JP2023131905A - Behavior estimation system, behavior estimation method, and program - Google Patents

Abstract

To provide a behavior estimation system a behavior estimation method, and a program which can accurately estimate a behavior of a moving object even when image data of the moving object partially lost is obtained.SOLUTION: The behavior estimation system comprises first acquisition means 41 which acquires an image of a moving object captured by an imaging apparatus, second acquisition means 42 which acquires information about a state of the moving body measured by a measurement instrument, and estimation means 44 which estimates a behavior of the moving object within a prescribed period on the basis of an image of the moving object captured within the prescribed period and information about a state of the moving object within the prescribed period.SELECTED DRAWING: Figure 3

Description

Application for application of Article 30, Paragraph 2 of the Patent Act Presentation of behavior estimation system, behavior estimation method, and program at a meeting Name of meeting: Information Processing Society of Japan 84th National Convention Venue: Ehime University Johoku Campus Hybrid event Date: Reiwa March 3rd to 5th, 4th year

The present invention relates to a behavior estimation system, a behavior estimation method, and a program.

In recent years, for example, in the fields of medicine, nursing care, security, work/behavior management, information provision services, etc., there has been an increase in Research is progressing on technology to estimate the behavior of people.

For example, in the technology described in Patent Document 1, a person area representing a person (moving object) is detected from a series of input images obtained by continuously photographing the person, and the person remains still for a predetermined period of time or more. If the person has fallen, it is assumed that the person has fallen.

Re-table No. 2018-030024

However, with such technology, if there is another object (for example, another moving object or installed object) in front of the moving object to be detected, at least a portion of the moving object may be hidden by the other object. When the moving object is imaged (image data with at least part of the moving object missing), the area of the moving object to be detected cannot be accurately detected, and as a result, the behavior of the moving object cannot be accurately estimated. There was a risk that it would be difficult to do so.

The present invention has been made in view of the above problems, and provides an action estimation system that can accurately estimate the behavior of a moving object even when image data in which at least a portion of the moving object is missing is obtained. The purpose is to provide behavior estimation methods and programs.

In order to solve the above problems, the present invention firstly provides a first acquisition unit that acquires an image of a moving object captured by an imaging device, and a second acquisition unit that acquires information regarding the state of the moving object measured by a measuring device. an estimation means for estimating the behavior of the moving object within the predetermined period based on an acquisition means, an image of the moving object captured within the predetermined period, and information regarding the state of the moving object within the predetermined period; (Invention 1).

According to this invention (invention 1), the behavior of the moving object within the predetermined period is estimated based on images of the moving object captured within the predetermined period and information regarding the state of the moving object within the predetermined period. Therefore, for example, even if at least part of a moving object is missing in a captured image, the behavior of the moving object can be estimated by using information about the state of the moving object in addition to the image of the moving object. become. Thereby, even if image data in which at least a portion of the moving object is missing is obtained, the behavior of the moving object can be accurately estimated.

In the above invention (invention 1), the estimating means includes an image of the moving object captured within the predetermined period, information regarding the state of the moving object within the predetermined period, and an image of the moving object and/or the moving object. The behavior of the moving object within the predetermined period may be estimated based on a learned model based on machine learning using information regarding the state of as learning data (invention 2).

According to this invention (invention 2), the behavior of the moving object can be easily estimated based on the behavior of the moving object estimated by the learned model using the image of the moving object and/or the information regarding the state of the moving object.

In the above invention (invention 2), the estimating means uses the images of the moving object captured within the predetermined period to obtain a first learning model that is a trained model based on machine learning using the images of the moving object as learning data. The estimated probability of the predetermined behavior of the moving object obtained by inputting it into a predetermined model and the information about the state of the moving object within the predetermined period are based on machine learning using the information about the state of the moving object as learning data. By newly calculating the estimated probability of the predetermined action of the moving object using the estimated probability of the predetermined action of the moving object obtained by inputting it into a second trained model that is a trained model, The behavior of the moving object within a period may be estimated (invention 3).

According to this invention (Invention 3), the estimated probability of a predetermined behavior of a moving object obtained by inputting images of a moving object captured within a predetermined period to the first trained model, and the estimated probability of a predetermined behavior of a moving object within a predetermined period are The behavior of the moving body within a predetermined period can be estimated using the estimated probability of the predetermined behavior of the moving body obtained by inputting the information regarding the state into the second learned model.

In the above invention (invention 2), the estimating means includes an image of the moving object captured within the predetermined period and information regarding the state of the moving object within the predetermined period. The behavior of the moving object within the predetermined period may be estimated by inputting information regarding the state to a third learned model that is a learned model based on machine learning using learning data (invention 4).

According to this invention (Invention 4), by inputting an image of a moving object captured within a predetermined period and information regarding the state of the moving object within the predetermined period to the third trained model, behavior can be estimated.

The above inventions (Inventions 1 to 4) include a detection means for detecting at least one part of the moving body based on the acquired image, and the estimating means detects at least one part of the moving body within the predetermined period. The behavior of the moving object within the predetermined period may be estimated based on information regarding the movement mode of the moving object and information regarding the state of the moving object within the predetermined period (invention 5).

According to this invention (Invention 5), the behavior of the moving object within the predetermined period is determined by using information regarding the movement mode of at least one part of the moving object within the predetermined period and information regarding the state of the moving object within the predetermined period. can be estimated more accurately.

In the above invention (invention 5), the detection means may detect at least one joint of the moving object as at least one part of the moving object (invention 6).

According to this invention (invention 6), the behavior of a moving object can be estimated using information regarding the movement mode of at least one joint of the moving object.

In the above inventions (Inventions 1 to 6), the information regarding the state of the moving object may include at least one of the position of the moving object, the acceleration of the moving object, the angular velocity of the moving object, and the temperature of the moving object (Invention 7). ).

According to this invention (invention 7), the behavior of the moving object can be estimated using at least one of the position of the moving object, the acceleration of the moving object, the angular velocity of the moving object, and the temperature of the moving object.

Second, the present invention provides the steps for a computer to acquire an image of a moving object imaged by an imaging device, a step of acquiring information regarding the state of the moving object measured by a measuring device, and estimating the behavior of the moving object within the predetermined period based on an image of the moving object and information regarding the state of the moving object within the predetermined period. (Invention 8).

Thirdly, the present invention provides a computer with a function of acquiring an image of a moving object imaged by an imaging device, a function of acquiring information regarding the state of the moving object measured by a measuring device, and a function of acquiring an image of a moving object imaged within a predetermined period. A program for realizing a function of estimating the behavior of the moving object within the predetermined period based on an image of the moving object and information regarding the state of the moving object within the predetermined period is provided. ).

According to the behavior estimation system, behavior estimation method, and program of the present invention, even if image data in which at least a portion of the moving body is missing is obtained, the behavior of the moving body can be accurately estimated.

1 is a diagram schematically showing the basic configuration of an action estimation system according to an embodiment of the present invention. FIG. 2 is a block diagram showing the configuration of an estimation device. FIG. 2 is a functional block diagram for explaining functions that play a major role in the behavior estimation system. It is a figure showing the example of composition of the 1st acquisition data. It is a figure which shows the example of a structure of 2nd acquired data. FIG. 3 is a diagram showing an example of the relationship between the distance from the measuring device and the received signal strength. It is a figure which shows an example of the detection result of at least one site|part of a moving body. It is a flowchart which shows an example of the process of an estimation means. FIG. 3 is a diagram showing an example of the configuration of first learning data. It is a figure which shows the example of a structure of 2nd learning data. FIG. 3 is a diagram showing an example of the relationship between estimation coefficients and estimation accuracy. It is a figure showing an example of composition of estimation data. It is a flowchart which shows an example of main processing of the action estimation system concerning one embodiment of the present invention. It is a figure showing an example of composition of action data. (a), (b) is a figure which shows the example of division between an estimation device and an estimation server about each function of an action estimation system.

Hereinafter, one embodiment of the present invention will be described in detail with reference to the accompanying drawings. However, this embodiment is an example, and the present invention is not limited thereto.

(1) Basic configuration of behavior estimation system FIG. 1 is a diagram schematically showing the basic configuration of a behavior estimation system according to an embodiment of the present invention. As shown in FIG. 1, in the behavior estimation system according to the present embodiment, when a moving object (in this embodiment, the subject T) exists in a predetermined space SP such as indoors, a predetermined position in the space SP (in the figure In example 1, an image of the subject T captured by the imaging device 10 provided in the upper center of the space SP and information regarding the state of the subject T measured by the measuring device 20 are collected by the estimation device 30. It is now being acquired.

Further, the estimation device 30 estimates the behavior of the target person T within the predetermined period based on images captured within the predetermined period and information regarding the state of the target person T within the predetermined period. It has become. Here, each of the imaging device 10 and the measuring device 20, and the estimation device 30 are connected to a communication network NW (network) such as the Internet or a LAN (Local Area Network).

The imaging device 10 may be, for example, an imaging device (for example, a digital camera, a digital video camera, etc.) that captures moving images and/or still images, and is configured to capture an image of the space SP at a predetermined position within the space SP. It is provided. Further, the imaging device 10 performs imaging processing at a predetermined frame rate (for example, 30 fps (flames per second), etc.) and transmits the captured image to the estimation device 30 via the communication network NW. It is configured. Note that the imaging device 10 may perform imaging processing when receiving a predetermined imaging instruction signal from the estimation device 30.

Here, the imaging device 10 is an imaging device (for example, an omnidirectional camera, etc.) that captures an omnidirectional image (for example, an image with a circumference of 360° (in the example of FIG. 1, a circumference of 360° in the horizontal direction)). Good too. In this case, by expanding the imaging range, it is possible to grasp the behavior of the subject T over a wide range.

Further, the imaging device 10 may be an imaging device (for example, an infrared camera) that takes an infrared image. This makes it possible to detect the target person T in the captured image, for example, even if the target person T is present in an environment with poor visibility (for example, at night, in a dark place, in bad weather, etc.) .

Furthermore, the imaging device 10 may be an imaging device (for example, a stereo camera, etc.) that captures stereo images. Here, the stereo image may be, for example, a set of two images having a predetermined parallax.

Note that in this embodiment, the case where the space SP is imaged using one imaging device 10 is described as an example; The space SP may be imaged using two imaging devices (such as two imaging devices).

The measuring device 20 is a device that measures the condition of the subject T continuously or intermittently (eg, at predetermined intervals (eg, 300 milliseconds, 1 second, etc.)). Here, the information regarding the condition of the subject T may be, for example, information regarding the position of the subject T, a value representing the physical condition of the subject T, or the position of the subject T. The information may be a value obtained by substituting a value representing the person T's position or physical condition into a predetermined calculation formula, or may be information representing the position or degree of the physical condition of the subject T. Further, the information regarding the location of the subject T may include, for example, location information (for example, at least one of latitude, longitude, and altitude) measured using a location measurement technology such as GPS (Global Positioning System). However, it may also include information regarding received signal strength (RSSI) when a signal transmitted by one device is received by another device. Here, the information regarding the received signal strength may be, for example, a value of the received signal strength, a value obtained by substituting the value of the received signal strength into a predetermined calculation formula, It may also be information representing the degree of received signal strength.

The measuring device 20 may be, for example, a device that measures the position of the subject T (such as a GPS sensor), or may be a device that measures the position of the subject T (for example, a ), angular velocity in 3-axis directions (may be angular velocity of a predetermined part of subject T), heart rate (pulse), blood pressure, body temperature, amount of perspiration, number of steps, walking speed, posture , devices that measure exercise intensity (e.g. heart rate ÷ maximum heart rate) or calories burned (e.g. heart rate monitor, blood pressure monitor, thermometer, sweat meter, 3-axis acceleration sensor, 3-axis gyro sensor, motion sensor, etc.) It may be.

In addition, as shown in FIG. 1, the measuring device 20 includes a first measuring device 21 that can be carried by the subject T or can be worn on the body of the subject T, and a plurality of measuring devices ( In the example of FIG. 1, there are three second measuring devices 22, and a plurality of second measuring devices 22 that wirelessly communicate with the first measuring device 21 when the first measuring device 21 exists in the space SP. It may be composed of.

In this case, when the first measuring device 21 exists in the space SP, the first measuring device 21 connects the plurality of second measuring devices using a predetermined wireless communication method (for example, wireless LAN (for example, Wi-Fi (registered trademark))). 22 may be configured to be able to perform wireless communication. In order to perform wireless communication with the plurality of second measuring devices 22, the first measuring device 21 also sends a wireless signal (for example, probe requests, etc.) at predetermined intervals (eg, every few hundred milliseconds, etc.). Furthermore, the first measuring device 21 may be, for example, a device that can be worn by the subject T (for example, a wearable device), or a portable device that the subject T can carry. Furthermore, the first measuring device 21 includes, for example, a mobile terminal, a smartphone, a PDA (Personal Digital Assistant), a personal computer, and a television receiver equipped with a two-way communication function (a so-called multi-function smart television). ) may be a communication device operated by an individual user.

The plurality of second measuring devices 22 can wirelessly communicate with the first measuring device 21 within the space SP using a predetermined wireless communication method (for example, wireless LAN (for example, Wi-Fi (registered trademark))). It may be provided at any possible position. Further, the plurality of second measurement devices 22 may be, for example, devices that relay wireless communication between two or more first measurement devices 21 existing in the space SP, or may be devices that relay wireless communication between the first measurement devices 21 and the space SP. It may be a device that relays wireless communication with other devices (not shown) existing in the SP, or it may be a device that relays wireless communication between the first measuring device 21 and other devices (for example, not shown) connected via the communication network NW. , estimating device 30, etc.). Furthermore, the plurality of second measurement devices 22 may be packet capture devices.

Further, one of the first measuring device 21 and the plurality of second measuring devices 22 is provided with an RSSI circuit that detects the received signal strength (RSSI) when the other one receives the signal transmitted. You can leave it there. Furthermore, this signal may include identification information (for example, MAC address, etc.) of the measuring device (first measuring device 21 or second measuring device 22) that transmitted the signal, and the received signal strength of this signal is detected by the RSSI circuit, the detected received signal strength and the identification information of the device that transmitted this signal are associated with each other, and the measuring device that received this signal (first measuring device 21 Alternatively, it may be stored in a storage device (not shown) provided in the second measuring device 22).

Note that although the case where wireless communication is performed between the first measuring device 21 and the plurality of second measuring devices 22 using Wi-Fi (registered trademark) is described here as an example, the communication method is as follows. It is not limited to this case. For example, a wireless communication method such as Bluetooth (registered trademark), ZigBee (registered trademark), UWB, optical wireless communication (for example, infrared) may be used, or a wired communication method such as USB may be used.

Every time the measuring device 20 (the first measuring device 21 and/or the plurality of second measuring devices 22) detects the condition of the subject T, it sends information regarding the detected condition to the estimation device 30 via the communication network NW. configured to send.

In addition, in this embodiment, although the case where one 1st measuring device 21 is provided in the subject T is demonstrated as an example, the several 1st measuring device 21 may be provided in the subject T. . Furthermore, in the present embodiment, the case where three second measuring devices 22 are provided in the space SP is described as an example, but the number of second measuring devices 22 provided in the space SP is two. It may be less than or equal to four or more.

Furthermore, the imaging device 10 and the measuring device 20 may be configured to directly communicate with the estimation device 30 using a wired or wireless connection method, or with a predetermined relay device (not shown). The estimating device 30 may be configured to communicate with the estimating device 30 via the relay device by transmitting and receiving information using a wired or wireless connection method.

The estimation device 30 communicates with each of the imaging device 10 and the measuring device 20 via the communication network NW, and transmits the image captured by the imaging device 10 and information regarding the state of the subject T measured by the measuring device 20. The information is configured to be acquired over time via the communication network NW. The estimation device 30 is configured by an individual user, such as a mobile terminal, a smartphone, a PDA, a personal computer, a television receiver with a two-way communication function (including a so-called multi-function smart TV), etc. It may also be a terminal device that is operated.

(2) Configuration of estimation device The configuration of the estimation device 30 will be described with reference to FIG. 2. FIG. 2 is a block diagram showing the internal configuration of the estimation device 30. As shown in FIG. 2, the estimation device 30 includes a CPU (Central Processing Unit) 31, a ROM (Read Only Memory) 32, a RAM (Random Access Memory) 33, a storage device 34, a display processing section 35, It includes a display section 36, an input section 37, and a communication interface section 38, and is provided with a bus 30a for transmitting control signals or data signals between each section.

When power is turned on to the estimation device 30, the CPU 31 loads various programs stored in the ROM 32 or the storage device 34 into the RAM 33 and executes them. In the present embodiment, the CPU 31 reads and executes a program stored in the ROM 32 or the storage device 34, thereby controlling the first acquisition means 41, the second acquisition means 42, the detection means 43, and the estimation means 44 (see FIG. ).

The storage device 34 is, for example, a non-volatile storage device such as a flash memory, an SSD (Solid State Drive), a magnetic storage device (for example, an HDD (Hard Disk Drive), a floppy disk (registered trademark), a magnetic tape, etc.), or an optical disk. It may be a volatile storage device such as a RAM, and stores programs executed by the CPU 31 and data referenced by the CPU 31. The storage device 34 also stores first acquired data (shown in FIG. 4), second acquired data (shown in FIG. 5), first learning data (shown in FIG. 8), and second learning data (shown in FIG. 9), which will be described later. ) and estimated data (shown in FIG. 12) are stored.

The display processing section 35 displays the display data given from the CPU 31 on the display section 36. The display unit 36 is, for example, an LCD (Liquid Crystal Display) monitor including thin film transistors arranged pixel by pixel in a matrix, and displays data on the display screen by driving the thin film transistors based on display data. indicate.

When the estimation device 30 is a button input type device, the input unit 37 includes a button group including a plurality of instruction input buttons such as a direction instruction button and a decision button for accepting user operation input, and a button group such as a numeric keypad. It includes a button group including a plurality of instruction input buttons, and includes an interface circuit for recognizing press (operation) input of each button and outputting it to the CPU 31.

When the estimation device 30 is a touch panel input type device, the input unit 37 mainly receives touch panel type input by touching the display screen with a fingertip or a pen. The touch panel input method may be a known method such as a capacitive method.

Further, when the estimation device 30 is a device capable of voice input, the input unit 37 may be configured to include a microphone for voice input, or may be configured to include voice input via an external microphone. An interface circuit for outputting the information to the CPU 31 may be provided. Furthermore, when the estimation device 30 is a device that can input moving images and/or still images, the input section 37 may be configured to include a digital camera or a digital video camera for image input, It may also include an interface circuit for receiving image data captured by an external digital camera or digital video camera and outputting it to the CPU 31.

The communication interface unit 38 includes an interface circuit for communicating with other devices (for example, the imaging device 10 and the measuring device 20, etc.) via the communication network NW.

(3) Overview of each function in the behavior estimation system The functions realized by the behavior estimation system of this embodiment will be described with reference to FIG. 3. FIG. 3 is a functional block diagram for explaining functions that play a major role in the behavior estimation system of this embodiment. In the functional block diagram of FIG. 3, the first acquisition means 41, the second acquisition means 42, and the estimation means 44 correspond to the main components of the behavior estimation system of the present invention. Other means (detection means 43) are not necessarily essential components, but are components for making the present invention more preferable.

The first acquisition means 41 has a function of acquiring an image of the subject T (moving object) captured by the imaging device 10.

The functions of the first acquisition means 41 are realized, for example, as follows. First, for example, when the subject T exists in the space SP, the imaging device 10 performs imaging processing at a predetermined frame rate (for example, 30 fps, etc.), and each time the imaging processing is performed, an image of the captured image is The data is transmitted to the estimation device 30 via the communication network NW. Here, the image data of the image captured by the imaging device 10 is associated with the imaging date and time and identification information of the imaging device 10 (for example, the serial number and MAC (Media Access Control) address of the imaging device 10). It may also be transmitted to the estimation device 30.

On the other hand, each time the CPU 31 of the estimation device 30 receives (acquires) image data transmitted from the imaging device 10 via the communication interface unit 38, the CPU 31 associates the received image data with the imaging date and time of the image. For example, the first acquired data shown in FIG. 4 is stored in the state. The first acquired data is data in which image data of an image is associated with each other for each imaging date and time of the image. In this way, the first acquisition means 41 can acquire the image of the subject T captured by the imaging device 10.

Note that the CPU 31 may acquire a stereo image of the subject T (moving object) captured by the imaging device 10. In this case, a three-dimensional model of the subject T can be generated based on the stereo image of the subject T, so that at least one It becomes possible to determine the position (coordinates) of two parts in three-dimensional space. This makes it possible to capture the position of at least one part of the subject T more accurately compared to the case where a two-dimensional model of the subject T is used. The accuracy of estimating the behavior of person T can be improved.

Further, for example, when two imaging devices spaced apart in the horizontal and/or vertical directions are provided in the space SP, the CPU 31 causes each imaging device to take images at substantially the same timing. (For example, it may be two images with the same imaging date and time, or two images with a time difference within a predetermined range (for example, several milliseconds to tens of milliseconds, etc.) may be obtained as a set of images (that is, stereo images) having a predetermined parallax.

The second acquisition means 42 has a function of acquiring information regarding the state of the subject T (moving object) measured by the measuring device 20.

Here, the information regarding the state of the subject T (moving body) may include at least one of the position of the subject T, the acceleration of the subject T, the angular velocity of the subject T, and the temperature of the subject T. Note that the information regarding the location of the subject T may include location information (for example, at least one of latitude, longitude, and altitude) measured using a location measurement technology such as GPS, It may also include information regarding the received signal strength (RSSI) when the other measuring device receives a signal transmitted by one measuring device among the plurality of second measuring devices 22. Thereby, the behavior of the subject T can be estimated using at least one of the position of the subject T, the acceleration of the subject T, the angular velocity of the subject T, and the temperature of the subject T.

The functions of the second acquisition means 42 are realized, for example, as follows. First, for example, when the subject T exists in the space SP, the first measurement device 21 measures the state of the subject T (for example, the position of the subject T and the three-axis direction or the two-axis direction of the subject T). at least one of the following: the acceleration of Every time the measurement is performed (every second, etc.), information regarding the measured state is transmitted to the estimation device 30 via the communication network NW. Here, the information regarding the condition of the subject T measured by the first measuring device 21 corresponds to the measurement date and time and the identification information of the first measuring device 21 (for example, the serial number and MAC address of the first measuring device 21). It may be transmitted to the estimating device 30 in a state where it is attached.

Here, the information regarding the condition of the subject T includes information regarding the position of the subject T, and the information regarding the position of the subject T is between the first measuring device 21 and the plurality of second measuring devices 22. A case will be described in which the received signal strength (RSSI) of a signal transmitted and received in communication is included. For example, when the subject T is present in the space SP, the first measuring device 21 performs wireless communication with each of the plurality of second measuring devices 22, and communicates with each of the plurality of second measuring devices 22 at a predetermined interval ( For example, each time a wireless signal (such as a beacon signal) transmitted at intervals of several hundred milliseconds is received, the value of the received signal strength (RSSI) of the wireless signal detected (measured) by the RSSI circuit is calculated. , is stored in a storage device (not shown) provided in the first measuring device 21. Here, the value of RSSI corresponds to the measurement date and time (for example, the date and time when the first measurement device 21 received a wireless signal corresponding to the RSSI value from any of the second measurement devices 22) and the value of the RSSI. It may be stored in a state in which it is associated with the identification information (for example, the serial number, MAC address, etc. of the second measuring device 22) of any second measuring device 22 that transmitted the wireless signal. Then, each time a predetermined period (for example, every 3 seconds, etc.) elapses, the first measuring device 21 generates information regarding the state of the subject T measured by the first measuring device 21 within the predetermined period (i.e., the relevant state). The first measurement device 21 may transmit the RSSI values of all the wireless signals received from the plurality of second measurement devices 22 within a predetermined period to the estimation device 30 via the communication network NW. Here, the information regarding the condition of the subject T measured by the first measuring device 21 within a predetermined period is the identification information of the first measuring device 21 (for example, the serial number, MAC address, etc. of the first measuring device 21). It may be transmitted to the estimation device 30 in a matched state.

Note that here, the first measurement device 21 transmits the RSSI value when receiving the wireless signal transmitted from each of the plurality of second measurement devices 22 to the estimation device 30 as information regarding the state of the subject T. Although the case has been described as an example, each of the plurality of second measuring devices 22 estimates the RSSI value when receiving the wireless signal transmitted from the first measuring device 21 as information regarding the state of the subject T. It may also be transmitted to the device 30.

On the other hand, each time the CPU 31 of the estimation device 30 receives (acquires) information regarding the state of the subject T transmitted from the measuring device 20 via the communication interface section 38, the CPU 31 of the estimation device 30 converts the received information regarding the state into the information. For example, the second acquired data shown in FIG. 5 is stored in association with the measurement date and time. The second acquired data is data in which information regarding the condition of the subject T (in the illustrated example, "measurement information") is described in association with each measurement date and time. In this way, the second acquisition means 42 acquires the state of the subject T measured by the measuring device 20 (here, the position of the subject T, the acceleration of the subject T in the three-axis direction or the two-axis direction, At least one of the angular velocity of the subject T in three or two axial directions and the temperature of the subject T can be acquired.

Note that, based on the function of the second acquisition means 42, the CPU 31 of the estimation device 30 calculates the position of the subject T, the average and/or variance of the acceleration of the subject T in three-axis directions or two-axis directions, and the subject person's position. At least one of the average and/or variance of the angular velocity in three or two axis directions of T and the temperature of the subject T may be acquired.

式（１）中、Ｐ_ｒはＲＳＳＩ（ｄＢｍ）を示し、Ｐ_ｔは第２計測装置２２の送信電力（ｄＢｍ）を示し、Ｇ_ｒは第１計測装置２１の受信アンテナの利得（ｄＢｉ）を示し、Ｇ_ｔは第２計測装置２２の送信アンテナの利得（ｄＢｉ）を示し、Ｌは自由空間損失（ｄＢｍ）を示している。このＬは式（２）でもとめられ、式（２）中、ｄは第１計測装置２１と第２計測装置２２との距離（ｍ）を示し、ｆは無線信号の周波数（Ｈｚ）を示し、ｃは光速（＝２．９９７９２４５８×１０^８）（ｍ／ｓ）を示している。式（１）及び式（２）によってもとめられる距離とＲＳＳＩの値との関係は、例えば図６に示す対数関数で表される。図６に示すように、第１計測装置２１と第２計測装置２２との距離が短いほど、ＲＳＳＩ（図の例では、受信信号強度）の値が大きいことがわかる。 Further, in the present embodiment, the received signal strength (RSSI) of the signal transmitted and received in communication between the first measuring device 21 and the plurality of second measuring devices 22 is included in the information regarding the position of the subject T. Using this received signal strength (RSSI), the distance between the first measuring device 21 and each of the plurality of second measuring devices 22, and the position of the first measuring device 21 in the space SP (that is, the position of the subject T) location). Specifically, the distance between the first measuring device 21 and each of the plurality of second measuring devices 22 can be calculated using, for example, the following equations (1) and (2).
P _r =P _t +G _r +G _t -L...(1)

In equation (1), P _r indicates RSSI (dBm), P _t indicates the transmission power (dBm) of the second measuring device 22, and G _r indicates the gain (dBi) of the receiving antenna of the first measuring device 21. , G _t indicates the gain (dBi) of the transmitting antenna of the second measuring device 22, and L indicates the free space loss (dBm). This L is determined by equation (2), where d indicates the distance (m) between the first measuring device 21 and the second measuring device 22, and f indicates the frequency (Hz) of the wireless signal. , c indicates the speed of light (=2.99792458×10 ⁸ ) (m/s). The relationship between the distance determined by equations (1) and (2) and the RSSI value is expressed, for example, by a logarithmic function shown in FIG. 6. As shown in FIG. 6, it can be seen that the shorter the distance between the first measuring device 21 and the second measuring device 22, the larger the value of RSSI (received signal strength in the illustrated example).

Thereby, using the received signal strength (RSSI) of the signal between the first measuring device 21 and each of the plurality of second measuring devices 22, the first measuring device 21 and each of the plurality of second measuring devices 22 can be connected to each other. It becomes possible to find the distance of Furthermore, it is possible to determine the position of the first measuring device 21 in the space SP (that is, the position of the subject T) using the distance between the first measuring device 21 and each of the plurality of second measuring devices 22. Become.

The detection means 43 has a function of detecting at least one part of the subject T (moving object) based on the acquired image.

Further, the detection means 43 may detect at least one joint of the subject T (moving body) as at least one part of the subject T. Thereby, the behavior of the subject T can be estimated using information regarding the movement mode of at least one joint of the subject T.

The function of the detection means 43 is realized, for example, as follows. For example, based on the function of the first acquisition means 41, the CPU 31 of the estimation device 30 stores each of at least one body part of the subject T every time the image data transmitted from the imaging device 10 is stored as the first acquisition data. A node corresponding to the position of may be set in the image, and the coordinates of each node in the image may be calculated. Here, the setting of each node and the calculation of the coordinates may be performed using, for example, feature point (key point) detection technology based on deep learning (eg, OpenPose, etc.). Further, the coordinates of each node may be two-dimensional coordinates, or may be three-dimensional coordinates, for example, when a three-dimensional model of the subject T is generated from image data (stereo image). .

Note that, for example, when the imaging device 10 captures an omnidirectional image, the CPU 31 develops the omnidirectional image captured by the imaging device 10 into a panoramic image, and configures node settings and nodes for the developed panoramic image. You may also calculate the coordinates of .

Further, when a stereo image of the subject T is acquired, the CPU 31 performs pre-processing (for example, noise removal, etc.) on each image of the set of images having a predetermined parallax, and A three-dimensional model of the subject T may be generated by performing the stereo matching process.

For example, as shown in FIG. 7, the CPU 31 selects at least one region of the subject T (in the illustrated example, "neck", "right shoulder", "left shoulder", "right buttock") in the image captured by the imaging device 10. ”, “left buttock”, etc.) (in the example shown, the coordinates in the X direction and the coordinate in the Y direction), and the nodes corresponding to each of the detected parts (14 nodes in the example shown) Set. Here, the coordinates of each node may be stored in the first acquired data in a state where each node is associated with the image in which it is set. In addition, the CPU 31 may, for example, when at least one part of the subject T cannot be detected because the part is imaged so as to be hidden by another object (for example, another moving object, an installed object, etc.). The coordinates of the node corresponding to the part may be stored as NULL data in the first acquired data.

Note that at least one region of the subject T (“neck,” “right shoulder,” “left shoulder,” “right buttock,” “left buttock,” etc.) is detected here using the image captured by the imaging device 10. However, for example, at least one joint of the subject T (for example, "right shoulder joint", "right elbow joint", "right hip joint", "right knee joint," "left shoulder joint," "left elbow joint," "left hip joint," "left knee joint," etc.) may be detected.

The estimation means 44 estimates the target person within the predetermined period based on images of the target person T (moving object) captured within the predetermined period and information regarding the state of the target person T (moving object) within the predetermined period. Equipped with a function to estimate the behavior of T (moving object).

Here, the estimation means 44 includes an image of the subject T (moving body) captured within a predetermined period, information regarding the state of the subject T (moving body) within the predetermined period, and an image of the subject T (moving body). and/or a trained model based on machine learning using information regarding the state of the subject T (moving body) as learning data, and/or estimating the behavior of the subject T (moving body) within the predetermined period. good. Thereby, the behavior of the target person T can be easily estimated based on the behavior of the target person T estimated by the trained model using the image of the target person T and/or information regarding the state of the target person T. .

Further, the estimation means 44 uses the images of the subject T (moving object) captured within a predetermined period to obtain a first learning model that is a trained model based on machine learning using the image of the subject T (moving object) as learning data. The estimated probability of a predetermined behavior of the target person T (moving object) obtained by inputting the information into the predetermined model and the state of the target person T (moving object) within the predetermined period are The estimated probability of the given action of the target person T (moving object) obtained by inputting the information about the given action into the second trained model, which is a trained model based on machine learning using as learning data, The behavior of the subject T (moving object) within the predetermined period may be estimated by newly calculating the estimated probability of the predetermined behavior. Thereby, the estimated probability of the predetermined behavior of the subject T obtained by inputting the image of the subject T taken within the predetermined period into the first trained model and the state of the subject T within the predetermined period are calculated. The behavior of the target person T within a predetermined period can be estimated using the estimated probability of the predetermined behavior of the target person T obtained by inputting information about the second learned model into the second learned model.

Furthermore, the estimation means 44 calculates the images of the subject T (moving body) captured within a predetermined period and the information regarding the state of the subject T (moving body) within the predetermined period. Estimating the behavior of the target person T (moving object) within a predetermined period by inputting information regarding the state of the target person T (moving object) to the third trained model, which is a trained model based on machine learning using learning data. You may. As a result, by inputting images of the subject T captured within a predetermined period and information regarding the state of the subject T within the predetermined period to the third trained model, the image of the subject T within the predetermined period is Behavior can be estimated.

Furthermore, the estimating means 44 calculates, based on information regarding the movement mode of at least one part of the subject T (moving body) within a predetermined period and information regarding the state of the subject T (moving body) within the predetermined period, The behavior of the subject T (moving object) within the predetermined period may be estimated.

The function of the estimating means 44 is realized, for example, as follows. Note that here, the estimation means 44 uses an image of the subject T taken within a predetermined period, information regarding the state of the subject T within the predetermined period, and a machine that uses the image of the subject T as learning data. Estimating the behavior of the target person T within the predetermined period based on a first learned model based on learning and a second trained model based on machine learning using information regarding the state of the target person T as learning data. Let's explain the case.

An example of the processing of the estimation means 44 in this embodiment will be described with reference to the flowchart in FIG. 8. First, the CPU 31 of the estimation device 30 estimates the behavior of the subject T within a predetermined period based on images of the subject T (moving object) captured within the predetermined period (step S100). Specifically, the CPU 31 of the estimation device 30 accesses the first acquired data, and every time a predetermined period (for example, 3 seconds, etc.) elapses starting from a predetermined imaging date and time, the CPU 31 of the estimation device 30 performs imaging within the predetermined period. Extract all image data. Here, all of the extracted image data is associated with the coordinates of each node corresponding to at least one part of the subject T detected based on the function of the detection means 43. Next, the CPU 31 selects at least one region of the subject T (in this embodiment, "neck", "right shoulder", "left shoulder", "right buttock", "left buttock") from all the extracted image data. etc.) further extract the coordinates of each node corresponding to the node. Thereby, information regarding the time course of the position of at least one body part of the subject T within a predetermined period is obtained. Here, the information regarding the time transition of the position of at least one part of the subject T within the predetermined period is an example of "information regarding the movement mode of at least one part of the moving body within the predetermined period" of the present invention.

Next, the CPU 31 calculates the coordinates of each node corresponding to at least one body part of the subject T (information regarding the time course of the position of at least one body part of the subject T within a predetermined period). The behavior of the subject T within the predetermined period is estimated by inputting information regarding the movement mode of the two body parts into a first learned model based on machine learning using as first learning data. Note that, here, the coordinates of each node corresponding to at least one body part of the subject T are input into the first learned model, so that one or more actions (in this embodiment, "remove the exterior") , "Connect the cable," "Put a sticker on the main body," "Read the barcode," "Assemble," etc.).

An example of the first learning data is shown in FIG. The first learning data shown in FIG. 9 shows that the subject T performs a predetermined action (in the example shown in the figure, "remove the exterior", "connect the cable", "apply a sticker to the main body", "read the barcode"), At least one part of the subject T (in the example shown, the "neck", "right shoulder", "left shoulder", "right buttock", "left buttock") while performing any of the "assembly", etc.) ) is described in a state where the time transition of the position is associated with the relevant action (action label). Thereby, as a result of machine learning using the first learning data, the first learning that shows the relationship between the time transition within a predetermined period of the position of at least one body part of the subject T who performs a predetermined action and the said action. The completed model is configured.

Note that the data described in the first learning data may include data when the same subject performs different actions, or may include data when different subjects perform the same action.

In addition, the CPU 31 uses a model (first 1 trained model).

In this case, for example, when a predetermined model learning instruction is input using the input unit 37, the CPU 31 may perform model learning using the first learning data shown in FIG. The CPU 31 may perform learning using, for example, a time-series neural network model. Here, as the time-series compatible neural network, for example, RNN (Recurrent Neural Network), LSTM (Long Short-Term Memory), which is an advanced type of RNN, etc. can be applied. Further, the CPU 31 may be configured to use, for example, a graph neural network (GNN) model, a convolutional neural network (CNN) model, a support vector machine (SVM) model, a fully connected neural network (FNN) model, a gradient boosting (HGB) model, a wave network Learning may be performed using any one of a plurality of models such as the (WN) model. Note that in this embodiment, behavior estimation is performed by using any of the graph convolutional neural network (GCN) model, graph attention network (GAT) model, and graph convolution LSTM (GC-LSTM) model, which are derivatives of GNN. It is possible to improve accuracy.

In this way, the CPU 31 inputs the time course of the position of at least one body part of the subject T within a predetermined period into the first learned model, thereby adjusting the image of the subject T captured within the predetermined period. Based on this, the actions of the target person T during the specified period (here, "remove the exterior," "connect the cable," "apply a sticker to the main unit," "read the barcode," "assemble," etc.) It can be estimated.

Next, the CPU 31 of the estimation device 30 estimates the behavior of the subject T within the predetermined period based on information regarding the state of the subject T (moving object) within the predetermined period (step S102). Specifically, the CPU 31 of the estimating device 30 accesses the second acquired data, and every time a predetermined period (for example, 3 seconds, etc.) elapses starting from a predetermined measurement date and time, the CPU 31 of the estimation device 30 accesses the second acquired data and calculates the second data within the predetermined period. 2. Extract the measurement information (information regarding the condition of the subject T) stored in the acquired data. Note that the measurement date and time that is the starting point when extracting measurement information from the second acquired data may be the same date and time as the imaging date and time that is the starting point when extracting image data from the first acquired data. Then, the CPU 31 converts the extracted measurement information (information regarding the time transition of information regarding the condition of the subject T within a predetermined period) into a second learning method based on machine learning using the information regarding the condition of the subject T as second learning data. By inputting the information to the learned model, the behavior of the target person T within the predetermined period is estimated. Note that here, by inputting information regarding the state of the subject T to the second trained model, one or more actions (in this embodiment, "remove the exterior", "connect the cable", "main body") The estimated probabilities P (sensor) for each of the following tasks (such as ``sticking a sticker to the object'', ``reading a barcode'', ``assembling'', etc.) are calculated.

An example of the second learning data is shown in FIG. The first learning data shown in FIG. 10 shows that the subject T performs a predetermined action (in the example shown in the figure, "remove the exterior", "connect the cable", "apply a sticker to the main body", "read the barcode"), The time course of the state of the subject T (at least one of the subject T's position in the space SP, acceleration, angular velocity, temperature, etc.) while performing the action ( This is data that is written in association with a behavior label). As a result of machine learning using the second learning data, a second learned model is constructed that shows the relationship between the time course of the state of the subject T performing a predetermined action within a predetermined period and the said action. Ru.

Note that the data described in the second learning data may include data when the same subject performs different actions, or may include data when different subjects perform the same action.

In addition, the CPU 31 uses a model (first model) used to estimate the behavior of the subject T based on the state of the subject T by machine learning using information regarding the time transition of the state of the subject T as second learning data. 2 trained model).

In this case, for example, when a predetermined model learning instruction is input using the input unit 37, the CPU 31 may perform model learning using the second learning data shown in FIG. For example, the CPU 31 is configured to use a time-series neural network model, a GNN model, a CNN model, an SVM model, an FNN model, an HGB model, a WN model, and a convolutional LSTM (GC-LSTM) model, as well as the first learned model described above. Learning may be performed using any one of a plurality of models such as .

In this way, the CPU 31 inputs the time transition of the condition of the subject T within the predetermined period into the second trained model, and thereby calculates the period of time based on the information regarding the condition of the subject T within the predetermined period. It is possible to estimate the actions of the target person T (here, ``remove the exterior'', ``connect the cable'', ``apply a sticker to the main body'', ``read the barcode'', ``assemble'', etc.).

In the present embodiment, the CPU 31 learns the models provided in the estimation device 30 (the first learned model and the second learned model), and estimates the behavior of the subject T using the learned models. Although the case is described as an example, the present invention is not limited to this case. For example, the first trained model and/or the second trained model may be provided in a device other than the estimation device 30. In this case, the CPU 31 transmits information regarding the time course of the position of at least one body part of the subject T within the predetermined period and/or information regarding the time course of the state of the subject T within the predetermined period to another device. Information regarding behavior estimated by the trained model may be inputted to a provided trained model and may be received (acquired) from another device.

Next, the CPU 31 of the estimation device 30 performs a predetermined action based on the behavior estimated based on the image of the subject T (moving body) and the behavior estimated based on the information regarding the state of the subject T (moving body). The behavior of the subject T (moving object) within the period is estimated (step S104). Specifically, the CPU 31 of the estimation device 30 calculates the estimated probability P of a predetermined behavior of the subject T obtained by inputting images of the subject T captured within a predetermined period into the first learned model. (skeleton) and the estimated probability P(sensor) of a predetermined behavior of the target person T obtained by inputting information regarding the state of the target person T within the predetermined period into the second trained model. , the behavior of the subject T within the predetermined period is estimated. For example, the CPU 31 uses the following formula for each of multiple actions (here, "remove the exterior,""connect the cable,""apply a sticker to the main body,""read the barcode,""assembly," etc.) Estimated probability P is newly calculated using (3).
P=P(skeleton)×w+P(sensor)×(1-w)…(3)
In equation (3), w (0<w<1) represents an estimation coefficient representing the weight of image data. For example, the CPU 31 calculates the estimated probability P for each of the plurality of actions, and selects the action with the highest estimated probability P among the plurality of actions (for example, "remove the exterior") as the action of the target person T within a predetermined period. It may be estimated.

FIG. 11 shows an example of the relationship between the estimation coefficient w and the standardized estimation accuracy for each missing rate of the part of the subject T in the image data. In the example shown in FIG. 11, the estimation accuracy for each missing rate is highest within the range of 0.55<w<0.65. Further, as the missing rate of the part of the subject T in the image data increases, the value of the estimation coefficient w that provides the highest estimation accuracy increases.

In this way, the CPU 31 determines the behavior of the target person T within a predetermined period based on the behavior estimated based on the image of the target person T and the behavior estimated based on the information regarding the condition of the target person T. Behavior can be estimated.

Note that the CPU 31 stores information regarding the time transition of at least one body part of the subject T within a predetermined period, information regarding the time transition of measurement information of the subject T within the predetermined period, and information regarding the time transition of the measurement information of the subject T within the predetermined period. The behavior of the subject T may be stored in the estimated data shown in FIG. 12, for example. The estimated data shown in FIG. 12 includes a time change of at least one body part of the subject T within a predetermined period, a time change of measurement information of the subject T within the predetermined period, and a time change of the measurement information of the subject T within the predetermined period. This is data that describes actions in a state where they are associated with each other.

Further, the CPU 31 may present information regarding the estimated behavior of the target person T within the predetermined period. For example, when the CPU 31 estimates the behavior of the target person T within a predetermined period, the CPU 31 may display information regarding the estimated behavior of the target person T within the predetermined period on the display unit 36, for example. Here, the information regarding the behavior of the target person T within a predetermined period may be composed of text data, image data, etc. Furthermore, if the information regarding the behavior of the target person T within the predetermined period is configured as audio data, the CPU 31 outputs the information regarding the behavior of the target person T within the predetermined period from an audio output device such as a speaker. It's okay.

Furthermore, the CPU 31 may transmit information regarding the behavior of the target person T within a predetermined period to another computer (for example, a server, etc.) via the communication network NW.

(4) Flow of main processing of the behavior estimation system of this embodiment Next, an example of the flow of main processing performed by the action estimation system of this embodiment will be described with reference to the flowchart of FIG. 13.

First, the CPU 31 of the estimation device 30 acquires an image of the subject T (moving object) imaged by the imaging device 10 based on the function of the first acquisition means 41 (step S200). Next, the CPU 31 of the estimation device 30 determines the state of the subject T (moving body) measured by the measuring device 20 (for example, the position of the subject T and the state of the subject T) based on the function of the second acquisition means 42. Information regarding at least one of the acceleration in the triaxial direction or biaxial direction, the angular velocity in the triaxial direction or biaxial direction of the subject T, and the temperature of the subject T is acquired (step S202). Note that, after step S200, the CPU 31 of the estimation device 30 may detect at least one part of the subject T (moving object) using the acquired image based on the function of the detection means 43.

Next, based on the function of the estimating means 44, the CPU 31 of the estimation device 30 determines the image of the subject T (moving body) captured within a predetermined period and the state of the subject T (moving body) within the predetermined period. The behavior of the subject T (moving object) within the predetermined period is estimated using the information and (step S204). Here, the CPU 31 of the estimation device 30 uses an image of the subject T captured within a predetermined period, information regarding the state of the subject T within the predetermined period, and an image of the subject T as first learning data. The target person T within the predetermined period is based on the first trained model based on machine learning based on the condition of the target person T and the second trained model based on machine learning using information regarding the state of the target person T as second learning data. The behavior of the person may be estimated. Further, the CPU 31 of the estimation device 30 may learn a model (first learned model) by machine learning using images of the subject T as first learning data, or may learn a model (first learned model) using information regarding the state of the subject T as the first learning data. The model (second learned model) may be learned by machine learning using second learning data.

Note that, after step S204, the CPU 31 of the estimation device 30 may display information regarding the estimated behavior of the target person T within a predetermined period, for example, on the display unit 36. Further, the CPU 31 of the estimation device 30 may transmit information regarding the behavior of the target person T within a predetermined period to another computer (for example, a server, etc.) via the communication network NW.

As described above, according to the behavior estimation system, behavior estimation method, and program of the present embodiment, an image of the subject T captured within a predetermined period, information regarding the state of the subject T within the predetermined period, Based on this, the behavior of the subject T within the predetermined period is estimated, so even if at least a part (at least one part) of the subject T is missing in the captured image By using information regarding the state of the subject T in addition to the image of the subject T, it becomes possible to estimate the behavior of the subject T. Thereby, even if image data in which at least a portion (at least one region) of the subject T is missing is obtained, the behavior of the subject T can be accurately estimated.

Modifications of the above-described embodiment will be described below.
(Modified example)
In the above embodiment, the estimation means 44 uses the image of the subject T captured within a predetermined period, the information regarding the state of the subject T within the predetermined period, and the image of the subject T as the first learning data. The target person T within the predetermined period is based on the first trained model based on machine learning based on the condition of the target person T and the second trained model based on machine learning using information regarding the state of the target person T as second learning data. Although the case where the behavior of the person is estimated is described as an example, the present invention is not limited to this case. For example, the estimation means 44 determines that the movement mode of the subject T within a predetermined period in the captured image satisfies a first condition corresponding to a predetermined action, and the information regarding the state of the subject T within the predetermined period is , when the second condition corresponding to the predetermined action is satisfied, it may be estimated that the subject T is performing the predetermined action. Here, the first condition is, for example, that the position of at least one body part of subject T within a predetermined period is included within the range of the position of at least one body part of subject T that corresponds to a predetermined action. It can also be about being present. Further, the second condition may be, for example, that information regarding the state of the subject T within a predetermined period is included within the range of information regarding the state of the subject T corresponding to a predetermined action. .

In this case, the CPU 31 of the estimation device 30 refers to the behavior data shown in FIG. 14, and based on the captured images captured within the predetermined period and the information regarding the state of the subject T within the predetermined period, The behavior of the subject T within a predetermined period may be estimated. Behavioral data corresponds to multiple actions (in the example shown in the figure, ``remove the exterior'', ``connect the cable'', ``apply a sticker to the main unit'', ``read the barcode'', ``assemble'', etc.) At least one part of the subject T where the action is estimated to be taking place (in the above embodiments, "neck", "right shoulder", "left shoulder", "right buttock", "left buttock", etc.) This is data that describes the temporal transition of the range of the position of and the temporal transition of the range of measurement information of the subject T when it is estimated that the corresponding action is being performed. . The behavior data may be stored in the storage device 34, for example.

For example, the CPU 31 determines that the time transition of each position of at least one body part of the subject T within a predetermined period is included in the range of the position of each body part corresponding to any behavior in the behavior data ( (the first condition is met), and the time transition of information regarding the state of the subject T within the predetermined period is included in the range of measurement information corresponding to any of the actions (the second condition is met) If it is determined that this is the case, it may be estimated that the subject T has performed any of the actions (for example, "remove the exterior") within the predetermined period. Further, the CPU 31 may store information regarding the estimated behavior in the RAM 33 or the storage device 34, for example.

In this way, the behavior estimation system, behavior estimation method, and program according to this modification can exhibit the same effects as the above-described embodiment.

Note that the program of the present invention may be stored in a computer-readable storage medium. The storage medium in which this program is recorded may be the ROM 32, RAM 33, or storage device 34 of the estimation device 30 shown in FIG. Further, the storage medium may be, for example, a CD-ROM or the like that can be read by being inserted into a program reading device such as a CD-ROM drive. Furthermore, the storage medium may be a magnetic tape, a cassette tape, a flexible disk, an MO/MD/DVD, or a semiconductor memory.

The embodiments and modifications described above are described to facilitate understanding of the present invention, and are not described to limit the present invention. Therefore, each element disclosed in the above embodiments and modifications is intended to include all design changes and equivalents that fall within the technical scope of the present invention.

For example, in the embodiment described above, the CPU 31 of the estimation device 30, based on the function of the estimation means 44, collects images of the subject T captured within a predetermined period and information regarding the state of the subject T within the predetermined period. Based on a first trained model based on machine learning using images of the target person T as learning data, and a second trained model based on machine learning using information regarding the state of the target person T as learning data. Although the case where the behavior of the subject T within the predetermined period is estimated is described as an example, the present invention is not limited to this case. For example, based on the function of the estimating means 44, the CPU 31 of the estimation device 30 uses the images of the subject T taken within a predetermined period and the information regarding the state of the subject T within the predetermined period. The behavior of the target person T within a predetermined period may be estimated by inputting the image and information regarding the state of the target person T to a third trained model that is a trained model based on machine learning using as learning data. . In this case, the CPU 31 of the estimation device 30 performs machine learning using the image of the subject T and information regarding the state of the subject T as learning data (for example, learning data that is a combination of the first learning data and the second learning data). , may have a function of learning a model (third trained model) used to estimate the behavior of the subject T based on the captured image of the subject T and information regarding the state. Here, the model used for machine learning may be the same as the first learned model or the second learned model.

Further, in the above-described embodiment, the case where the moving object is a person (target person T) has been described as an example, but the present invention is not limited to this case. The moving object may be any object that can be a target of behavior estimation; for example, it may be an animal other than a person, or it may be an object such as a working machine, a vehicle, or a flying object.

Furthermore, in the embodiment described above, the case where the neck, right shoulder, left shoulder, right buttock, and left buttock of the subject T (moving body) are detected as the parts of the subject T is described as an example, but other parts ( For example, the head, hands, feet, joints, etc.) may be detected.

Further, in the above-described embodiment, the case where the behavior of one moving object (subject T) in the space SP is estimated is explained as an example, but even if the behavior of each of a plurality of moving objects in the space SP is estimated. good.

Furthermore, in the embodiment described above, any of the actions of "removing the exterior", "connecting the cable", "applying a sticker to the main body", "reading the barcode", and "assembling" is performed by the moving object (target person T ) has been described as an example, but the content of the behavior is not limited to this.

Further, in the embodiment described above, the case where one estimating device 30 is provided has been described as an example, but the present invention is not limited to this case. For example, a plurality of estimating devices 30 may be provided, and in this case, the operation details and processing results on any estimating device 30 may be presented in real time on the other estimating devices 30, Processing results etc. in any estimation device 30 may be shared among a plurality of estimation devices 30.

Furthermore, in the embodiment described above, the information regarding the time transition of the position of at least one part of the subject T within the predetermined period is the "information regarding the movement mode of at least one part of the moving object within the predetermined period" of the present invention. Although the corresponding case has been described as an example, it is not limited to this case. "Information regarding the movement mode of at least one part of a moving object within a predetermined period" may be, for example, information regarding the time course of the angle of at least one part of the subject T within a predetermined period, or It may also be information regarding the time course of the distance between multiple parts of the subject T.

Further, in the embodiment described above, the estimation device 30 is configured to implement the functions of the first acquisition means 41, the second acquisition means 42, the detection means 43, and the estimation means 44, but the present invention is not limited to this configuration. For example, a computer or the like (for example, a general-purpose personal computer, a server, etc.) communicatively connected to the estimation device 30 via a communication network such as the Internet or a LAN may perform at least one of the means 41 to 44. It is also possible to have a configuration that realizes the function. For example, each function of the functional block diagram shown in FIG. It may be divided arbitrarily between the server and the server.

Furthermore, the function of at least one of the above-mentioned means 41 to 44 may be realized by the imaging device 10 and/or the measuring device 20.

The behavior estimation system, behavior estimation method, and program of the present invention as described above can accurately estimate the behavior of a moving object even when image data in which at least a portion of the moving object is missing is obtained. For example, there are business management systems that analyze the behavior of moving objects (e.g., workers, working machines, etc.), information provision systems that provide appropriate information according to the behavior of moving objects, and Since it can be suitably used for monitoring systems for people (people, pets, etc.), its industrial applicability is extremely large.

10...Imaging device 20...Measuring device 30...Estimation device 41...First acquisition means 42...Second acquisition means 43...Detection means 44...Estimation means T...Target person

Claims (9)

a first acquisition means for acquiring an image of a moving object captured by the imaging device;
a second acquisition means for acquiring information regarding the state of the moving object measured by a measuring device;
estimating means for estimating the behavior of the moving object within the predetermined period based on an image of the moving object captured within the predetermined period and information regarding the state of the moving object within the predetermined period;
An action estimation system equipped with The estimation means uses, as learning data, an image of the moving object captured within the predetermined period, information regarding the state of the moving object within the predetermined period, and an image of the moving object and/or information regarding the state of the moving object. The behavior estimation system according to claim 1, wherein the behavior of the moving object within the predetermined period is estimated based on a learned model based on machine learning based on the learned model. The estimation means is configured to obtain an image of the moving object captured within the predetermined period by inputting the image of the moving object into a first trained model that is a trained model based on machine learning using the image of the moving object as learning data. A second learned model is a trained model based on machine learning that uses the estimated probability of a predetermined action of the moving object and the state of the moving object within the predetermined period as learning data. The estimated probability of the predetermined action of the moving object obtained by inputting it into the model is used to newly calculate the estimated probability of the predetermined action, thereby estimating the action of the moving object within the predetermined period. The behavior estimation system according to claim 2. The estimation means uses an image of the moving object captured within the predetermined period and information regarding the state of the moving object within the predetermined period, and uses the image of the moving object and the information regarding the state of the moving object as learning data. The behavior estimation system according to claim 2, wherein the behavior of the moving object within the predetermined period is estimated by inputting the behavior to a third learned model that is a learned model based on machine learning. comprising a detection means for detecting at least one part of the moving object based on the acquired image,
The estimating means estimates the behavior of the moving object within the predetermined period based on information regarding the movement mode of at least one part of the moving object within the predetermined period and information regarding the state of the moving object within the predetermined period. The behavior estimation system according to any one of claims 1 to 4, which estimates the behavior estimation system. The behavior estimation system according to claim 5, wherein the detection means detects at least one joint of the moving body as at least one part of the moving body. 7. The behavior estimation system according to claim 1, wherein the information regarding the state of the moving object includes at least one of the position of the moving object, the acceleration of the moving object, the angular velocity of the moving object, and the temperature of the moving object. The computer is
acquiring an image of the moving object captured by the imaging device;
acquiring information regarding the state of the moving object measured by a measuring device;
estimating the behavior of the moving object within the predetermined period based on images of the moving object captured within the predetermined period and information regarding the state of the moving object within the predetermined period;
A behavior estimation method that executes each step. to the computer,
A function to acquire an image of a moving object captured by an imaging device,
a function of acquiring information regarding the state of the moving object measured by a measuring device;
a function of estimating the behavior of the moving object within the predetermined period based on images of the moving object captured within the predetermined period and information regarding the state of the moving object within the predetermined period;
A program to make this happen.

Images