Measurement device, measurement method, and program

Abstract

The present invention comprises: a data acquisition unit 241 that acquires position time series data indicating positions of a contact part 11 of a measurement unit 1 that comes into contact with an object W to be measured, said positions occurring during a measurement period when the contact part 11 is pressed against the object W to be measured, and pressure time series data indicating pressure applied to the contact part 11 during the measurement period; and a state identification unit 242 that analyzes the pressure time series data and the position time series data to identify an internal state parameter of the object W to be measured.

Description

Measurement device, measurement method, and program

[0001] The present invention relates to a measuring device, a measuring method, and a program for measuring the internal state of a measurement object.

[0002] A conventional method for measuring the hardness of an object by measuring the reaction force generated by pressing a contact into the object has been known. In this conventional measurement method, the hardness of the object is evaluated by comparing the measured reaction force per contact area with a reference value (see, for example, Patent Document 1).

[0003] Japanese Patent Application Laid-Open No. 2006-250633

[0004] There are cases where it is necessary to understand the internal state of an object that is not homogeneous. Conventional measurement methods can measure the surface hardness of an object, but when the object is not homogeneous inside, such as food, it is not possible to measure the distribution of the internal state of the object.

[0005] The present invention has been made in consideration of these points, and has as its object to make it possible to identify the distribution of the internal state of a measurement object.

[0006] A first aspect of the present invention provides a measurement device for measuring internal state parameters of a measurement object, the measurement device comprising: a contact portion of a measurement unit that contacts the measurement object; a memory unit that stores reference data indicating the relationship between reference position time series data indicating the position of the contact portion while the contact portion is pressed into a reference object, reference pressure time series data indicating the pressure applied to the contact portion while the contact portion is pressed into the reference object, and the internal state parameters of the reference object; a data acquisition unit that acquires position time series data indicating the position of the contact portion during a measurement period when the contact portion is pressed into the measurement object and pressure time series data indicating the pressure applied to the contact portion during the measurement period; and a state identification unit that identifies the internal state parameters of the measurement object by analyzing the position time series data and the pressure time series data using the reference data.

[0007] The state specifying unit may output, as the internal state parameter, information indicating a distribution of hardness or viscosity inside the measurement object, or a thickness of each of a plurality of layers inside the measurement object.

[0008] The object to be measured may have multiple layers each with different characteristics, the data acquisition unit may acquire the position time series data and the pressure time series data during a measurement period when the contact portion is pressed in a direction that is not parallel to the multiple layers, and the state identification unit may identify an internal state parameter of the object to be measured in association with the position at which the contact portion is pressed into the object to be measured.

[0009] The memory unit may store, as the reference data, a machine learning model that outputs an internal state parameter of the object to be measured when the position time series data and the pressure time series data are input, and the state identification unit may input the position time series data and the pressure time series data to the machine learning model and identify the internal state parameter output from the machine learning model as the internal state parameter of the object to be measured.

[0010] The object to be measured may have multiple layers, each with different properties, and the machine learning model may be created by deep learning using multiple training data indicating the relationship between the pressure time series data and the position time series data for multiple reference objects, each of which corresponds to the multiple layers and is a combination of multiple components with different physical properties.

[0011] The storage unit may store a plurality of reference data corresponding to a plurality of reference objects each having a different internal state parameter, and the state identification unit may identify, among the plurality of reference data, an internal state parameter corresponding to reference data that most closely resembles a relationship between the position time-series data and the pressure time-series data as the internal state parameter of the object to be measured.

[0012] The apparatus may further include a state evaluation unit that identifies information about an internal state of the object to be measured based on the internal state parameter.

[0013] The object to be measured may be food, and the information regarding the internal state may be information regarding one or more of the chewiness, state of decay, degree of ripeness, and freshness of the food.

[0014] The measurement unit may have a plurality of contact portions arranged in a direction perpendicular to a direction in which the contact portions are pressed into the object to be measured, and which simultaneously contact different parts of the object to be measured, the data acquisition unit acquires a plurality of position time series data indicating positions of the plurality of contact portions during a measurement period when the plurality of contact portions are pressed into the object to be measured, and a plurality of pressure time series data indicating pressures applied to each of the plurality of contact portions during the measurement period, and the state identification unit may analyze the plurality of position time series data and the plurality of pressure time series data to identify internal state parameters of the object to be measured.

[0015] The measurement unit is arranged on the surface of a rotating drum and has a plurality of contact portions that sequentially come into contact with different portions of the object to be measured as the rotating drum rotates, and the data acquisition unit acquires a plurality of position time series data indicating positions of the plurality of contact portions during a measurement period when the plurality of contact portions are sequentially pressed into the object to be measured as the rotating drum rotates, and a plurality of pressure time series data indicating pressures applied to each of the plurality of contact portions during the measurement period, and the state identification unit may analyze the plurality of position time series data and the plurality of pressure time series data to identify internal state parameters of the object to be measured.

[0016] A measurement method of a second aspect of the present invention includes the steps of: acquiring, by a computer, position time series data indicating the position of a contact portion of a measurement unit that contacts an object to be measured during a measurement period when the contact portion is pressed into the object to be measured; and pressure time series data indicating the pressure applied to the contact portion during the measurement period; and identifying internal state parameters of the object to be measured by analyzing the pressure time series data and the position time series data using reference data indicating the relationship between reference position time series data indicating the position of the contact portion while the contact portion is pressed into a reference object, reference pressure time series data indicating the pressure applied to the contact portion while the contact portion is pressed into the reference object, and internal state parameters of the reference object.

[0017] A third aspect of the present invention is a program for causing a computer to execute the steps of acquiring position time series data indicating the position of a contact portion of a measurement unit that comes into contact with an object to be measured during a measurement period when the contact portion is pressed into the object to be measured, and pressure time series data indicating the pressure applied to the contact portion during the measurement period; and identifying internal state parameters of the object to be measured by analyzing the pressure time series data and the position time series data using reference data indicating the relationship between reference position time series data indicating the position of the contact portion while the contact portion is pressed into a reference object, reference pressure time series data indicating the pressure applied to the contact portion while the contact portion is pressed into the reference object, and internal state parameters of the reference object.

[0018] A fourth aspect of the present invention is a measurement device that measures internal state parameters of a measurement object, and includes: a contact portion of a measurement unit that contacts the measurement object; a memory unit that stores an arithmetic expression that indicates the relationship between position time series data that indicates the position of the contact portion while the contact portion is pressed into the object, pressure time series data that indicates the pressure applied to the contact portion while the contact portion is pressed into the object, and the internal state parameters of the object; a data acquisition unit that acquires position time series data that indicates the position of the contact portion during a measurement period when the contact portion is pressed into the measurement object and pressure time series data that indicates the pressure applied to the contact portion during the measurement period; and a state identification unit that identifies the internal state parameters of the measurement object by inputting the position time series data and the pressure time series data into the arithmetic expression.

[0019] A measurement method of a fifth aspect of the present invention includes the steps of: acquiring, executed by a computer, position time series data indicating the position of a contact portion of a measurement unit that contacts an object to be measured during a measurement period when the contact portion is pressed into the object to be measured; and pressure time series data indicating the pressure applied to the contact portion during the measurement period; and identifying an internal state parameter of the object to be measured by inputting the pressure time series data and the position time series data into an arithmetic expression indicating the relationship between the position time series data indicating the position of the contact portion while the contact portion is pressed into the object, the pressure time series data indicating the pressure applied to the contact portion while the contact portion is pressed into the object, and an internal state parameter of the object.

[0020] A sixth aspect of the present invention is a program for causing a computer to execute the steps of acquiring position time series data indicating the position of a contact portion of a measurement unit that comes into contact with an object to be measured during a measurement period when the contact portion is pressed into the object to be measured, and pressure time series data indicating the pressure applied to the contact portion during the measurement period; and identifying an internal state parameter of the object to be measured by inputting the pressure time series data and the position time series data into an arithmetic expression that indicates the relationship between the position time series data indicating the position of the contact portion while the contact portion is pressed into the object, the pressure time series data indicating the pressure applied to the contact portion while the contact portion is pressed into the object, and an internal state parameter of the object.

[0021] According to the present invention, it is possible to identify the distribution of the internal state of the measurement object.

[0022] FIG. 1 is a diagram for explaining an overview of the measurement device S. FIG. 2 is a diagram showing a change in pressure when the contact portion 11 moves toward the measurement object W and then moves away from the measurement object W. FIG. 3 is a diagram showing the configuration of the analysis unit 2. FIG. 4 is an example of data showing the relationship between position and pressure. FIG. 5 is a flowchart showing the flow of processing in the analysis unit 2. FIG. 6 is a diagram showing the configuration of a measurement unit 1A according to a first modified example. FIG. 7 is a diagram showing the configuration of a measurement unit 1B according to a second modified example.

[0023] 1 is a diagram illustrating an example of the overview of the measuring device S. The measuring device S is a device for measuring internal state parameters of a measurement object W placed on a floor F. The internal state parameters are data indicating the distribution of physical properties of the material constituting the measurement object W, such as the distribution of hardness or viscosity inside the measurement object W. If the interior of the measurement object W is composed of multiple layers, the internal state parameters are data indicating the thickness of each of these multiple layers.

[0024] The measurement object W may be any object, but an object whose interior is not homogeneous and whose hardness or viscosity varies depending on the position is suitable as the measurement object W. The measurement object W may be, for example, food products such as bread, fruits, and vegetables (thick fried tofu, tofu, fried tofu, etc.), clothing, food products contained in packaging, or industrial products.

[0025] The measurement device S includes a measurement unit 1 and an analysis unit 2. The measurement unit 1 moves based on control data transmitted from the analysis unit 2 and measures the pressure received by the measurement unit 1 while pushing the measurement object W, thereby generating position time series data indicating the position of the measurement unit 1 and pressure time series data indicating the pressure received by the measurement unit 1. In the position time series data, time is associated with the position of the measurement unit 1. In the pressure time series data, time is associated with the pressure received by the measurement unit 1. The measurement unit 1 transmits the generated position time series data and pressure time series data indicating the pressure to the analysis unit 2.

[0026] The analysis unit 2 has, for example, a processor, and identifies internal state parameters of the measurement object W by analyzing the position time series data and pressure time series data received from the measurement unit 1. The analysis unit 2 outputs information indicating the distribution of hardness or viscosity inside the measurement object W as the identified internal state parameters. The analysis unit 2 may identify the layer structure inside the measurement object W based on the distribution of hardness or viscosity inside the measurement object W, and output information indicating the positions of the identified multiple layers and the thickness of each of the multiple layers.

[0027] The measurement unit 1 has a contact section 11, an imaging section 12, and a control section 13. The contact section 11 is a section that comes into contact with the measurement target W, and has one or more contact bodies 111. The contact body 111 may be in any form as long as it can detect the pressure it is receiving, but the contact body 111 in this embodiment is an object that deforms when it receives pressure.

[0028] The photographing unit 12 generates photographed image data by photographing the contact object 111 and inputs the generated photographed image data to the control unit 13. The control unit 13 has, for example, a processor, and identifies the pressure being applied to the contact object 111 based on the shape or size of the image of the contact object 111 included in the photographed image data. The control unit 13 identifies the pressure based on the shape or size of the image of the contact object 111 by, for example, referring to data stored in a memory in which the shape or size of the contact object 111 is associated with the pressure. The control unit 13 generates pressure time series data in which the time output from the timing unit is associated with the pressure. The control unit 13 transmits the pressure time series data to the analysis unit 2.

[0029] 1 , when the contact unit 11 has a plurality of contact bodies 111, the control unit 13 identifies the pressure applied to each of the plurality of contact bodies 111. In this case, the control unit 13 may transmit pressure time-series data associated with each of the plurality of contact bodies 111 to the analysis unit 2.

[0030] The control unit 13 further determines the position of the measurement unit 1. For example, the measurement unit 1 is connected to a fixed member (e.g., a linear actuator) in a movable state, and the control unit 13 determines the position of the measurement unit 1 in the movement direction based on data output from a linear encoder of the fixed member. The movement direction of the measurement unit 1 is, for example, a direction toward the center position of the measurement target W (e.g., the up and down direction in FIG. 1 ), but may be another direction. The control unit 13 generates position time series data in which the time output from the timing unit is associated with the position. The control unit 13 transmits the position time series data to the analysis unit 2. If the contact unit 11 has multiple contact bodies 111, the control unit 13 may transmit position time series data associated with each of the multiple contact bodies 111 to the analysis unit 2.

[0031] 1(a) shows a state in which the contact portion 11 is not in contact with the measurement object W, and FIG. 1(b) shows a state in which the measurement unit 1 has moved from the state in FIG. 1(a) toward the measurement object W and is pressing the measurement object W. The control unit 13 transmits, for example, pressure time series data and position time series data generated after the contact portion 11 comes into contact with the measurement object W to the analysis unit 2.

[0032] In the example shown in Fig. 1, the measurement object W includes a packaging material W1 and an object W2 contained in the packaging material W1. A gas (e.g., air) is sealed between the packaging material W1 and the object W2. When the measurement unit 1 presses such a measurement object W, the pressure received by the contact portion 11 differs between a state in which the contact portion 11 is not in contact with the object W2 via the packaging material W1 and a state in which the contact portion 11 is in contact with the object W2 via the packaging material W1. Furthermore, if the interior of the object W2 is not homogeneous, the pressure received by the contact portion 11 also differs depending on the amount by which the contact portion 11 presses the object W2.

[0033] 2 is a diagram showing changes in pressure when the contact part 11 moves toward the measurement object W and then moves away from the measurement object W. The solid line indicates the pressure received by the contact part 11, and the dashed dotted line indicates the position in the movement direction of the measurement unit 1. Fig. 2(a) shows a case where the packaging material W1 is sealed, and Fig. 2(b) shows a case where there is a hole in the packaging material W1 and air escapes when the packaging material W1 is pressed.

[0034] 2 shows the pressure that the contact part 11 receives from the time when the contact part 11 is not in contact with the measurement object W as shown in FIG. 1(a), so the pressure is zero from the time when the measurement unit 1 starts moving until time T1. As shown in FIG. 2(a), after the contact part 11 comes into contact with the packaging material W1 at time T1, the pressure starts to rise, and the pace of the pressure rise becomes faster from time T2 when the contact part 11 comes into contact with the contained item W2 via the packaging material W1.

[0035] In the example shown in Fig. 2(b), the contact portion 11 presses the packaging material W1, causing the air to escape, resulting in a lower pressure than in the example shown in Fig. 2(a). After time T3, when the air has escaped and the contact portion 11 is no longer in contact with the contents W2 via the packaging material W1, the pressure received by the contact portion 11 becomes lower than that in the example shown in Fig. 2(a).

[0036] In this way, the manner in which the pressure changes over time differs depending on the internal state of the measurement object W. Therefore, the analysis unit 2 can identify the internal state of the measurement object W by analyzing the pressure time-series data.

[0037] 3 is a diagram showing an example of the configuration of the analysis unit 2. The analysis unit 2 has a first communication unit 21, a second communication unit 22, a storage unit 23, and a control unit 24. The control unit 24 has a data acquisition unit 241, a state identification unit 242, and a state evaluation unit 243.

[0038] The first communication unit 21 has a communication interface for transmitting and receiving data to and from the measurement unit 1. The first communication unit 21 may transmit and receive data to and from the measurement unit 1 via wireless communication, or may transmit and receive data to and from the measurement unit 1 via wired communication. The first communication unit 21 transmits control data input from the data acquisition unit 241 to the measurement unit 1. The first communication unit 21 inputs the position time series data and pressure time series data received from the measurement unit 1 to the data acquisition unit 241.

[0039] The second communication unit 22 has a communication interface for transmitting data input from the state identification unit 242 to an external device such as a computer or a printer. The second communication unit 22 transmits, for example, to the external device, the internal state parameters of the measurement object W identified by the state identification unit 242 and information indicating the distribution of physical properties inside the measurement object W identified based on the internal state parameters.

[0040] The storage unit 23 has storage media such as a read-only memory (ROM), a random access memory (RAM), and a solid-state drive (SSD). The storage unit 23 stores programs executed by the control unit 24. The storage unit 23 also stores the position time-series data and pressure time-series data transmitted from the measurement unit 1 via the first communication unit 21.

[0041] Furthermore, the storage unit 23 stores reference data indicating the relationship between reference position time-series data indicating the position of the contact unit 11 while the contact unit 11 is pressed into the reference object, reference pressure time-series data indicating the pressure applied to the contact unit 11 while the contact unit 11 is pressed into the reference object, and internal state parameters of the reference object. That is, the storage unit 23 stores reference data indicating the relationship between the position of the contact unit 11 and the pressure applied to the contact unit 11 while the contact unit 11 is pressed into a reference object whose internal state parameters are known. The position of the contact unit 11 is, for example, the position in the direction in which the contact unit 11 is pressed into the measurement target W. The storage unit 23 stores, for example, multiple pieces of reference data corresponding to multiple reference objects each having different internal state parameters.

[0042] The reference object is, for example, an object made up of a combination of multiple materials with known physical properties, and the reference data is time-series data indicating the relationship between position, pressure, and time measured while the contact portion 11 is pressed into the object. The reference position time-series data is time-series data of the position measured while the contact portion 11 is pressed into the reference object, and the reference pressure time-series data is time-series data of the pressure measured while the contact portion 11 is pressed into the reference object. The reference object is, for example, an object with a multi-layer structure made up by stacking multiple members each having different hardness, viscosity, and thickness. The reference object is an object in which the areas in which each of the multiple members exists are known.

[0043] The reference object does not have to be an actual object, but may be a virtual three-dimensional model. In this case, the reference data may be time-series data showing the relationship between position, pressure, and time, calculated by simulating the movement of the contact portion 11 being pressed into the three-dimensional model of the object.

[0044] The storage unit 23 may store an arithmetic expression that indicates the relationship between the position time-series data, pressure time-series data, and internal state parameters of the object measured while the contact unit 11 is pressed against the object. The object for which the arithmetic expression indicates the relationship is preferably the same type of object as the measurement target. Furthermore, a known method or formula can be used as the arithmetic expression. For example, when calculating hardness as the internal state parameter, the elastic modulus can be calculated from the position time-series data and pressure time-series data, and the elastic modulus can be converted to hardness using a known conversion method.

[0045] The storage unit 23 may store, as reference data, a machine learning model that outputs internal state parameters of the measurement target W when position time series data and pressure time series data are input. That is, the reference data may be a machine learning model created by learning using, as training data, position time series data and pressure time series data indicating the relationship between position, pressure, and time measured while the contact unit 11 is pressed into a number of objects each having different internal state parameters, and the internal state parameters. The machine learning model may be a machine learning model created by learning using, as training data, position time series data and pressure time series data calculated by simulating the movement of the contact unit 11 being pressed into a three-dimensional model, and the internal state parameters.

[0046] If the measurement target W has multiple layers with different properties, the machine learning model may be created by deep learning using multiple training data indicating the relationship between position time-series data and pressure time-series data for multiple reference objects each of which is a combination of multiple members corresponding to the multiple layers and each of which has different physical properties. By changing the combination of members corresponding to the multiple layers, multiple reference objects each with different internal state parameters can be easily created. Using such reference objects enables efficient learning using multiple training data.

[0047] When the contact unit 11 has a plurality of contact bodies 111 as shown in FIG. 1, the storage unit 23 may store a plurality of reference data in association with the plurality of contact bodies 111.

[0048] The control unit 24 includes, for example, a CPU (Central Processing Unit). The control unit 24 executes the programs stored in the storage unit 23 to function as a data acquisition unit 241, a state identification unit 242, and a state evaluation unit 243.

[0049] The data acquiring unit 241 acquires, via the first communication unit 21, position time series data indicating the position of the contact unit 11 during a measurement period when the contact unit 11 of the measurement unit 1 is pressed into the measurement object W, and pressure time series data indicating the pressure applied to the contact unit 11 during the measurement period. The measurement period is, for example, the period from when the contact unit 11 comes into contact with the measurement object W until when it is released. When the contact unit 11 has multiple contact bodies 111 as shown in FIG. 1 , the data acquiring unit 241 acquires multiple position time series data and multiple position time series data associated with each of the multiple contact bodies 111.

[0050] The position time series data and the pressure time series data may include position data and pressure data during a period when the contact portion 11 is not in contact with the measurement object W. The position time series data and the pressure time series data may include, for example, position data and pressure data during a period when the measurement unit 1 moves from the initial position of the measurement unit 1 toward the measurement object W, reaches a predetermined position, and then returns to the initial position.

[0051] When the object to be measured W has multiple layers each having different characteristics, the data acquisition unit 241 acquires position time series data and pressure time series data while the contact portion 11 is pressed in a direction that is not parallel to the multiple layers.

[0052] The state specifying unit 242 specifies the internal state parameters of the measurement object W corresponding to the relationship between the position time-series data and the pressure time-series data by using the reference data stored in the storage unit 23. The state specifying unit 242 specifies the internal state parameters of the measurement object W in association with the position at which the measurement object W is pressed.

[0053] The state identification unit 242 creates data indicating the relationship between the position of the contact unit 11 and the pressure received by the contact unit 11, for example, by associating the position and pressure at the same time in the position time series data and pressure time series data acquired by the data acquisition unit 241, and identifies the internal state parameters of the measurement object W based on the created data. If the contact unit 11 has multiple contact bodies 111, the state identification unit 242 may use reference data corresponding to the multiple contact bodies 111 to identify the internal state parameters of the measurement object W based on the multiple position time series data and the multiple pressure time series data corresponding to the multiple contact bodies 111.

[0054] FIG. 4 is an example of data showing the relationship between position and pressure. FIG. 4 corresponds to the position time series data and pressure time series data shown in FIG. 2. The solid lines correspond to the position time series data and pressure time series data shown in FIG. 2( a), and the dashed lines correspond to the position time series data and pressure time series data shown in FIG. 2( b). As can be seen by comparing the solid and dashed lines, the relationship between position and pressure varies depending on the internal state of the measurement object W. The state identification unit 242 may, for example, analyze a graph of the measured position and pressure to identify the internal state parameters, or may use reference data stored in the memory unit 23 to identify the internal state parameters of the measurement object W corresponding to the relationship between the position time series data and the pressure time series data.

[0055] The state identification unit 242 identifies, among the multiple reference data, the internal state parameter corresponding to the reference data that most closely matches the relationship between the position time-series data and the pressure time-series data as the internal state parameter of the measurement object W. For example, the state identification unit 242 calculates a correlation value between the reference data and data that indicates the relationship between position and pressure based on the measurement results, and determines that the internal state parameter associated with the reference data with the largest correlation value is the internal state parameter of the measurement object W.

[0056] The state identification unit 242 may input the position time series data and the pressure time series data into a machine learning model, and identify the internal state parameters output from the machine learning model as the internal state parameters of the object W to be measured.

[0057] The state specifying unit 242 may output information indicating the distribution of hardness or viscosity inside the measurement object W, or the thickness of each of multiple layers inside the measurement object W, which has been specified as the internal state parameter of the measurement object W. The state specifying unit 242 outputs image data or text data indicating the distribution of hardness levels and viscosity levels inside the measurement object W, for example, by referring to data indicating the relationship between the internal state parameter and the hardness level or viscosity level. The state specifying unit 242 creates data indicating the distribution of hardness levels and viscosity levels inside the measurement object W by specifying to which hardness or viscosity level the internal state parameter of hardness or viscosity corresponding to each of multiple positions on the measurement object W belongs.

[0058] The state specifying unit 242 may output the result of specifying the thickness of each of the plurality of layers, assuming that the position where the change in hardness or viscosity indicated by the internal state parameter is equal to or greater than a threshold value is the boundary position between the plurality of layers. The state specifying unit 242 may output data indicating the material of each of the plurality of layers by referring to data indicating the relationship between the internal state parameter and the type of material. By outputting such data by the state specifying unit 242, the user of the measurement device S can easily grasp the internal structure of the measurement target W.

[0059] The state identification unit 242 may identify the internal state parameters of the object to be measured by inputting the position time series data and pressure time series data acquired by the data acquisition unit 241 into an arithmetic formula stored in the memory unit 23.

[0060] The state evaluation unit 243 identifies information relating to the internal state of the measurement object W based on the internal state parameters. The state evaluation unit 243 outputs the identified information via the second communication unit 22. For example, if the measurement object W is a food product, the state evaluation unit 243 can identify information relating to the internal state of the food product based on the internal state parameters (hardness, viscosity, etc.).

[0061] If the food is fruit, the state evaluation unit 243 identifies, as information regarding the internal state, information indicating, for example, one or more of chewiness, state of decay, ripeness, and freshness using an evaluation index of 1 to 5. If the food is deep-fried tofu, the state evaluation unit 243 may identify, as information regarding the internal state, information indicating, for example, chewiness, state of decay, and the like using an evaluation index of 1 to 5. Furthermore, if the food is deep-fried tofu (deep-fried tofu for inari sushi), the state evaluation unit 243 may identify, as information regarding the internal state, information indicating resistance to tearing using an evaluation index of 1 to 5. Then, when outputting the information regarding the internal state (chewiness, state of decay, ripeness, freshness, and resistance to tearing) as an evaluation index of 1 to 5, the state evaluation unit 243 may use a rule-based algorithm or a trained machine learning model.

[0062] [Processing Flow in Analysis Unit 2] Fig. 5 is a flowchart showing an example of the processing flow in the analysis unit 2. The flowchart shown in Fig. 5 starts from the point when measurement of the internal state parameters of one measurement object W is started.

[0063] The data acquiring unit 241 transmits control data for moving the contact portion 11 to the measurement unit 1, and starts moving the contact portion 11 toward the measurement target W (S11). Thereafter, the data acquiring unit 241 acquires the position time series data and pressure time series data generated by the measurement unit 1 (S12, S13). The data acquiring unit 241 may acquire the position time series data and pressure time series data corresponding to the measurement period after the measurement by the measurement unit 1 is completed, or may acquire the position data and pressure data multiple times at regular time intervals.

[0064] When the data acquisition unit 241 acquires the position time series data and the pressure time series data, the state identification unit 242 analyzes the position time series data and the pressure time series data. The data acquisition unit 241, for example, inputs the position time series data and the pressure time series data into a machine learning model (S14) and identifies internal state parameters of the measurement object W based on the data output from the machine learning model (S15). Next, the state identification unit 242 outputs information indicating the distribution of hardness and viscosity inside the measurement object W or the layer structure inside the measurement object W as the identified internal state parameters (S16).

[0065] 5, following the output of the information on the internal state parameters (S16), the state evaluation unit 243 may output information on the internal state of the measurement object W based on these internal state parameters. As specific examples of information on the internal state of the measurement object W and methods for generating or outputting this information, the above-mentioned methods can be used.

[0066] [First Modification] Figure 6 is a diagram showing the configuration of a measurement unit 1A according to a first modification. The measurement unit 1A differs from the measurement unit 1 shown in Figure 1 in that it has multiple contact portions 11 (contact portion 11A, contact portion 11B, contact portion 11C), but is otherwise the same. The multiple contact portions 11 are arranged, for example, in a direction perpendicular to the pressing direction of the measurement unit 1, and simultaneously contact different portions of the measurement object W. The multiple contact portions 11 may be arranged on a curve that has the same shape as the contour line of the measurement object W.

[0067] Here, when the storage unit 23 is used, the storage unit 23 stores reference data indicating the relationship between the positions of the multiple contact portions 11 while the multiple contact portions 11 are pressed into a reference object whose internal state parameters are known, and the pressures applied to the multiple contact portions 11. The data acquisition unit 241 acquires multiple pieces of position time-series data indicating the positions of the multiple contact portions 11 in the pressing direction of the measurement unit 1 during a measurement period when the multiple contact portions 11 are pressed into the measurement target W, and multiple pieces of pressure time-series data indicating the pressures applied to the multiple contact portions 11 during the measurement period.

[0068] The state identification unit 242 analyzes the plurality of position time series data and the plurality of pressure time series data acquired by the data acquisition unit 241 to identify internal state parameters of the measurement object W. Furthermore, when the storage unit 23 is used, the state identification unit 242 uses reference data stored in the storage unit 23 to identify internal state parameters of the measurement object W corresponding to the relationship between the plurality of position time series data and the plurality of pressure time series data. The state identification unit 242 may identify the internal state parameters using the plurality of data indicating the relationship between position and pressure included in the reference data, or may identify the internal state parameters by inputting the plurality of position time series data and the plurality of pressure time series data into the reference data, which is a machine learning model.

[0069] 6 , when the measurement unit 1 has multiple contact parts 11, other contact parts 11 press on the measurement object W, causing the measurement object W to deform and press on the contact parts 11, applying pressure in directions other than the pressing direction to the contact parts 11. Each of the multiple contact parts 11 has multiple contact bodies 111, and the photographing unit 12 photographs the shapes of the multiple contact bodies 111 provided on the multiple contact parts 11, making it possible to measure multiple pressures applied to the contact parts 11 from multiple directions at multiple positions on the measurement object W.

[0070] In this case, the control unit 13 transmits time series data indicating the relationship between position and pressure corresponding to each of the multiple contact bodies 111 to the analysis unit 2, and the data acquisition unit 241 acquires multiple pieces of position time series data and multiple pieces of pressure time series data associated with the multiple contact bodies 111 of each of the multiple contact units 11. The state identification unit 242 identifies internal state parameters of the measurement object W based on the reference data corresponding to the multiple contact bodies 111 and the multiple pieces of position time series data and multiple pieces of pressure time series data. Since the measurement unit 1 and the analysis unit 2 have such configurations, the state identification unit 242 can identify internal state parameters based on pressure time series data at multiple positions on the measurement object W, and therefore can identify internal state parameters over a wide range of the measurement object W.

[0071] 7 is a diagram showing the configuration of a measurement unit 1B according to a second modification. The measurement unit 1B has a rotating drum 14 and a plurality of contact portions 11 that sequentially come into contact with different portions of the measurement object W as the rotating drum 14 rotates about a rotation axis C. The photographing portion 12 does not rotate as the rotating drum 14 rotates, but photographs the contact bodies 111 of the contact portions 11 that come into contact with the measurement object W. As an example, during the measurement period, the floor F moves as the rotating drum 14 rotates.

[0072] The data acquisition unit 241 acquires a plurality of position time series data indicating the positions of the plurality of contact portions 11 during a measurement period when the plurality of contact portions 11 are sequentially pressed into the measurement object W while the rotating drum 14 rotates, and a plurality of pressure time series data indicating the pressures applied to each of the plurality of contact portions 11 during the measurement period. As in the first modified example, the state identification unit 242 analyzes the plurality of position time series data and the plurality of pressure time series data to identify internal state parameters of the measurement object W. As in the first modified example, the state identification unit 242 can also use reference data to identify internal state parameters of the measurement object W that correspond to the relationship between the plurality of position time series data and the plurality of pressure time series data.

[0073] The data acquiring unit 241 may acquire a plurality of position time series data and a plurality of pressure time series data associated with a plurality of contact bodies 111 of the plurality of contact units 11. In this case, the state identifying unit 242 may identify internal state parameters of the measurement object W based on reference data corresponding to the plurality of contact bodies 111, the plurality of position time series data, and the plurality of pressure time series data.

[0074] With the measurement unit 1 and analysis unit 2 configured in this manner, the state identification unit 242 can identify internal state parameters using position time series data and pressure time series data obtained when each of the multiple contact units 11 comes into contact with a different position on the measurement object W at a different time. Therefore, even if the width of the measurement unit 1 is smaller than that of the analysis unit 2, the internal state parameters can be efficiently identified over a wide range of the measurement object W.

[0075] [Effects of the Measuring Device S] As described above, the measurement unit 1 has a data acquiring unit 241 that acquires position time-series data indicating the position of the contact portion 11 of the measurement unit 1 that contacts the measurement object W during a measurement period when the contact portion 11 is pressed into the measurement object W, and pressure time-series data indicating the pressure applied to the contact portion 11 during the measurement period. The state identifying unit 242 then identifies internal state parameters of the measurement object W that correspond to the relationship between the pressure time-series data and the position time-series data by using the reference data stored in the storage unit 23. This allows the measuring device S to identify with high accuracy the internal state of the measurement object W, such as the distribution of physical properties inside the measurement object W, or the presence or absence of holes or tears in the measurement object W.

[0076] Although the present invention has been described above using the embodiments, the technical scope of the present invention is not limited to the scope described in the above embodiments, and various modifications and changes are possible within the scope of the gist. For example, in the above description, the measurement object W has a packaging material W1 and a contained item W2, but the measurement object W does not have to have the packaging material W1.

[0077] All or part of the device can be configured by functionally or physically distributing or integrating in any unit. New embodiments resulting from any combination of multiple embodiments are also included in the embodiments of the present invention. The effects of the new embodiments resulting from the combination also have the effects of the original embodiments.

[0078] REFERENCE SIGNS LIST 1 measurement unit 2 analysis unit 11 contact unit 12 imaging unit 13 control unit 14 rotating drum 21 first communication unit 22 second communication unit 23 storage unit 24 control unit 111 contact body 241 data acquisition unit 242 state identification unit 243 state evaluation unit

Claims

1. A measuring device for measuring an internal state parameter of a measurement object, comprising: a contact portion of a measurement unit that contacts the measurement object; a memory portion that stores reference data indicating a relationship between reference position time series data indicating the position of the contact portion while the contact portion is pressed into a reference object, reference pressure time series data indicating the pressure applied to the contact portion while the contact portion is pressed into the reference object, and the internal state parameter of the reference object; a data acquisition portion that acquires position time series data indicating the position of the contact portion during a measurement period when the contact portion is pressed into the measurement object, and pressure time series data indicating the pressure applied to the contact portion during the measurement period; and a state identification portion that identifies the internal state parameter of the measurement object by analyzing the position time series data and the pressure time series data using the reference data.

2. The measuring device according to claim 1, wherein the state specifying unit outputs, as the internal state parameter, information indicating the distribution of hardness or viscosity inside the object to be measured, or the thickness of each of a plurality of layers inside the object to be measured.

3. The measuring device of claim 1, wherein the object to be measured has multiple layers, each having different characteristics; the data acquisition unit acquires the position time series data and the pressure time series data during a measurement period when the contact portion is pressed in a direction not parallel to the multiple layers; and the state identification unit identifies internal state parameters of the object to be measured in association with the position at which the contact portion is pressed into the object to be measured.

4. The measuring device described in claim 1, wherein the memory unit stores as the reference data a machine learning model that outputs internal state parameters of the object to be measured when the position time series data and the pressure time series data are input, and the state identification unit inputs the position time series data and the pressure time series data to the machine learning model and identifies the internal state parameters output from the machine learning model as the internal state parameters of the object to be measured.

5. The measuring device of claim 4, wherein the object to be measured has multiple layers each having different characteristics, and the machine learning model is created by deep learning using multiple teacher data indicating the relationship between the pressure time series data and the position time series data for multiple reference objects each of which corresponds to the multiple layers and is a combination of multiple components each having different physical properties.

6. The measuring device according to claim 1, wherein the memory unit stores a plurality of reference data corresponding to a plurality of reference objects each having a different internal state parameter, and the state identification unit identifies, among the plurality of reference data, an internal state parameter corresponding to reference data that is closest to the relationship between the position time series data and the pressure time series data as the internal state parameter of the object to be measured.

7. The measurement device according to claim 1, further comprising a state evaluation unit that identifies information regarding the internal state of the measurement object based on the internal state parameter.

8. The measuring device according to claim 7, wherein the object to be measured is food, and the information relating to the internal condition is information relating to one or more of the chewiness, state of decay, degree of ripeness, and freshness of the food.

9. The measuring device of claim 1, wherein the measuring unit has a plurality of contact portions arranged in a direction perpendicular to the direction in which the contact portions are pressed into the object to be measured, and which simultaneously contact different portions of the object to be measured; the data acquiring unit acquires a plurality of position time series data indicating positions of the plurality of contact portions during a measurement period when the plurality of contact portions are pressed into the object to be measured, and a plurality of pressure time series data indicating pressures applied to each of the plurality of contact portions during the measurement period; and the state identifying unit analyzes the plurality of position time series data and the plurality of pressure time series data to identify internal state parameters of the object to be measured.

10. The measuring device described in claim 1, wherein the measuring unit is arranged on the surface of a rotating drum and has a plurality of contact portions which sequentially come into contact with different portions of the object to be measured as the rotating drum rotates, the data acquisition unit acquires a plurality of position time series data indicating positions of the plurality of contact portions during a measurement period when the plurality of contact portions are sequentially pressed into the object to be measured as the rotating drum rotates, and a plurality of pressure time series data indicating pressures applied to each of the plurality of contact portions during the measurement period, and the state identification unit analyzes the plurality of position time series data and the plurality of pressure time series data to identify internal state parameters of the object to be measured.

11. A measurement method executed by a computer, comprising: a step of acquiring position time series data indicating a position of a contact portion of a measurement unit that contacts an object to be measured during a measurement period when the contact portion is pressed into the object to be measured, and pressure time series data indicating a pressure applied to the contact portion during the measurement period; and a step of identifying an internal state parameter of the object to be measured by analyzing the pressure time series data and the position time series data using reference data indicating a relationship between reference position time series data indicating the position of the contact portion while the contact portion is pressed into a reference object, reference pressure time series data indicating the pressure applied to the contact portion while the contact portion is pressed into the reference object, and an internal state parameter of the reference object.

12. A program for causing a computer to execute the steps of: acquiring position time series data indicating the position of a contact portion of a measurement unit that contacts an object to be measured during a measurement period when the contact portion is pressed into the object to be measured, and pressure time series data indicating the pressure applied to the contact portion during the measurement period; and identifying internal state parameters of the object to be measured by analyzing the pressure time series data and the position time series data using reference data indicating the relationship between reference position time series data indicating the position of the contact portion while the contact portion is pressed into a reference object, reference pressure time series data indicating the pressure applied to the contact portion while the contact portion is pressed into the reference object, and internal state parameters of the reference object.

13. A measuring device for measuring an internal state parameter of an object to be measured, comprising: a contact portion of a measuring unit that comes into contact with the object to be measured; a memory portion that stores an arithmetic expression indicating the relationship between position time series data indicating the position of the contact portion while the contact portion is pressed into the object, pressure time series data indicating the pressure applied to the contact portion while the contact portion is pressed into the object, and the internal state parameter of the object; a data acquiring portion that acquires position time series data indicating the position of the contact portion during a measurement period when the contact portion is pressed into the object to be measured and pressure time series data indicating the pressure applied to the contact portion during the measurement period; and a state identifying portion that identifies the internal state parameter of the object to be measured by inputting the position time series data and the pressure time series data into the arithmetic expression.

14. A measurement method executed by a computer, comprising the steps of: acquiring position time series data indicating the position of a contact portion of a measurement unit that contacts an object to be measured during a measurement period when the contact portion is pressed into the object to be measured, and pressure time series data indicating the pressure applied to the contact portion during the measurement period; and identifying an internal state parameter of the object to be measured by inputting the pressure time series data and the position time series data into an arithmetic expression indicating the relationship between the position time series data indicating the position of the contact portion while the contact portion is pressed into the object, the pressure time series data indicating the pressure applied to the contact portion while the contact portion is pressed into the object, and an internal state parameter of the object.

15. A program for causing a computer to execute the steps of: acquiring position time series data indicating the position of a contact portion of a measurement unit that contacts an object to be measured during a measurement period when the contact portion is pressed into the object to be measured, and pressure time series data indicating the pressure applied to the contact portion during the measurement period; and identifying an internal state parameter of the object to be measured by inputting the pressure time series data and the position time series data into an arithmetic expression indicating the relationship between the position time series data indicating the position of the contact portion while the contact portion is pressed into the object, the pressure time series data indicating the pressure applied to the contact portion while the contact portion is pressed into the object, and an internal state parameter of the object.

Images