image 502 and including excrement estimation information and the at least one captured image 502 (i.e., at least one captured image 502 that is the basis for estimating the excrement estimation information) (hereinafter referred to as "first data") is stored in the server-side storage section. This is indicated in FIG. 10 as step 201 (ie, S201) and step 202 (ie, S202).

Similarly, based on at least one soil learning model selected from the group consisting of the toilet bowl soiling learning model and the lens cover soiling learning model, the server-side control unit determines at least one soiling model selected from the group consisting of toilet bowl soiling and lens cover soiling. Information of one dirt is estimated from at least one captured image 502 included in the received captured image information 501 as dirt estimation information, and the dirt estimation information and the at least one captured image 502 (that is, dirt The server-side control unit controls the server-side storage unit so as to store data (hereinafter referred to as "second data") including at least one captured image 502 that is the basis for estimating the estimated information in the server-side storage unit. do. This is indicated in FIG. 10 as step 201 (ie, S201) and step 202 (ie, S202).

Finally, as indicated as step 203 (that is, S203) in FIG. The server-side control unit controls the server-side transmission unit so as to transmit from the storage unit to the user-side reception unit, and the user-side control unit controls the user-side reception unit so as to receive the first data.

Similarly, when the server-side control unit determines that the second data includes an image of at least one stain selected from the group consisting of toilet bowl stains and lens cover stains, the second data is stored in the server-side storage unit. The server-side control unit controls the server-side transmission unit so as to transmit the data to the user-side reception unit, and the user-side control unit controls the user-side reception unit so as to receive the second data.

(Description of dirt detection)
In FIG. 10, the image uploaded by the toilet bowl 1 in the server 101 is the text added to the image that is first saved after the start of imaging (for example, "13:05:
The text 503) represented as "00" is used to create a group of images classified for each excretion. The captured image information 501 is input to a recognition algorithm equipped with an excrement learning model capable of reacting to the characteristics of excrement and a dirt learning model capable of reacting to the characteristics of dirt. , the amount, color, number of drops, and the presence or absence of dirt. The output recognition result is stored in the database 110 .

If even one image of excrement is detected in the captured image 502 included in the captured image information 501, the result of excretion detection and the captured image 502 that is the basis for the detection are selected as the first data. Then, the result and the captured image 502 are displayed on the information processing device 201 on the user side.

Similarly, when even one image of dirt is detected in the captured image 502 included in the captured image information 501, the result of the detection of dirt (that is, estimated toilet dirt) and the basis thereof The captured image 502 is selected as the second data, and the result and the captured image 502 are displayed on the user-side information processing device 201 .

Only when an image of excrement or dirt is detected in the plurality of captured images 502 included in the captured image information 501, the recognition result (that is, the result of detecting excrement or the result of detecting dirt) and It is also possible to select the captured image 502 that serves as the basis for the result and display the result and the captured image 502 on the user-side information processing device 201 .

Also, when selecting a plurality of recognition results and images, all the results and images may be displayed on the user-side information processing device 201, or only representative results may be displayed. The representative result is an image with high reliability, which is a parameter of the recognition result (that is, an image with a high probability of being excrement when excrement is detected), an image with little blur, and an image with the largest object area. For example, various parameters may be used for selection.

As shown in FIG. 10, the system according to this embodiment has one learning model, and the one learning model includes an excrement learning model and a dirt learning model. In other words, the excrement image display system has one learning model, and the one learning model has the excrement learning model. Further, the one learning model has at least one dirt learning model selected from the group consisting of a toilet bowl dirt learning model and a lens cover dirt learning model.

Alternatively, as shown in FIG. 11, the system according to this embodiment may have two learning models corresponding to the waste learning model and the dirt learning model. In other words, the excrement image display system has two or more learning models, one of the learning models included in the two or more learning models is the excrement learning model, and is included in the two or more learning models. Another one of the learning models may be at least one dirt learning model selected from the group consisting of a toilet bowl dirt learning model and a lens cover dirt learning model.

In any case, the learning model is stored in the server-side storage unit.

Further, as shown in FIG. 12A, in an algorithm that inputs captured image information 501 and outputs the presence or absence of dirt as a recognition result, an image without dirt stored in advance and an image immediately after the start of shooting in the captured image information 501 and at least one image selected from the group consisting of the image immediately after washing with water, extracting the soiled area 603 in the image by image processing, and extracting the soiled area 603 in the image as shown in FIG. 12A. A mask image 604 obtained by binarizing (that is, drawing only in black and white) the dirt region 603 in the water puddle region 602 of 601 is created, and the dirt region 603 included in the mask image 604 is If the number of pixels in the corresponding white portion is equal to or greater than a specified threshold, it may be determined that there is dirt.

Note that the mask image 604 may be created as a mask image obtained by extracting an excrement region using an algorithm that outputs excretion information as a recognition result.

The mask image 604 may also be used in an algorithm for outputting excretion information as a recognition result to improve the recognition accuracy of excretion information by deleting the dirty area 603 by image processing, as shown in FIG. 12B. .

When the mask image 604 is displayed on the user-side information processing device 201, at least one region selected from the group consisting of the excrement region and the dirt region 603 is mosaiced by image processing, or a specific image is displayed. It may be displayed by replacing it with another image such as a texture. Accordingly, even a user who does not want to directly view at least one selected from the group consisting of excrement and dirt can be encouraged to view the captured image 502 .

(Description of display on user-side information processing device 201)
As described above, the user can display the excretion record on the display unit 202 by entering the ID and password on the web or application.

FIG. 4 shows an example of an excretion record illustrated on the display unit 202. As shown in FIG. As shown in FIG. 4, the time when the person sat on the toilet seat 2, when there was excretion, whether the excrement was stool (that is, stool) or urine (that is, urine), and if it was stool, the shape at that time were determined. The classification, color of stool, amount of stool, and number of drops of stool among the seven Bristol scale index categories are illustrated on a graph 205 . If it is urine, the excretion record is displayed on the display unit 202 as illustrated on the graph 205 with urine color and urine volume.

The graph 205 is composed of a three-stage graph with the time from 0:00:00 to 23:59:59 on the horizontal axis. In the upper part, the time spent sitting on the toilet seat from the imaging time indicated in the image name is displayed in rectangular waves.

In the middle row, the imaging time recorded in the image determined as urine by the presence/absence detection of feces and urine is read, and a bar-shaped figure is displayed on the graph. At this time, the figure may be displayed in the same color as the urine color determination result, and the height may be changed according to the amount of urine. Alternatively, as shown in FIG. Alternatively, text may be displayed in the form of annotations near the figure, as shown in FIG.

Furthermore, in the lower part, the imaging time described in the image determined as stool by feces/urine presence/absence detection is read, and a bar-shaped figure is displayed on the graph. At this time, the graphic may display the stool shape recognition determination result as text in the graphic, or may display a graphic in which the Bristol scale index is iconized in the shape of stool. The figure may be displayed in the same color as the stool color determination result, and the height may be changed according to the amount of stool. Text may be displayed in the form of annotations near the figure.

Further, as shown in FIG. 7, the user-side information processing apparatus 201 according to the present embodiment provides a basis for the determination result by clicking or touching a graphic representing the determination result displayed on the graph 205. It is possible to display the image that was obtained (that is, one captured image 502 that is the basis of the determination result). Note that the image may be displayed on the graph 205 from the beginning. The image displayed at this time may be displayed in a form in which a window separate from the graph pops up as an image display area 206, or as shown in FIG. may be displayed in

In this manner, in the excrement image display system according to Embodiment 1, when the excrement recognition result (that is, excrement estimation information) is displayed on the display unit 202, the user (eg, the care recipient) A viewer of excrement (eg, a caregiver) can easily recognize the presence or absence of the user's excrement, and as a result, can easily grasp the user's health condition.

Similarly, in the excreta image display system according to Embodiment 1, the stain recognition result (that is, at least one stain estimation information selected from the group consisting of toilet bowl stain information and lens cover stain information) is displayed on display unit 202. , a viewer (eg, a caregiver) can easily recognize the presence or absence of dirt, and as a result, is encouraged to clean the toilet bowl.

As shown in FIG. 2B, toilet-side information processing device 5 may have recognition unit 103 . In other words, the toilet-side information processing device 5 may also serve as the server 101 .

The toilet seat 2 shown in FIG. 2B causes data to be displayed on the display of the user information processing device.

The toilet seat 2 has an imaging element 4 and a toilet-side information processing device 5 .

In the toilet seat 2 shown in FIG. 2B, the toilet information processing device 5 includes a toilet seat transmission section, a toilet seat storage section, and a toilet seat control section. The toilet seat storage unit stores at least one dirt learning model selected from the group consisting of (I) an excrement learning model and (II) a toilet dirt learning model and a lens cover dirt learning model. On the other hand, the user-side information processing device further includes a user-side receiving section.

The toilet information processing device 5 included in the toilet seat 2 shown in FIG. 2B performs the following steps during operation.

First, the toilet seat control unit controls the image pickup device 4 so that the image pickup device 4 picks up an image of the inside of the toilet, thereby generating captured image information 501 .

Next, based on the excrement learning model, the toilet seat control unit estimates excrement information from at least one captured image 502 included in the captured image information 501 as excrement estimation information.

Further, based on at least one dirt learning model, the toilet seat control unit is selected from the group consisting of information on toilet bowl dirt adhering to the inner wall surface of the toilet bowl and information on lens cover dirt adhering to the outer peripheral surface of the lens cover. At least one piece of information is estimated from at least one captured image 502 included in the captured image information 501 as dirt estimation information.

Finally, the server-side controller controls the server-side transmitter to transmit at least one data selected from the group consisting of the first data and the second data to the user-side receiver. Similar to the above, the first data includes excrement estimation information and at least one captured image 502 (that is, at least one captured image 502 that is the basis for estimating the excrement estimation information), and the second data is It includes dirt estimation information and at least one captured image 502 (that is, at least one captured image 502 on which the dirt estimation information is estimated).

(Explanation of correction of judgment result)
In this embodiment, the determination result displayed on the display unit 202 can be corrected. As shown in FIG. 9, a determination result display box 207 in which determination result text is input is provided, and a pull-down box 208 is provided beside the determination result display box 207 . A list of judgment results is displayed by clicking or touching the pull-down box 208, and an erroneous judgment correcting unit corrects the judgment result displayed in the judgment result display box 207 by clicking or touching the judgment result to be confirmed by the user. 203 of the user-side information processing apparatus 201 of this embodiment. At this time, the figure representing the determination result displayed on the graph is also redrawn according to the content of the correction.

(Description of the hardware of the control unit)
The system of this embodiment is a computer system. A computer system is mainly composed of a processor (that is, a central processing unit that functions as a control unit) and a memory (that is, a storage unit) as hardware. In the server 101 and the user-side information processing device, the present system operates by executing a program recorded in a memory (ie, a storage unit) by a processor (ie, a central processing unit that functions as a control unit). Also in the toilet bowl side information processing device included in the toilet bowl, this system operates by executing the program recorded in the memory (that is, the storage unit) by the processor (that is, the central processing unit that functions as the control unit). .

The program may be pre-recorded in memory, may be provided through an electric communication line, or may be provided by being recorded in a non-temporary recording medium such as a computer system-readable memory card, optical disk, or hard disk drive. good too. A processor consists of one or more electronic circuits, including semiconductor integrated circuits (ICs) or large scale integrated circuits (LSIs). The integrated circuit such as IC or LSI referred to here is called differently depending on the degree of integration, and includes integrated circuits called system LSI, VLSI (Very Large Scale Integration), or ULSI (Ultra Large Scale Integration).

Furthermore, an FPGA (Field-Programmable Gate Array), which is programmed after the LSI is manufactured, or a logic device capable of reconfiguring the connection relationship inside the LSI or reconfiguring the circuit partitions inside the LSI, may also be adopted as the processor. can be done. A plurality of electronic circuits may be integrated into one chip, or may be distributed over a plurality of chips. A plurality of chips may be integrated in one device, or may be distributed in a plurality of devices. A computer system, as used herein, includes a microcontroller having one or more processors and one or more memories. Accordingly, the microcontroller also consists of one or more electronic circuits including semiconductor integrated circuits or large scale integrated circuits.

The excrement image display system of the present disclosure accurately determines the nature of excrement by using a determination algorithm that considers at least one dirt selected from the group consisting of toilet bowl dirt and lens cover dirt. The system prompts the toilet bowl to be cleaned by notifying the viewer (eg, nursing staff) of excrement of the toilet user (eg, care recipient) that the toilet bowl is dirty. As a result of cleaning, the nature of excreta is estimated more accurately.

A program that causes a computer to execute the excrement image display system of the present disclosure is also included in the content of the present disclosure.

1 toilet bowl 2 toilet seat 3 seating sensor 4 imaging device 42 lens 43 lens cover 5 toilet bowl side information processing device 101 server 102 learning unit 103 recognition unit 110 database 201 user side information processing device 202 display unit 205 graph 206 image display area 207 determination result display Box 208 Pull-down box 501 Captured image information 502 Captured image 503 Text 601 Toilet area 602 Water pool area 603 Dirt area 604 Mask image

Claims (15)

An excrement image display system for displaying the state of excrement on a display unit,
The excrement image display system includes:
toilet bowl,
a server, and a user-side information processing device comprising the display unit,
here,
The toilet comprises a toilet seat, an imaging device, and a toilet-side information processing device,
The imaging device comprises a light receiving unit, a lens, and a lens cover that protects the lens,
The toilet-side information processing device comprises a toilet-side data transmission unit and a toilet-side control unit,
the server comprises a server-side receiving unit, a server-side transmitting unit, a server-side storage unit, and a server-side control unit;
The server-side storage unit
(I) an excrement learning model; and (II) at least one dirt learning model selected from the group consisting of a toilet bowl dirt learning model and a lens cover dirt learning model;
and the user-side information processing device further comprises a user-side input unit, a user-side transmission unit, a user-side reception unit, a user-side storage unit, and a user-side control unit,
In the excrement image display system,
The toilet bowl-side control unit controls the imaging element so that the imaging element images the inside of the toilet bowl,
the toilet-side control unit controls the toilet-side data transmission unit so as to transmit the captured image captured by the imaging device to the server as captured image information;
The server-side control unit controls the server-side reception unit so as to receive the captured image information,
Based on the excrement learning model, the server-side control unit estimates the excrement information as excrement estimation information from at least one captured image included in the received captured image information,
Based on the at least one dirt learning model, the server-side control unit selects from a group consisting of information on toilet bowl dirt adhering to the inner wall surface of the toilet bowl and information on lens cover dirt adhering to the outer peripheral surface of the lens cover. estimating at least one information to be obtained from at least one captured image included in the received captured image information as dirt estimation information;
the server-side control unit controls the server-side transmission unit so as to transmit at least one data selected from the group consisting of first data and second data to the user-side reception unit;
here,
the first data includes the excrement estimation information and the at least one captured image;
the second data includes the dirt estimation information and the at least one captured image;
the user-side controller controls the user-side receiver to receive the at least one data;
The excrement image display system, wherein the user-side control unit controls the display unit so as to display the at least one data received by the user-side reception unit on the display unit. The at least one dirt learning model includes the toilet dirt learning model, and the dirt estimation information is information about toilet dirt adhering to the inner wall surface of the toilet.
The excrement image display system according to claim 1. the at least one dirt learning model includes the lens cover dirt learning model, and the dirt estimation information is information about lens cover dirt adhering to the outer peripheral surface of the lens cover,
The excrement image display system according to claim 1. When the at least one data received by the user-side receiving unit is displayed on the display unit, the displayed image includes an area in which the image of excrement and the image of dirt contained in the image are captured. Mosaic is applied by image processing,
The excrement image display system according to claim 1. When the at least one data received by the user-side receiving section is displayed on the display section, the displayed image includes both the image of excrement and the image of dirt contained in the image. replaced by an image that does not contain any objects or dirt,
The excrement image display system according to claim 1. A toilet seat for displaying data received by a user-side receiving unit included in a user-side information processing device on a display unit included in the user-side information processing device,
The toilet seat includes an imaging device and an information processing device,
The imaging device comprises a light receiving unit, a lens, and a lens cover that protects the lens,
The information processing device includes a toilet seat-side transmission unit, a toilet seat-side storage unit, and a toilet seat-side control unit,
The toilet seat side storage unit
(I) an excrement learning model; and (II) at least one dirt learning model selected from the group consisting of a toilet bowl dirt learning model and a lens cover dirt learning model;
and the user-side information processing device further comprises a user-side receiving unit,
In the information processing device, during operation,
The toilet seat-side control unit controls the image pickup device so that the image pickup device picks up an image of the inside of the toilet, whereby the toilet seat-side control unit generates the captured image information,
Based on the excrement learning model, the toilet seat control unit estimates excrement information as excrement estimation information from at least one captured image included in the captured image information,
Based on the at least one dirt learning model, the toilet seat-side control unit selects from a group consisting of information on toilet bowl dirt adhering to the inner wall surface of the toilet bowl and information on lens cover dirt adhering to the outer peripheral surface of the lens cover. estimating at least one information to be obtained from at least one captured image included in the captured image information as dirt estimation information;
the server-side control unit controls the server-side transmission unit so as to transmit at least one data selected from the group consisting of first data and second data to the user-side reception unit;
here,
the first data includes the excrement estimation information and the at least one captured image;
the second data includes the dirt estimation information and the at least one captured image;
toilet seat. The at least one dirt learning model includes the toilet dirt learning model, and the dirt estimation information is information about toilet dirt adhering to the inner wall surface of the toilet.
A toilet seat according to claim 6. the at least one dirt learning model includes the lens cover dirt learning model, and the dirt estimation information is information about lens cover dirt adhering to the outer peripheral surface of the lens cover,
A toilet seat according to claim 6. When the at least one data received by the user-side receiving unit is displayed on the display unit, the displayed image includes an area in which the image of excrement and the image of dirt contained in the image are captured. Mosaic is applied by image processing,
A toilet seat according to claim 6. When the at least one data received by the user-side receiving section is displayed on the display section, the displayed image includes both the image of excrement and the image of dirt contained in the image. replaced by an image that does not contain any objects or dirt,
A toilet seat according to claim 6. A program for operating an excrement image display system that displays the state of excrement on a display unit,
The excrement image display system includes:
toilet bowl,
a server, and a user-side information processing device comprising the display unit,
here,
The toilet comprises a toilet seat, an imaging device, and a toilet-side information processing device,
The imaging device comprises a light receiving unit, a lens, and a lens cover that protects the lens,
The toilet-side information processing device comprises a toilet-side data transmission unit and a toilet-side control unit,
the server comprises a server-side receiving unit, a server-side transmitting unit, a server-side storage unit, and a server-side control unit;
The server-side storage unit
(I) an excrement learning model; and (II) at least one dirt learning model selected from the group consisting of a toilet bowl dirt learning model and a lens cover dirt learning model;
and the user-side information processing device further comprises a user-side input unit, a user-side transmission unit, a user-side reception unit, a user-side storage unit, and a user-side control unit,
Said program
The toilet bowl-side control unit controls the imaging element so that the imaging element images the inside of the toilet bowl,
The toilet-side control unit controls the toilet-side data transmission unit so that the captured image captured by the imaging device is transmitted to the server as captured image information,
The server-side control unit controls the server-side reception unit so as to receive the captured image information,
Based on the excrement learning model, the server-side control unit estimates the excrement information as excrement estimation information from at least one captured image included in the received captured image information,
Based on the at least one dirt learning model, the server-side control unit selects from a group consisting of information on toilet bowl dirt adhering to the inner wall surface of the toilet bowl and information on lens cover dirt adhering to the outer peripheral surface of the lens cover. estimating at least one information to be obtained from at least one captured image included in the received captured image information as dirt estimation information;
the server-side control unit controls the server-side transmission unit so as to transmit at least one data selected from the group consisting of first data and second data to the user-side reception unit;
here,
the first data includes the excrement estimation information and the at least one captured image;
the second data includes the dirt estimation information and the at least one captured image;
the user-side controller controls the user-side receiver to receive the at least one data;
The user-side control unit controls the display unit such that the at least one data received by the user-side reception unit is displayed on the display unit. The at least one dirt learning model includes the toilet dirt learning model, and the dirt estimation information is information about toilet dirt adhering to the inner wall surface of the toilet.
12. A program according to claim 11. the at least one dirt learning model includes the lens cover dirt learning model, and the dirt estimation information is information about lens cover dirt adhering to the outer peripheral surface of the lens cover,
A program according to claim 1. When the at least one data received by the user-side receiving unit is displayed on the display unit, the displayed image includes an area in which the image of excrement and the image of dirt contained in the image are captured. Mosaic is applied by image processing,
A program according to claim 1. When the at least one data received by the user-side receiving section is displayed on the display section, the displayed image includes both the image of excrement and the image of dirt contained in the image. replaced by an image that does not contain any objects or dirt,
A program according to claim 1.

Images