Measurement data management device, measurement data management method, program, and measurement system

Abstract

The present disclosure provides a measurement system (10) capable of automatically identifying a user. One aspect of the present disclosure relates to a measurement system (10) comprising a load measurement device (50) and a measurement data management device (100), wherein: the load measurement device (100) measures, at at least three measurement positions, a load to a platform which a user steps on, and transmits, to the measurement data management device (100), load data that has been generated on the basis of the measured load; and the measurement data management device (100) generates, on the basis of the load data, first time-series data of barycentric coordinates of the load for a period during which the user is on the platform, and associates and stores the first time-series data with identification information of the user.

Description

[Technical Field]

[0001] The present disclosure relates to a measurement data management device, a measurement data management method, a program, and a measurement system. [Background technology]

[0002] With the recent advances in information and communication technology, it is now widely used in various technical fields. For example, information and communication technology is not only applied to information processing devices such as personal computers (PCs), smartphones, and tablets, but is also being used to enhance the functionality of household appliances and other home devices.

[0003] For example, Patent Documents 1 to 3 describe scales with additional functions such as a balance evaluation function. Patent Documents 1 to 3 disclose scales that not only measure the weight of a user, but also measure the balance state of the user based on center of gravity data and fluctuation data of the user's load at multiple positions on the platform.

[0004] Furthermore, with the spread of IoT (Internet of Things) technology, IoT technology is being used not only in industrial equipment but also in home appliances. For example, home appliances can be connected to information processing terminals such as users' personal computers (PCs), smartphones, and tablets via wired or wireless connections to transmit various data to the information processing terminals or receive various control data for remotely operating the home appliances from the information processing terminals. By connecting the home appliances to the information processing terminals in this way, various data acquired by the home appliances can be stored in the information processing terminals, making it possible to apply more advanced information processing technologies such as AI (artificial intelligence) technology. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] Japanese Patent Application Laid-Open No. 2014-140640 [Patent Document 2] Japanese Patent Application Laid-Open No. 2012-61049 [Patent Document 3] Japanese Patent Application Laid-Open No. 2013-226225 Summary of the Invention [Problem to be solved by the invention]

[0006] For example, a household scale may be shared by a family member. When measurement data or weight data measured using the scale is recorded, it is desirable to be able to automatically identify which member of the family the measurement data or weight data belongs to.

[0007] One objective of the present disclosure is to provide a measurement system that can automatically identify a user. [Means for solving the problem]

[0008] One aspect of the present disclosure is a load measuring device and a measurement data management device, wherein the load measuring device measures a load on a platform on which a user stands at at least three measurement points, and transmits load data generated based on the measured load to the measurement data management device, and the measurement data management device generates first time series data of center of gravity coordinates of the load during a period when the user stands on the platform based on the load data, generating second time series data by rotating the first time series data by 180 degrees; The first time-series data is associated with the user's identification information. and the second time series data The present invention relates to a measurement system that stores the [Effects of the Invention]

[0009] According to the present disclosure, it is possible to provide a measurement system that can automatically identify a user. [Brief explanation of the drawings]

[0010] [Figure 1] FIG. 1 is a schematic diagram illustrating a measurement system according to one embodiment of the present disclosure. [Figure 2]FIG. 2 is a schematic diagram illustrating a measurement mechanism of a load measuring device according to an embodiment of the present disclosure. [Figure 3] 3A and 3B are diagrams illustrating center of gravity swing and weight swing according to one embodiment of the present disclosure. [Figure 4] FIG. 4 is a block diagram illustrating a hardware configuration of a measurement data managing device according to an embodiment of the present disclosure. [Figure 5] FIG. 5 is a block diagram illustrating a functional configuration of a measurement data managing device according to an embodiment of the present disclosure. [Figure 6] FIG. 6 is a diagram illustrating a state in which a user moves up and down relative to a load measuring device according to an embodiment of the present disclosure. [Figure 7] FIG. 7 is a schematic diagram illustrating different trajectories of the barycentric coordinates depending on the installation location according to an embodiment of the present disclosure. [Figure 8] 8A and 8B are schematic diagrams illustrating different trajectories of the center of gravity coordinates depending on the placement of a load measuring device according to one embodiment of the present disclosure. [Figure 9] 9A to 9D are diagrams illustrating trajectories of barycentric coordinate time-series data according to an embodiment of the present disclosure. [Figure 10] FIG. 10 is a schematic diagram illustrating the architecture of a machine learning model according to one embodiment of the present disclosure. [Figure 11] FIG. 11 is a schematic diagram illustrating a machine learning model that generates feature amount data from barycentric coordinate time-series data according to an embodiment of the present disclosure. [Figure 12] FIG. 12 is a schematic diagram illustrating a machine learning model that generates feature amount data from barycentric coordinate time-series data rotated by 180 degrees according to an embodiment of the present disclosure. [Figure 13] FIG. 13 is a flowchart illustrating a measurement data management process according to an embodiment of the present disclosure. DETAILED DESCRIPTION OF THE INVENTION

[0011] Hereinafter, embodiments of the present disclosure will be described with reference to the drawings.

[0012] In the following embodiment, a measurement system is disclosed that can automatically identify a user to be measured and manage measurement results in association with the user's identification information.

[0013] [Summary] A measurement system according to an embodiment of the present disclosure includes a load measuring device (e.g., a weight scale) that measures the weight of a user, and a measurement data management device (e.g., an information processing terminal such as a personal computer (PC), smartphone, or tablet) that is communicatively connected to the load measuring device and receives and stores the measurement results. The load measuring device continues to measure the load of the user on the platform at at least three measurement points from the time the user begins to place one foot on the platform until the time both feet leave the platform, and transmits load time-series data for the period while the user is on the platform to the measurement data management device.

[0014] When the load time series data is acquired from the load measuring device, the measurement data management device generates center of gravity coordinate time series data of the user's load based on the acquired load time series data. It is known that the tendency of a person's center of gravity coordinate to change when ascending or descending from a platform or the like has characteristics unique to each individual and can be used to identify the individual. The measurement data management device associates the center of gravity coordinate time series data with the user's identification information and stores it in advance. When the measurement data management device subsequently receives load time series data to be identified from the load measuring device, it can automatically identify the user based on the center of gravity coordinate time series data derived from the received load time series data.

[0015] This makes it possible for the measurement system according to this embodiment to automatically identify a user who stands on the load measuring device, and store load data indicating the weight or load of the user in association with the user's identification information.

[0016] [Measurement system] As shown in FIG. 1, the measurement system 10 includes a load measuring device 50 and a measurement data management device 100 that is connected to the load measuring device 50 for communication.

[0017] The load measuring device 50 is typically realized as a weighing scale, measures the load of the user while standing on the platform, and transmits load time series data indicating the measured load to the measurement data management device 100 as the measurement result.

[0018] 2, the load measuring device 50 may include load sensors 52_1, 52_2, 52_3, and 52_4 (hereinafter, collectively referred to as load sensors 52) at the four corners of a rectangular platform 51. The load sensors 52_1, 52_2, 52_3, and 52_4 measure the loads W1, W2, W3, and W4 [kg weight] of the user standing on the platform 51 at a predetermined sampling rate. The load measuring device 50 converts the measured load W1 ti ,W2 ti ,W3 ti ,W4 ti The load measuring device 50 transmits the load time-series data [kg weight] to the measurement data managing device 100. The load measuring device 50 may transmit the impedance values ​​measured by each load sensor 52 to the measurement data managing device 100 as the load time-series data.

[0019] For example, the load measuring device 50 may store time-series data (W1 t1 ,W2 t1 ,W3 t1 ,W4 t1 ),(W1 t2 ,W2 t2 ,W3 t2 ,W4 t2 ),···,(W1 tn ,W2 tn ,W3 tn ,W4 tn) to the measurement data managing device 100. Here, the predetermined sampling rate may be a predetermined time unit such as 20 Hz, and the data length may correspond to a predetermined duration such as 1 byte. Based on the received load time-series data, the measurement data managing device 100 can calculate the user's weight value as well as the user's center of gravity position (coordinates (x, y) in FIG. 2, etc.).

[0020] For example, the change in center of gravity coordinates during the period from when the user starts to stand on one foot on the platform 51 until both feet are stationary, i.e., the center of gravity fluctuation, can be represented as a trajectory as shown in Fig. 3A. Also, the change in body weight value during that period, i.e., the body weight fluctuation, can be represented as a trajectory as shown in Fig. 3B. It is known that the center of gravity fluctuation and body weight fluctuation can be used to identify individuals, and in the following embodiment, the measurement data management device 100 automatically identifies the user based on the center of gravity fluctuation and / or body weight fluctuation calculated from the load time-series data measured by the load measuring device 50, and stores the measured body weight value of the user in association with the user's identification information.

[0021] In the illustrated embodiment, the load measuring device 50 includes four load sensors 52_1, 52_2, 52_3, and 52_4, but the present disclosure is not necessarily limited to this. In order to determine the position of the center of gravity of the user, the load measuring device 50 only needs to include at least three load sensors 52. Furthermore, the platform 51 does not necessarily need to be table-shaped, and may have a flat shape such as a mat.

[0022] The measurement data managing device 100 is typically realized by an information processing terminal such as a personal computer (PC), a smartphone, or a tablet, and is capable of wired / wireless communication with the load measuring device 50. When the measurement data managing device 100 acquires the load time series data of the user from the load measuring device 50, it generates time series data of the center of gravity coordinate of the load during the period when the user is on the platform based on the acquired load time series data, and stores the time series data of the center of gravity coordinate in association with the user's identification information.

[0023] 1, in the measurement data managing device 100, center of gravity coordinate time series data #1 and #2 of user #1 and user #2 are initially registered in the user measurement data in association with user #1 and user #2, respectively. After the users are registered, when load time series data is acquired from the load measuring device 50, the measurement data managing device 100 generates center of gravity coordinate time series data from the acquired load time series data, and automatically identifies user #1 and user #2 based on the generated center of gravity coordinate time series data by referring to the registered center of gravity coordinate time series data #1 and #2 of user #1 and user #2. After identifying the users, the measurement data managing device 100 stores the measured weight values ​​in the user measurement data in association with the identified users.

[0024] Here, the determination of whether the barycentric coordinate time series data of the identification target matches the barycentric coordinate time series data registered in the user measurement data may be performed, for example, based on the feature quantities of the barycentric coordinate time series data output from a machine learning model. Specifically, the measurement data management device 100 may acquire the barycentric coordinate time series data of the identification target using a machine learning model trained to generate feature quantity data from the barycentric coordinate time series data, and determine whether the barycentric coordinate time series data of the identification target and the registered barycentric coordinate time series data belong to the same user based on the similarity between the feature quantity data of the registered barycentric coordinate time series data and the barycentric coordinate time series data.

[0025] In the above-described embodiment, the user is identified based on the center-of-gravity coordinate time series data. However, as will be described in detail below, the user may be identified based on the center-of-gravity coordinate time series data and the weight time series data. In the above-described embodiment, the load measuring device 50 transmits the measurement results of each load sensor 52 to the measurement data managing device 100. However, the present disclosure is not necessarily limited to this. For example, the load measuring device 50 may calculate the user's center-of-gravity coordinate time series data and / or the weight time series data based on the measurement results of each load sensor 52, and transmit the calculated center-of-gravity coordinate time series data and / or the weight time series data to the measurement data managing device 100 as the load time series data. Furthermore, the load measuring device 50 and the measurement data managing device 100 do not necessarily need to be physically separated, and may be physically integrated as a measuring device.

[0026] Here, the measurement data managing device 100 may be realized by a computing device such as a personal computer (PC), a smartphone, or a tablet, and may have a hardware configuration such as that shown in Fig. 4. That is, the measurement data managing device 100 has a drive device 101, a storage device 102, a memory device 103, a processor 104, a user interface (UI) device 105, and a communication device 106, which are interconnected via a bus B.

[0027] The programs or instructions that realize the various functions and processes described below in the measurement data managing device 100 may be stored in a removable storage medium such as a CD-ROM (Compact Disk-Read Only Memory) or flash memory. When the storage medium is set in the drive device 101, the programs or instructions are installed from the storage medium to the storage device 102 or memory device 103 via the drive device 101. However, the programs or instructions do not necessarily have to be installed from the storage medium, and may be downloaded from any external device via a network or the like.

[0028] The storage device 102 is realized by a hard disk drive or the like, and stores installed programs or instructions as well as files, data, etc. used to execute the programs or instructions.

[0029] The memory device 103 is realized by a random access memory, a static memory, or the like, and when a program or an instruction is activated, it reads and stores the program, instruction, data, or the like from the storage device 102. The storage device 102, the memory device 103, and the removable storage medium are all non-transitory storage media. They may be collectively referred to as "medium."

[0030] The processor 104 may be realized by one or more CPUs (Central Processing Units), GPUs (Graphics Processing Units), processing circuitry, etc., which may be composed of one or more processor cores, and performs various functions and processes of the measurement data management device 100 described below in accordance with programs, instructions, data such as parameters required to execute the programs or instructions, etc. stored in the memory device 103.

[0031] The user interface (UI) device 105 may be composed of input devices such as a keyboard, mouse, camera, microphone, etc., output devices such as a display, speaker, headset, printer, etc., and input / output devices such as a touch panel, and realizes an interface between a user and the measurement data managing device 100. For example, a user operates the measurement data managing device 100 by operating a keyboard, mouse, etc. to use a GUI (Graphical User Interface) displayed on a display or touch panel.

[0032] The communication device 106 is realized by various communication circuits that execute wired and / or wireless communication processing with external devices, the Internet, a LAN (Local Area Network), a cellular network, or other communication networks.

[0033] However, the above-described hardware configuration is merely an example, and the measurement data managing device 100 according to the present disclosure may be realized by any other appropriate hardware configuration.

[0034] [Measurement data management device] Next, a measurement data managing device 100 according to an embodiment of the present disclosure will be described with reference to Figures 5 to 12. Figure 5 is a block diagram showing the functional configuration of the measurement data managing device 100 according to an embodiment of the present disclosure.

[0035] 5, the measurement data management device 100 includes a load data acquisition unit 110, a center of gravity data generation unit 120, and a measurement data management unit 130. For example, one or more functional units of the load data acquisition unit 110, the center of gravity data generation unit 120, and the measurement data management unit 130 may be realized by one or more processors 104 executing one or more programs or instructions stored in one or more memory devices 103.

[0036] The load data acquiring unit 110 acquires load data generated by measuring the load on the platform 51 on which the user stands at at least three measurement points. Specifically, the load data acquiring unit 110 acquires time-series data (W1 t1 ,W2 t1 ,W3 t1 ,W4 t1 ),(W1 t2 ,W2 t2 ,W3 t2 ,W4 t2 ),···,(W1 tn ,W2 tn ,W3 tn ,W4 tn ) may be acquired as the load data.

[0037] In a typical usage of the load measuring device 50, when a user steps onto the platform 51, the user first places one foot on the platform 51 and moves the center of gravity forward (tag (0)), as shown in Fig. 6. The weight swing caused by this movement may indicate a sudden increase in the load value, as shown in the figure.

[0038] Thereafter, the user supports his / her body with one foot placed on platform 51, and moves his / her center of gravity forward by placing the other foot on platform 51. The weight swing caused by this movement fluctuates up and down around the actual weight value of the user, as shown in the figure.

[0039] Thereafter, the user stands still with both feet on the platform 51 (Count_1), and the weight fluctuation stabilizes at the user's actual weight value as shown in the figure, and the weight value is determined (Count_2).

[0040] The center-of-gravity data generating unit 120 generates time-series data of the center-of-gravity coordinates of the load during the period when the user is on the platform 51, based on the load data. Specifically, for the platform 51 shown in FIG. t1 ,W2 t1 ,W3 t1 ,W4 t1 ),(W1 t2 ,W2 t2 ,W3 t2 ,W4 t2 ),···,(W1 tn ,W2 tn ,W3 tn ,W4 tn ) is acquired as load data, the center of gravity data generating unit 120 calculates x ti [mm]=A×(W1 ti +W2 ti ) / (W1 ti +W2 ti +W3 ti +W4 ti ) y ti [mm]=B×(W1 ti +W3 ti ) / (W1 ti +W2 ti W3 ti +W4ti ) according to the coordinates of the center of gravity (x ti ,y ti ) may be calculated. z ti [kg weight]=(W1 ti +W2 ti +W3 ti +W4 ti ) according to the weight value z at time ti (1≦i≦n) ti may be calculated.

[0041] The centroid data generating unit 120 calculates the centroid coordinates (x ti ,y ti ) based on the centroid coordinate time series data (x t1 ,y t2 ),(x t2 ,y t2 ),···,(x tn ,y t n ) may be generated. The center-of-gravity data generating unit 120 may also generate weight values ​​z ti Based on the weight time series data z t1 ,z t2 ,···,z tn may be generated.

[0042] In one embodiment, the center of gravity data generating unit 120 may generate a two-dimensional image showing the trajectory of the center of gravity coordinates as the center of gravity coordinate time-series data. Alternatively, the center of gravity data generating unit 120 may generate an image showing the center of gravity time-series data associated with the weight time-series data. For example, the weight time-series data may be normalized to a range of 0 to 1, and the normalized weight value at each time point may be associated with the center of gravity coordinate at that time point. When the trajectory of the center of gravity coordinates is visualized, the image of the trajectory may be colorized or grayscaled depending on the normalized weight value.

[0043] The measurement data management unit 130 stores the barycentric coordinate time series data in association with the user's identification information. Specifically, the measurement data management unit 130 associates the barycentric coordinate time series data of the user to be registered with the identification information of the user, and registers the barycentric coordinate time series data of the user as user measurement data. For example, when starting to use the measurement system 10, the measurement data management unit 130 may initially associate the barycentric coordinate time series data of the user with the user's identification information and register it as user measurement data.

[0044] In this way, when load data is acquired from the load measuring device 50 after the user registration is completed, the measurement data management unit 130 determines whether the center of gravity coordinate time series data generated from the acquired load data matches any of the center of gravity coordinate time series data of the registered user. If the generated center of gravity coordinate time series data matches any of the registered user's center of gravity coordinate time series data, the measurement data management unit 130 stores the weight value calculated from the load data in association with the user's identification information.

[0045] On the other hand, if the generated center of gravity coordinate time series data does not match the center of gravity coordinate time series data of a registered user, the measurement data management unit 130 may determine that a new user is on the platform 51 and inquire whether to register the user as a new user. When instructed to register the user as a new user, the measurement data management unit 130 associates the generated center of gravity coordinate time series data with the user's identification information and registers it in the user measurement data.

[0046] Alternatively, if the generated center-of-gravity coordinate time series data does not match the center-of-gravity coordinate time series data of a registered user, the measurement data management unit 130 may inquire as to whether the installation position of the platform 51 has been changed. For example, as shown in FIG. 7 , if the platform 51 is placed at the entrance to the bathroom, it is assumed that the user will step on the platform 51 and then move straight into the bathroom. On the other hand, if the platform 51 is placed below the sink, it is assumed that the user will step on the platform 51 and then move backward to get off the platform 51. Due to such differences in the movements of ascending and descending the platform 51, the center-of-gravity coordinate position time series data of the same user for each ascending and descending movement may also differ, as shown in FIG. 7 . For this reason, the measurement data management unit 130 may inquire of the user as to whether the discrepancy between the generated center-of-gravity coordinate time series data and the center-of-gravity coordinate time series data of the registered user is due to a change in the installation position of the platform 51. When notified that the installation location of the platform 51 has been changed, the measurement data management unit 130 may notify the user to specify the user of the generated center of gravity coordinate time series data, and register the installation location information indicating the installation location of the platform 51 and the center of gravity coordinate time series data in the user measurement data, associated with the identification information of the specified user.

[0047] In one embodiment, the measurement data management unit 130 may generate center-of-gravity coordinate time series data by rotating the center-of-gravity coordinate time series data by 180 degrees, and store the center-of-gravity coordinate time series data rotated by 180 degrees as user measurement data in association with the user's identification information. For example, as described with reference to FIG. 2, the platform 51 may have an orientation depending on the arrangement of the load sensor 52. In this case, the center-of-gravity coordinate time series data for a case in which the user stands on the platform 51 in the moving direction shown in FIG. 8A and a case in which the user stands on the platform 51 in the moving direction shown in FIG. 8B are considered to have trajectories rotated by 180 degrees for the same user. In order to register the center-of-gravity coordinate time series data rotated by 180 degrees for the same user, the measurement data management unit 130 may store the center-of-gravity coordinate time series data rotated by 180 degrees together with the user's center-of-gravity coordinate time series data in association with the user's identification information.

[0048] In one embodiment, the measurement data management unit 130 may store the center of gravity coordinate time series data in association with the weight time series data. As described above, the center of gravity data generation unit 120 may generate weight time series data indicating fluctuations in the user's weight value while the user is on the platform 51 based on the load data, and generate center of gravity coordinate time series data associated with the generated weight time series data. In this case, the measurement data management unit 130 may store the center of gravity coordinate time series data associated with the weight time series data.

[0049] For example, the association may be performed by superimposing weight time-series data on the center of gravity coordinate time-series data. Specifically, the center of gravity data generation unit 120 may normalize each user's weight time-series data within a range of 0 to 1 and associate the normalized value at each time point with the center of gravity coordinate at that time point. Here, the center of gravity data generation unit 120 may quantize the normalized values ​​of 0 to 1 into one of the quantization levels, assign a color or grayscale to each level, and color the trajectory of the center of gravity coordinate time-series data. In this way, the center of gravity coordinate time-series data associated with the weight time-series data can be represented as a colored or grayscaled image showing the trajectory of the center of gravity coordinate, as shown in FIGS. 9A to 9D.

[0050] For example, the center-of-gravity data generation unit 120 may quantize the weight time-series data into four weight levels (for example, for the weight value w normalized to 0 to 1, 0 < w ≤ 0.25, 0.25 < w ≤ 0.5, 0.5 < w ≤ 0.75, 0.75 < w ≤ 1, etc.), and assign colors #1 to #4 to each quantization level. In this case, for example, for the trajectory of the center-of-gravity coordinate time-series data as shown in FIG. 9A, the center-of-gravity data generation unit 120 may generate a color image in which colors # to #4 are assigned to the trajectory, as shown in FIG. 9B. Similarly, for the trajectory of the center-of-gravity coordinate time-series data as shown in FIG. 9C, the center-of-gravity data generation unit 120 may generate a color image in which colors #1 to #4 are assigned to the trajectory, as shown in FIG. 9D. Note that each quantization level of the weight time-series data does not necessarily have to be depicted by color or grayscale, and any other superimposition method that can identify each quantization level of the normalized weight value may be applied.

[0051] The measurement data management unit 130 may store an image showing the trajectory of the center of gravity thus colored or grayscaled as the center-of-gravity coordinate time-series data associated with the weight time-series data. Further, the measurement data management unit 130 may store an image showing the trajectory obtained by rotating the trajectory of the colored or grayscaled center of gravity of each user by 180 degrees, in association with the identification information of the user.

[0052] In one embodiment, the measurement data managing device 100 may use a machine learning model trained to convert barycentric coordinate time-series data into feature data. Specifically, the machine learning model may be any known machine learning model trained to receive as input an image showing the trajectory of barycentric coordinate time-series data and output feature values ​​of the image. The machine learning model may be any type of machine learning model that outputs similar feature values ​​for similar trajectories. For example, the machine learning model may be configured with an architecture such as that shown in FIG. 10 (e.g., EfficientNet B0). Alternatively, the machine learning model may be a machine learning model that receives as input an image showing the colored or grayscale trajectory of barycentric coordinate time-series data associated with weight time-series data and outputs feature values ​​of the image. The machine learning model may be provided in the measurement data managing device 100 or on a cloud.

[0053] As shown in Figure 11, when registering users #1, #2, and #3 in user measurement data, the measurement data management unit 130 inputs an image showing the trajectory of the center of gravity coordinate time series data of users #1, #2, and #3, or an image showing the colored or grayscale trajectory of the center of gravity coordinate time series data associated with weight time series data, into a machine learning model to obtain each user's feature data #1, #2, and #3, associates the obtained feature data #1, #2, and #3 with the identification information of users #1, #2, and #3, and stores the obtained feature data #1, #2, and #3 as user measurement data.

[0054] 12, the measurement data management unit 130 may acquire an image showing a trajectory obtained by rotating the trajectory of the center of gravity coordinate time series data of users #1, #2, and #3 by 180 degrees, or an image showing a trajectory obtained by rotating the colored or grayscaled trajectory of the center of gravity coordinate time series data associated with weight time series data by 180 degrees, and input these images into the machine learning model.The measurement data management unit 130 may then acquire each of the feature amount data #1, #2, and #3 from the machine learning model, and store the acquired feature amount data #1, #2, and #3 as user measurement data in association with the identification information of users #1, #2, and #3.

[0055] In this way, when each user's center of gravity coordinate time series data (and / or center of gravity coordinate time series data rotated 180 degrees) and / or center of gravity coordinate time series data associated with weight time series data (and / or center of gravity coordinate time series data associated with weight time series data rotated 180 degrees) are registered as user measurement data, the measurement data management unit 130 can use a machine learning model to determine which of the users registered in the user measurement data the unknown load data to be identified obtained from the load measuring device 50 belongs to.

[0056] Specifically, the measurement data management unit 130 inputs the center of gravity coordinate time series data generated from the load data of the identification target into a machine learning model and acquires feature data from the machine learning model. Then, when feature data whose similarity to the acquired feature data (for example, the cosine distance between two feature vectors) is greater than a predetermined threshold is registered in the user measurement data, the measurement data management unit 130 may determine that the acquired feature data belongs to the user of the registered feature data, and record a weight value calculated from the load data of the identification target in the user measurement data in association with the identification information of the specified user.

[0057] On the other hand, if no feature amount data having a similarity to the acquired feature amount data greater than a predetermined threshold is registered in the user measurement data, i.e., if the similarity between the acquired feature amount data and any registered feature amount data is equal to or less than a predetermined threshold, the measurement data management unit 130 may determine that the acquired feature amount data belongs to an unregistered user and may inquire of the user whether to register the unregistered user as a new user. When instructed to register the user as a new user, the measurement data management unit 130 may register the feature amount data and / or the center of gravity coordinate time series data in the user measurement data in association with the user's identification information.

[0058] Alternatively, if the similarity between the acquired feature amount data and any registered feature amount data is equal to or less than a predetermined threshold, the measurement data management unit 130 may determine that the installation location of the platform 51 has been changed, and inquire whether the installation location of the platform 51 has been changed. If the installation location has been changed and it is notified that the load data belongs to a registered user, the measurement data management unit 130 may update the installation location information with the changed installation location, and may register the feature amount data and / or the center of gravity coordinate time-series data in the user measurement data in association with the user's identification information.

[0059] According to the above-described measurement system 10, it is possible to automatically identify a user who stands on the load measuring device 50, and record load data indicating the weight or load of the user in association with the user's identification information.

[0060] [Measurement data management processing] Next, a measurement data management process according to an embodiment of the present disclosure will be described with reference to Fig. 13. The measurement data management process is executed by the above-described measurement data managing device 100, and more specifically, may be realized by one or more processors 104 of the measurement data managing device 100 executing one or more programs or instructions stored in one or more memory devices 103.

[0061] FIG. 13 is a flowchart illustrating a measurement data management process according to an embodiment of the present disclosure.

[0062] As shown in FIG. 13, in step S101, the measurement data managing device 100 acquires load data of the user. Specifically, the measurement data managing device 100 acquires load data indicating the load of the user standing on the platform 51 from the load measuring device 50. The load data may be, for example, time-series data (W1 t1 ,W2 t1 ,W3 t1 ,W4 t1 ),(W1 t2 ,W2 t2 ,W3 t2 ,W4 t2 ),···,(W1 tn ,W2 tn ,W3 tn ,W4 tn ) may also be used.

[0063] In step S102, the measurement data managing device 100 generates time-series data of center of gravity coordinates based on the load data. Specifically, the measurement data managing device 100 calculates, for each time point ti, x ti [mm]=A×(W1 ti +W2 ti ) / (W1 ti +W2 ti +W3 ti +W4 ti ) y ti [mm]=B×(W1 ti +W3 ti ) / (W1 ti +W2 ti W3 ti +W4 ti ) according to the coordinates of the center of gravity (x ti ,y ti ) is calculated, z ti [kg weight]=(W1 ti +W2 ti +W3 ti +W4ti ) according to the weight value z at time ti (1≦i≦n) ti In this way, the measurement data managing device 100 can calculate the time series data of the center of gravity coordinates (x t1 ,y t1 ),(x t2 ,y t2 ),···,(x tn ,y tn ) and weight time series data z t1 ,z t2 ,···,z tn and may be generated.

[0064] In step S103, the measurement data managing device 100 generates feature data from the barycentric coordinate time series data using a machine learning model trained to extract features from the barycentric coordinate time series data. Specifically, the measurement data managing device 100 may use a machine learning model trained to output features from an image showing the trajectory of the barycentric coordinates of the barycentric coordinate time series data to input an image showing the trajectory of the barycentric coordinates derived from the barycentric coordinate time series data acquired in step S102 into the machine learning model, and acquire feature data of the image from the machine learning model.

[0065] Furthermore, the measurement data management device 100 may generate feature data from the center of gravity coordinate time series data and the weight time series data using a machine learning model trained to extract features from the center of gravity coordinate time series data and the weight time series data. Specifically, the measurement data management device 100 may generate an image by associating the weight time series data with the trajectory of the center of gravity coordinate of the center of gravity coordinate time series data, input the generated image to a machine learning model, and acquire feature data of the image from the machine learning model. The image generated by associating the weight time series data with the trajectory of the center of gravity coordinate of the center of gravity coordinate time series data may be obtained, for example, by synchronizing the center of gravity coordinate time series data and the weight time series image with respect to time. Specifically, the image may be generated by colorizing or grayscaling the trajectory of the center of gravity coordinate according to each level of the weight value with respect to time.

[0066] In step S104, the measurement data managing device 100 determines whether the acquired feature data has been registered in the user measurement data. Specifically, if the similarity (e.g., the cosine distance between two feature vectors) between the feature acquired in step S103 and the feature of any user registered in the user measurement data is greater than a predetermined threshold (step S104: YES), the measurement data managing device 100 determines that the acquired load data belongs to a registered user, and in step S105, stores the weight value derived from the load data in association with the identification information of the user.

[0067] On the other hand, if the similarity between the feature acquired in step S103 and the feature of any user registered in the user measurement data is equal to or less than a predetermined threshold (step S104: NO), the measurement data managing device 100 determines that the acquired load data is that of an unregistered user and inquires of the user to register the user in the user measurement data, or determines that the installation location of the load measuring device 50 may have changed and inquires of the user whether the installation location of the load measuring device 50 has changed. When instructed to register the user in the user measurement data, the measurement data managing device 100 associates the feature data acquired in step S103 with the identification information of the user and registers it in the user measurement data in step S106. Alternatively, when notified that the installation location of the load measuring device 50 has been changed, the measurement data managing device 100 may associate the identification information designated by the user with installation location information indicating the changed installation location, and store the weight value derived from the load data.

[0068] According to the measurement data management process described above, it becomes possible to automatically identify a user standing on the load measuring device 50, and record load data indicating the weight or load of the user in association with the user's identification information.

[0069] In addition, the following supplementary notes are provided in relation to the above description. (Appendix 1) a load data acquisition unit that acquires load data generated by measuring the load on the platform on which the user stands at at least three measurement points; a center-of-gravity data generating unit that generates first time-series data of center-of-gravity coordinates of the load during a period when the user is standing on the platform based on the load data; a measurement data management unit that stores the first time series data in association with the user's identification information; A measurement data management device having the above. (Appendix 2) The measurement data management device described in Appendix 1, wherein the measurement data management unit generates second time series data by rotating the first time series data by 180 degrees, and stores the second time series data in association with the user's identification information. (Appendix 3) The measurement data management device described in Appendix 1 or 2, wherein the center of gravity data generation unit generates time series data of the user's weight during the period based on the load data, and associates the time series data of the weight with the first time series data. (Appendix 4) 4. The measurement data managing device according to claim 1, wherein the measurement data managing unit stores installation location information indicating an installation location of the stand. (Appendix 5) 5. The measurement data management device according to claim 2, wherein the measurement data management unit acquires first feature data indicating the first time series data and second feature data indicating the second time series data using a trained machine learning model, and stores the first feature data and the second feature data in association with identification information of the user. (Appendix 6) The measurement data management device described in Appendix 5, wherein when time series data of the center of gravity coordinates of the user to be identified is acquired, the measurement data management unit acquires feature data of the target to be identified that indicates the acquired time series data, compares the feature data of the target to be identified with the first feature data and the second feature data, and identifies the user to be identified based on the comparison result. (Appendix 7) The measurement data management device described in Appendix 6 or 7, wherein, when the similarity between the feature of the object to be identified and the first feature data and the second feature data is equal to or less than a predetermined threshold, the measurement data management unit inquires whether to register the object user to be identified as a new user. (Appendix 8) The measurement data management device described in Appendix 6 or 7, wherein, when the similarity between the feature of the object to be identified and the first feature data and the second feature data is equal to or less than a predetermined threshold, the measurement data management unit inquires whether the installation position of the stand has been changed. (Appendix 9) The measurement data management device according to any one of appendices 3 to 8, wherein the measurement data management unit acquires third feature data indicating third time series data generated by associating the first time series data with the weight time series data using a trained machine learning model, and stores the third feature data in association with the user's identification information. (Appendix 10) Obtaining load data generated by measuring a load on a platform on which a user stands at at least three measurement points; generating first time-series data of the center of gravity coordinates of the load during a period when the user is standing on the platform based on the load data; storing the first time series data in association with identification information of the user; The computer performs the measurement data management method. (Appendix 11) Obtaining load data generated by measuring a load on a platform on which a user stands at at least three measurement points; generating first time-series data of the center of gravity coordinates of the load during a period when the user is standing on the platform based on the load data; storing the first time series data in association with identification information of the user; A program that causes a computer to execute the following. (Appendix 12) a load measuring device; a measurement data management device; and the load measuring device measures the load on the platform on which the user stands at at least three measurement points, and transmits load data generated based on the measured load to the measurement data management device; The measurement data management device generates first time series data of the center of gravity coordinates of the load during the period when the user is standing on the platform based on the load data, and stores the first time series data in association with the user's identification information.

[0070] Although the examples of the present disclosure have been described in detail above, the present disclosure is not limited to the specific embodiments described above, and various modifications and variations are possible within the scope of the gist of the present disclosure as set forth in the claims.

[0071] The disclosures of the specification, drawings and abstract contained in Japanese Patent Application No. 2022-118071, filed on July 25, 2022, are incorporated herein by reference in their entirety. [Industrial Applicability]

[0072] The measurement system according to the present disclosure can be used in a weight scale that is shared by multiple users. [Explanation of symbols]

[0073] 10 Measurement System 50 Load measuring device 100 Measurement data management device 110 Load data acquisition unit 120 Center of gravity data generation unit 130 Measurement Data Management Unit

Claims

1. a load data acquisition unit that acquires load data generated by measuring the load on the platform on which the user stands at at least three measurement points; a center-of-gravity data generating unit that generates first time-series data of center-of-gravity coordinates of the load during a period when the user is standing on the platform based on the load data; a measurement data management unit that generates second time series data by rotating the first time series data by 180 degrees, associates the second time series data with the user's identification information, and stores the first time series data and the second time series data; A measurement data management device having the above.

2. 2. The measurement data management device according to claim 1, wherein the center of gravity data generation unit generates time series data of the user's weight during the period based on the load data, and associates the time series data of the weight with the first time series data.

3. The measurement data management device according to claim 1 , wherein the measurement data management unit stores installation location information indicating an installation location of the table.

4. 2. The measurement data management device according to claim 1, wherein the measurement data management unit acquires first feature data representing the first time series data and second feature data representing the second time series data using a trained machine learning model, and stores the first feature data and the second feature data in association with identification information of the user.

5. 5. The measurement data management device of claim 4, wherein, when time series data of the center of gravity coordinates of the user to be identified is acquired, the measurement data management unit acquires feature data of the user to be identified that indicates the acquired time series data, compares the feature data of the user to be identified with the first feature data and the second feature data, and identifies the user to be identified based on the comparison result.

6. 6. The measurement data management device of claim 5, wherein when the similarity between the feature of the object to be identified and the first feature data and the second feature data is equal to or less than a predetermined threshold, the measurement data management unit inquires whether to register the object user as a new user.

7. 6. The measurement data management device according to claim 5, wherein, when the similarity between the feature of the object to be identified and the first feature data and the second feature data is equal to or less than a predetermined threshold, the measurement data management unit inquires whether the installation position of the table has been changed.

8. 3. The measurement data management device according to claim 2, wherein the measurement data management unit acquires third feature data indicating third time series data generated by associating the first time series data with the weight time series data using a trained machine learning model, and stores the third feature data in association with the user's identification information.

9. Obtaining load data generated by measuring a load on a platform on which a user stands at at least three measurement points; generating first time-series data of the center of gravity coordinates of the load during a period when the user is standing on the platform based on the load data; generating second time series data by rotating the first time series data by 180 degrees, and storing the first time series data and the second time series data in association with the user's identification information; The computer performs the measurement data management method.

10. Obtaining load data generated by measuring a load on a platform on which a user stands at at least three measurement points; generating first time-series data of the center of gravity coordinates of the load during a period when the user is standing on the platform based on the load data; generating second time series data by rotating the first time series data by 180 degrees, and storing the first time series data and the second time series data in association with the user's identification information; A program that causes a computer to execute the following.

11. a load measuring device; a measurement data management device; and the load measuring device measures the load on the platform on which the user stands at at least three measurement points, and transmits load data generated based on the measured load to the measurement data management device; The measurement data management device generates first time series data of the center of gravity coordinates of the load during the period when the user is standing on the platform based on the load data, generates second time series data by rotating the first time series data by 180 degrees, and stores the first time series data and the second time series data in association with the user's identification information.

12. A load data acquisition unit that acquires load data generated by measuring the load on a platform on which a user stands at at least three measurement points; a center-of-gravity data generating unit that generates first time-series data of center-of-gravity coordinates of the load during a period when the user is standing on the platform based on the load data; a machine learning model that is trained to receive as input an image showing a trajectory of time-series data of the center of gravity coordinates of the load and output a feature quantity of the image; a measurement data management unit that stores the first time-series data in association with the user's identification information; a machine learning model that is trained to receive an input image showing a trajectory of time-series data of the center of gravity coordinate of the load and output a feature amount of the image, The measurement data management unit When time series data of the center of gravity coordinates of the user to be identified is acquired, time series data is generated by rotating the time series data by 180 degrees; Using the trained machine learning model, feature amount data indicating time series data of the center of gravity coordinates of the user to be identified and feature amount data indicating time series data obtained by rotating the time series data of the center of gravity coordinates of the user to be identified by 180 degrees are acquired; A measurement data management device that compares two feature data related to the user to be identified with a first feature indicating the first time series data obtained using the trained machine learning model, and identifies the user to be identified based on the comparison result.

Images