Feeling evaluation method and feeling evaluation device

Abstract

To provide a feeling evaluation method to evaluate a feeling or physical properties producing the feeling by adding a predetermined time weighting value according to the acquisition timing of a feature amount, as well as an evaluation device for a feeling or physical properties producing the feeling.SOLUTION: A feeling evaluation method includes: an acquisition step of repeatedly operating a motion body brought into contact with an object to acquire time-series data of dynamic physical amounts generated between the object and the motion body; a feature amount calculation step of calculating a feature amount for each predetermined period from the acquired time-series data of the dynamic physical amounts; and a feeling value identification step of adding, to the feature amount, a predetermined time weighting value according to the time zone corresponding to the feature amount calculated from the acquired time-series data of the dynamic physical amounts and utilizing the feature amount with the time weighting value added to identify a feeling value of the object.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to a tactile evaluation method and a tactile evaluation device. [Background technology]

[0002] There is an evaluation method in which a moving object is moved while being brought into contact with a cosmetic applied to the skin, etc., to detect vibrations generated, and the tactile sensation of the cosmetic when used is evaluated based on the change over time in the frequency spectrum of the detected vibrations (Patent Document 1). In addition, there is a method for evaluating the feel of a specific agent at a plurality of timings, such as immediately after application of the specific agent and after a specific time has elapsed (Patent Document 2). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] International Publication No. 2019 / 039466 [Patent Document 2] JP 2016-204343 A Summary of the Invention [Problem to be solved by the invention]

[0004] Patent Document 1 merely quantifies the change in vibration intensity over time, and is unable to properly evaluate the characteristics of the tactile sensation and its changes. Although Patent Document 2 evaluates the feel of a specified agent at each of a number of different timings, there is still room for research into how to evaluate the feel of a specific agent based on the physical quantities generated at each timing in a way that is closer to the feel that a person actually feels.

[0005] The present invention has been made in consideration of the above-mentioned problems, and relates to a tactile evaluation method and device for specifying the tactile value of an object by adding a predetermined time weighting value to a feature quantity calculated from time series data of mechanical physical quantities generated between an object by moving the object and a moving body, depending on which predetermined period of the acquired time series data of mechanical physical quantities. [Means for solving the problem]

[0006] The present invention relates to a tactile evaluation method including an acquisition step of acquiring time series data of mechanical physical quantities occurring between an object and a moving body by repeatedly moving the moving body in contact with the object, a feature amount calculation step of calculating feature amounts for each predetermined period from the acquired time series data of the mechanical physical quantities, and a tactile value specifying step of adding a predetermined time weighting value to the feature amount depending on which time period in the acquired time series data of the mechanical physical quantities the feature amount corresponds to, and specifying the tactile value of the object using the feature amount to which the time weighting value has been added.

[0007] The present invention also relates to a tactile evaluation device including an acquisition means for acquiring time series data of mechanical physical quantities occurring between an object and the moving body by repeatedly moving a moving body in contact with the object, a feature calculation means for calculating a feature value from the acquired time series data of the mechanical physical quantities at predetermined time intervals, and a tactile value identification means for adding a predetermined time weighting value to the feature value depending on which time period in the acquired time series data of the mechanical physical quantities the feature value corresponds to, and for identifying the tactile value of the object using the feature value to which the time weighting value has been added. Effect of the Invention

[0008] According to the method provided by the present invention, a predetermined time weighting value is added depending on which time period the acquired time series data of mechanical physical quantities corresponds to, and the feature quantity to which the time weighting has been added can be used to identify the tactile value of an object. [Brief description of the drawings]

[0009] [Figure 1] (1) is an image of a finger with a sensor attached touching the surface of the evaluator's skin during washing, and (2) is an image of a finger with a sensor attached touching the surface of the evaluator's skin during rinsing. [Diagram 2] 13A is a diagram showing a time-series waveform signal of Sample 1, and FIG. 13B is a diagram showing a time-series waveform signal of Sample 3. FIG. [Diagram 3] 1 is a conceptual diagram illustrating calculation of a plurality of types of feature amounts from a time-series waveform signal. [Figure 4] 13A is a diagram for explaining the spike signal amount of Sample 1, FIG. 13B is a diagram for explaining the spike signal amount of Sample 2, and FIG. 13C is a diagram for explaining the spike signal amount of Sample 3. FIG. [Diagram 5] (1) is a graph showing the correlation between the average value and variation of Sample 1, (2) is a graph showing the correlation between the average value and variation of Sample 2, (3) is a graph showing the correlation between the average value and variation of Sample 3, and (4) is a graph showing the correlation between the average value and variation of Sample 4. [Figure 6] FIG. 13 is a diagram showing the distance from the origin of each plot. [Figure 7] FIG. 1 is a diagram showing a coordinate system in which the horizontal axis represents x' values ​​and the vertical axis represents y'' values. [Figure 8] (1) is a graph using the coordinate system of FIG. 7 (beginning of rinsing (first period)), and (2) is a graph using the coordinate system of FIG. 7 (first half of rinsing (fourth period)). [Figure 9] (1) is a graph using the coordinate system of FIG. 7 (middle rinsing stage (fifth period)), and (2) is a graph using the coordinate system of FIG. 7 (late rinsing stage (seventh period)). [Figure 10] 8 is a graph (end of rinsing (ninth period)) using the coordinate system of FIG. 7. [Figure 11](1) is a diagram summarizing Figures 8 to 10 of Sample 1, (2) is a diagram summarizing Figures 8 to 10 of Sample 2, (3) is a diagram summarizing Figures 8 to 10 of Sample 3, and (4) is a diagram summarizing Figures 8 to 10 of Sample 4. [Figure 12] (1) is a conceptual diagram of a time variation graph with the x' value as the representative, and (2) is a time variation graph with the x' value as the representative. [Figure 13] (1) is a graph of time variation with the x' value as the representative (time change of each cleaning agent (during cleaning)), and (2) is a graph of time variation with the x' value as the representative (time change of each cleaning agent (during rinsing)). [Figure 14] (1) is a graph explaining a method for determining a tactile value, and (2) is a graph (diagram) explaining a method for determining a tactile value. [Figure 15] (1) is a graph explaining a method for determining a tactile value, and (2) is a graph (diagram) explaining a method for determining a tactile value. [Figure 16] (1) is a graph (diagram) explaining a method for determining a tactile value, and (2) is a diagram explaining the method for determining a tactile value. [Figure 17] 1 is a graph using first tactile values ​​and second tactile values. [Figure 18] FIG. 13 is a diagram for explaining the amount of deviation of the variation from the regression line. [Figure 19] (1) is a graph using a coordinate system with the horizontal axis being the x' value and the vertical axis being the y value (beginning of rinsing (first period)), and (2) is a graph using a coordinate system with the horizontal axis being the x' value and the vertical axis being the y value (first half of rinsing (fourth period)). [Figure 20] (1) is a graph using a coordinate system with the horizontal axis being the x' value and the vertical axis being the y value (middle of rinsing (5th period)), and (2) is a graph using a coordinate system with the horizontal axis being the x' value and the vertical axis being the y value (late rinsing (7th period)). [Figure 21] This is a graph (end of rinsing (9th period)) using a coordinate system with the x' value on the horizontal axis and the y value on the vertical axis. [Figure 22] 1 is a flowchart of a tactile evaluation method. [Figure 23]FIG. 1 is a block diagram of a tactile evaluation device. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0010] Hereinafter, preferred embodiments of the present invention will be described with reference to the drawings. The drawings of the present embodiment are all intended to explain the technical concept, configuration, and operation of the present invention, and are not intended to specifically limit the configuration. In addition, in all drawings, similar components are given similar reference numerals, and duplicated descriptions are omitted as appropriate.

[0011] The evaluation method in this embodiment (hereinafter, sometimes referred to as this method) will be outlined below. The tactile evaluation method of this embodiment includes an acquisition step, a feature amount calculation step, and a tactile value specification step. The "acquisition step" is a step of repeatedly moving a moving object in contact with an object to acquire time-series data of a mechanical physical quantity occurring between the object and the moving object. Examples of "objects" include human skin surfaces, artificial skin surfaces, hair, and scalps, as well as tableware and building materials for housing, and in particular, objects to which a specific agent described below is applied. The "moving object" is another object that is brought into contact with the object, and examples of such objects include the finger 20, the palm, a skin care product such as a massager, or a measuring tool that is brought into contact. The sensor 30 shown in FIG. 1 is attached to the finger 20, the palm, the skin care product, or the measuring tool. The finger 20 that is brought into contact with the object may be any position on the finger 20, but the finger pad is preferred because the tactile sensation is often confirmed with the finger pad. When checking the tactile sensation of an object, not only one finger but also multiple fingers may be used, and the palm may be brought into contact with the skin surface because the tactile sensation is sometimes confirmed by contacting not only the fingers but also the entire palm. The measuring tool is a tool that is brought into contact with the skin surface in the same way as the finger. The structure and composition of the part of the measuring tool that is brought into contact with the skin surface are not limited, but by forming the part with artificial skin that imitates human skin, it becomes possible to measure physical quantities that are close to those when a person touches the surface of an object with a finger or palm. "Repeated motion" refers to a motion of continuously contacting a moving body with an object, or a motion of intermittently contacting a moving body with an object. Specifically, it refers to a motion of repeatedly moving a finger in an outward direction approximately vertically on an object (tapping motion), a motion of repeatedly sliding a finger horizontally on an object, etc. In the case of a sliding motion, any of the following motions may be used: a repetitive motion in one direction, a repetitive motion by reciprocating, a repetitive motion in a circular motion, etc. Furthermore, a repetitive motion refers to a single motion (e.g., sliding once in the horizontal direction) followed by a certain period of time irregularly followed by the single motion, i.e., a case where a single motion is repeated at the same interval, or a case where a single motion is performed multiple times at irregular intervals. "Time series data of mechanical physical quantities occurring between an object and a moving body" refers to physical quantities related to mechanics occurring between the object and the moving body by repeatedly moving the moving body (e.g., a finger or a palm) in contact with the object. Examples include the magnitude of the elastic force or frictional force that the moving body receives from the object (skin), the amount of current or voltage caused by the displacement of the object, the pressure that the moving body receives, the distortion caused by the moving body, or the vibration amount (amplitude), frequency, speed and acceleration of the vibrating moving body. The tactile sensation evaluated in this method is a tactile sensation felt through receptors such as fingers by contacting the surface of an object with a finger or palm, such as sticky, refreshing, moist, and tight. The term "acquire" will be described in more detail below, but for example, it means that the sensor 30 shown in FIG. 1 detects vibrations and deformations caused by continuously contacting a moving body with an object, and outputs the vibrations and deformations as electrical signals to a computing device.

[0012] The "characteristics calculation step" is a step of calculating characteristics for each predetermined period from the acquired time-series data of the mechanical physical quantities. "Calculating the feature amount at every predetermined time period" means calculating the feature amount at every predetermined time period from the start of moving the moving body relative to the object, from the acquired time-series data of the mechanical physical amount. The calculation content will be described later. The "feature amount" is, for example, a mechanical physical quantity, a statistic of a mechanical physical quantity, a spike signal amount that is a value equal to or greater than a predetermined threshold value of a mechanical physical quantity or a statistic of a mechanical physical quantity, and a feature amount calculated from an average value or a variation that is a statistic of a mechanical physical quantity, but is not limited thereto. The mechanical physical quantity includes not only the measured value itself but also a calculated value (e.g., a value obtained by removing noise, a value multiplied by a coefficient, etc.). The statistics of the mechanical physical quantity includes not only a statistic using the measured value of the mechanical physical quantity itself (e.g., an average value, a standard deviation, a variance value, a maximum value, a minimum value, a kurtosis, a skewness, etc.) but also a statistic using a calculated value of the mechanical physical quantity. The statistics also include a value that can be calculated by frequency analysis of the mechanical physical quantity, for example, a power value of the spectrum at each time. The feature amount may be calculated after frequency decomposition or filtering of the waveform of the acquired mechanical physical quantity. The spike signal amount includes not only a value calculated from the measured value of the mechanical physical quantity itself but also a value calculated from a value calculated from the mechanical physical quantity. The spike amount will be described later. The feature amount calculated from the average value or the variation, which is a statistical amount of the mechanical physical amount, is, for example, the distance from the origin calculated from the correlation between the average value and the variation (the distance from the origin will be described later).

[0013] The "tactile value identification process" is a process in which a predetermined time weighting value is added to a feature depending on which time period in the acquired time series data of the mechanical physical quantity the feature corresponds to, and the feature to which the time weighting value has been added is used to identify the tactile value of the object. "Depending on which time period in the time series data of the acquired mechanical physical quantity the feature corresponds to" means depending on which time period in the time series data of the acquired mechanical physical quantity the feature corresponds to, for example, depending on which time period from when the moving body starts to move on the object, such as when the moving body starts to move on the object or a predetermined time has elapsed since the moving body starts to move on the object. The time period may be the same as the predetermined time when the feature is calculated, or may be different. The term "depending on" means that the time weighting value to be added is determined depending on which time period from when the moving body starts to move on the object, as will be described in detail later. The "predetermined time weighting value," which will be described in detail later, refers to when a moving object (e.g., a finger) in contact with an object is repeatedly moved to obtain time-series data of mechanical physical quantities occurring between the object and the moving object to evaluate the tactile sensation, and the degree to which the same physical quantity affects the tactile sensation differs immediately after the start of the movement and at the end of the movement. The predetermined time weighting value indicates this "degree." The predetermined time weighting value may be a value for each hour, or may be a ratio to a weighting value at a reference time (e.g., immediately after the start of the movement). The "addition process", the details of which will be described later, is a process of adding a predetermined time weighting value to the feature amount for each predetermined period described above. The "tactile value of an object" is a value related to the tactile sensation determined using a feature amount calculated every predetermined period of time. Details will be described later, but for example, it is a value obtained by adding a predetermined time weighting value to a mechanical physical quantity. Examples of the "tactile sensation" include "sticky," "refreshing," "moist," "sticky," "moisture," "dry," "firm," "elastic," "soft and hard," "soft and chewy," "plump," "smooth," "slippery," "oily," "rich," "comfortable," "dry," "stiff," "tight," "tight," "squeaky," and "slippery," but are not limited to these. The tactile sensation referred to here is the tactile sensation felt through receptors such as fingers by touching an object with fingers or palms (hereinafter referred to as "fingers, etc."), and may also include thermal sensations such as coldness and warmth felt by touching an object with fingers, etc. Here, "sticky" refers to the feeling of the skin sticking to the finger when the finger touches the object. "Refreshing" refers to the feeling of the skin sticking to the finger without any sticky feeling when the finger touches the object. "Moist" refers to the feeling of the skin being slightly damp and smooth when the finger touches the object. "Sticky" refers to the feeling of the skin being slightly sticky when the finger touches the object. "Moist" refers to the feeling of the skin being moderately moist when the finger touches the object. "Dry" refers to the feeling of the skin being dry when the finger touches the object. "Firmness" refers to the feeling of the skin being stiff and taut when the finger touches the object. "Elasticity" refers to the feeling of the skin bouncing back without any sinking feeling when the finger touches the object. "Hardness" refers to the degree to which the skin is deformed when the finger touches the skin surface. "Sticky" refers to the feeling that your fingers stick to the surface when you touch it. "Soft" refers to the feeling that your fingers stick to the surface a little and then bounce back when you touch it. "Fluffy" refers to the feeling that your fingers are slightly elastic and don't stick to the skin surface when you touch it, giving a refreshing feeling. "Smooth" refers to the feeling that your fingers move without any snags when you touch it to an object, especially when a cream has been applied, and that your fingers move naturally."Slippery" refers to the feeling that the finger slips when it touches an object, especially when a cosmetic agent is applied, due to the adhesiveness of the agent. "Smooth" refers to the feeling that the skin is soft and the finger moves without catching when it touches an object. "Oily" refers to the feeling that the finger sticks to the object when it touches it. "Rich" refers to the feeling that the finger feels the texture of the agent when it touches an object, especially when a cosmetic agent is applied, the texture when it spreads, and the weight when it spreads. "Familiarity" refers to the feeling that the finger does not feel strange when it touches an object, especially when a cosmetic agent is applied, and the feeling that the finger does not feel the texture of the agent on the object. "Dry" refers to the feeling that the finger feels dry when it touches an object. "Rough" refers to the feeling that the finger feels rough when it touches an object. "Tight" refers to the feeling that the finger feels taut when it touches an object. "Kyuto" or "kyu" refers to the feeling of resistance when you touch an object with your finger and start to slide it, or while you are sliding it. "Squeak" refers to the feeling of catching and rubbing. "Slip" refers to the feeling of your finger sliding easily with little catching.

[0014] The feature amount calculated from the time series data of the mechanical physical quantity at each predetermined period may be calculated from all the acquired time series data of the mechanical physical quantity or may be calculated from a part of the acquired time series data of the mechanical physical quantity. The "predetermined period" may be always the same period or may be different depending on the time period when the moving body is brought into contact with the object (immediately after the start of contact, after a predetermined time has passed, etc.). For example, the period immediately after the start of contact may be a short period (fine time interval), and the period may be gradually coarser (coarser time interval). Also, the period may be different for each stage (for example, during washing and during rinsing), which will be described later. A plurality of types of feature amounts may be calculated at each predetermined period from the time series data of the mechanical physical quantity. The predetermined periods for all the plurality of types of feature amounts do not need to be the same, and the evaluation by this method can be performed even if the predetermined periods are different for each of the plurality of types of feature amounts. To further explain the point that a predetermined time weighting value is added depending on which time period in the time series data of the acquired mechanical physical quantity corresponds to the feature amount in the tactile value specifying step described above, the addition process may be performed on the feature amount in all time periods, or may not be performed on the feature amount in the time period immediately after the start of contact, or may be performed only on the feature amount in a specific time period. Also, the "time period" may be a fixed period of time or may be a different period of time.

[0015] <Method of acquiring time series data of mechanical physical quantities> A method for acquiring time series data of mechanical physical quantities occurring between an object and a moving body is shown. In this embodiment, time series data of mechanical physical quantities occurring between an object to which a specific agent is applied and a moving body is acquired, but the acquisition method is similar in the case of an object to which a specific agent is not applied. Here, the term "predetermined agent" refers to a liquid (including mist), paste, solid, or powder, etc., that is applied to an object. Examples of the agent include external skin agents, cosmetics, sheet-type skin care cosmetics, dishwashing detergent, household detergent, body cleansing agent, face wash, hair cleansing agent, and other cleansing agents, and examples thereof include skin care cosmetics such as lotion, milky lotion, cream, beauty essence, massage, pack, lip cream, eye care sheet, mouth sheet, pack mask, sheet-type lotion, sheet-type makeup remover, etc.; makeup cosmetics such as foundation, makeup base, liquid foundation, oily foundation, powder foundation, concealer, control color, eye shadow, blusher, lipstick, lip gloss, lip liner, and body decollete; UV protection cosmetics such as sunscreen milky lotion, sunscreen gel, and sunscreen cream, body cleansing agent; solid soap, hand soap, and body soap, but are not limited thereto.

[0016] 1, in this embodiment, the "object" is the skin of a person's arm (skin surface of the person's arm), and the "moving body" is a finger 20 in contact with the person's arm. The "predetermined agent" to be applied to the object is a cleaning agent 40. As shown in FIG. 1, a sensor 30 is attached to a finger 20. When the finger 20, particularly the finger pad (hereinafter, the finger pad is also referred to as the finger), touches the skin of the arm to which the cleaner 40 has been applied, and the touched finger 20 is moved, vibrations and deformations occur in the finger skin, and the sensor 30 detects the vibrations and deformations and outputs them as electrical signals to the computing device 10. Since the vibrations and deformations occurring in the finger skin differ depending on the tactile sensation, the electrical signals outputted depending on the tactile sensation also differ. The outputted electrical signals correspond to the "mechanical physical quantity" of the present invention, and the electrical signals are acquired over time. The sensor 30 is only required to acquire the mechanical physical quantity generated by the touch with the finger 20, and may be, for example, a sensor (such as an acceleration sensor, angular velocity sensor, vibration sensor, or force sensor) capable of acquiring the acceleration or force generated by removing the finger 20 from the skin. Acquiring a physical quantity over time includes not only a mode in which the sensor 30 continuously acquires analog information (electrical signal in this embodiment) indicating the physical quantity over a predetermined time, but also a mode in which the sensor 30 acquires the physical quantity as digital information multiple times at short intervals such as on the order of milliseconds or sub-milliseconds. When the sensor 30 continuously acquires the physical quantity as analog information, it is preferable that the computing device 10 samples and discretizes the analog information at predetermined short intervals such as on the order of milliseconds or sub-milliseconds.

[0017] 1, in this embodiment, the finger 20 equipped with the sensor 30 is brought into contact with the skin (skin surface) of the arm, and the finger 20 is repeatedly slid in a substantially horizontal direction against the skin. The repeated movement may be in any form, such as repeated movement in one direction, repeated movement by reciprocating, or repeated movement in a circular motion. 1(1), a cleaning agent 40 is applied to the skin of an arm, and the mechanical physical quantity occurring between the skin of the arm and the finger 20 is obtained when the finger 20 is repeatedly moved. That is, in FIG. (1), the mechanical physical quantity between an object (corresponding to the skin of the arm) to which a specific agent (corresponding to the cleaning agent 40) has been applied and a moving body (corresponding to the finger 20) is obtained, and the object is cleaned by moving the moving body. Also, as shown in Fig. 1(2), the mechanical physical quantities occurring between the arm skin and the finger 20 are acquired when a constant amount of water is run over the arm skin to which the cleaner 40 has been applied while the finger 20 is repeatedly moved. That is, in Fig. 1(2), water is run over an object (corresponding to the arm skin) to which a specific agent (corresponding to the cleaner 40) has been applied, the mechanical physical quantities between the object and the moving body (corresponding to the finger 20) are acquired, and the object is rinsed by moving the moving body. When using a measuring tool for measurement, a sensor 30 is provided on a measuring tool (not shown) instead of the finger 20 in Fig. 1, the measuring tool is brought into contact with the skin (skin surface), and the measuring tool is moved so as to slide horizontally against the skin (skin surface), thereby making it possible to measure the generated physical quantity in a similar manner. When measuring by moving the measuring tool in contact with the skin, the movement of the measuring tool may be controlled using a predetermined device (not shown) so that the movement is uniform.

[0018] FIG. 2 shows the electrical signal waveform of the acquired mechanical physical quantity. As shown in FIG. 2, in this embodiment, the vibration occurring between the object and the moving body is acquired as the mechanical physical quantity. In FIG. 2(1), the applied predetermined agent is Sample 1, and in FIG. 2(2), the applied predetermined agent is Sample 3. Comparing FIG. 2(1) and FIG. 2(2), for example, the signal strength is uniform within a certain range in FIG. 2(1), whereas the signal strength is clearly strong and weak in FIG. 2(2), and the shape of the acquired signal waveform differs depending on the applied predetermined agent. Therefore, by analyzing the acquired mechanical physical quantity (signal waveform), it is possible to evaluate the tactile sensation or the physical properties that generate the tactile sensation.

[0019] <Feature Quantification Method> A method for calculating the feature amount will be described. FIG. 3 shows the signal waveform of the acquired mechanical physical quantity. The acquired signal waveform is cut out at predetermined intervals. The dotted frame shown in the signal waveform indicates each predetermined interval. Note that the dotted frame shown in FIG. 3 is only a part, and the dotted frame is actually provided continuously. That is, the cut-out predetermined interval is provided continuously. For each cut-out predetermined interval (first period, second period, third period, etc.), the average signal strength, standard deviation, and spike amount are calculated based on the signal waveform (electrical signal) included in the predetermined interval. The average signal strength is the average of the absolute values ​​of the signal within the predetermined interval, and the standard deviation is obtained from the calculated average. Note that in this embodiment, the average signal strength is described as the "x value", the standard deviation of the signal strength is described as the "y' value", and the spike amount is described as the "y'' value. In this embodiment, the predetermined interval is set to 2 seconds. If the specified interval is too short, there will not be enough time to move the moving body back and forth, and if it is too long, the specified agent will change and there is a possibility that multiple characteristics will be included within the period, so in this embodiment it is set to 2 seconds, but the number is not limited to this.

[0020] Here, a method for calculating spikes and the amount of spikes will be described. Figure 4(1) shows the signal waveform of the vibration generated between the object and the moving body when the specified agent applied is Sample 1, Figure 4(2) shows the signal waveform of the vibration generated between the object and the moving body when the specified agent applied is Sample 2, and Figure 4(3) shows the signal waveform of the vibration generated between the object and the moving body when the specified agent applied is Sample 3. For each signal waveform, attention is paid to the portion where the signal strength exceeds a first predetermined value (for example, 0.2 [V] or -0.2 [V]) and the timing where the signal strength exceeds a second predetermined value (for example, 0.4 [V] or -0.4 [V]). A signal that exceeds a predetermined threshold (in the case of FIG. 4, the first predetermined value and the second predetermined value) is defined as a "spike". In the case of FIG. 4 (1), spikes that exceed the first predetermined value occur from the middle onwards, and spikes that exceed the second predetermined value do not occur. From these, it can be said that the vibration is sharp. In the case of FIG. 4 (2), spikes that exceed the first predetermined value do not occur in the first half, and spikes that exceed the second predetermined value do not occur. From these, it can be said that the density of the signal waveform is high and the vibration is not sharp (the vibration is constant). In the case of FIG. 4 (3), many spikes that exceed the first predetermined value occur, and many spikes that exceed the second predetermined value occur from the middle onwards. From these, it can be said that the density of the signal waveform is sparse, the difference in vibration strength is large, and the vibration is sharp. An index showing the amount of this "spike" is calculated as a "spike amount". The spike amount is calculated by determining whether or not there is a signal with a larger absolute value than a signal within a predetermined time before and after the signal at a certain time, and if there is no signal with a larger absolute value, the signal is regarded as a "spike". The sum of the absolute values ​​of the signals is calculated as the "spike amount" at a certain time. In this embodiment, the certain time is set to a predetermined interval (time surrounded by a dotted line frame) shown in FIG. 3, and the spike amount is calculated with the predetermined time before and after as 0.075 sec. Therefore, in this embodiment, the average for each predetermined time, the standard deviation for each predetermined time, and the spike amount for each predetermined time are calculated from the acquired mechanical physical quantity. Note that these values ​​may be calculated from the value of the acquired mechanical physical quantity itself, or may be calculated from a value (calculated value) after a predetermined noise removal, since the acquired mechanical physical quantity is likely to contain noise. In this embodiment, the spike amount is calculated with the predetermined time before and after as 0.075 sec, but this is not limited to this, and the interval may be shorter or longer than 0.075 sec, and may be determined according to the acquired mechanical physical quantity. As another index representing the amount of spikes, for example, kurtosis calculated from a histogram of the waveform signal in the relevant section may be used. Also, as long as it represents the characteristics of the spikes, it is not limited to this.

[0021] <Indices showing correlations between multiple types of features> It is also possible to evaluate the tactile sensation of an object that changes over time or the physical properties that produce the tactile sensation based on an index showing the correlation between multiple types of feature amounts. Here, an index showing the correlation between multiple types of feature amounts will be described. Each of the multiple types of feature amounts is calculated by the method described in the feature amount calculation method. In addition, when calculating this index, a time weighting value and a perceived intensity coefficient, which will be described later, are taken into consideration in the calculation, making it possible to perform an evaluation that is closer to the tactile sensation that a person actually feels. Here, we will explain a method for evaluating the tactile sensation of an object that changes over time, or the physical properties that produce the tactile sensation, based on an index that indicates the correlation between multiple types of feature amounts.

[0022] An "index showing the correlation between multiple types of feature quantities" is an index showing the relationship between, for example, two types of feature quantities among multiple types of feature quantities, and refers to a regression line created by regression analysis, which is an analytical method of creating a most suitable line and formulating it, a criterion for determining whether or not two types of feature quantities are included in a predetermined confidence interval that is reliable as data, and the correlation between two types of feature quantities in a specific interval with respect to the entire correlation. In addition, a regression line showing the correlation between two types of feature quantities among multiple types of feature quantities obtained from a population may be described as a standard regression line. Note that examples of regression analysis methods include simple regression, multiple regression, nonlinear regression, logistic regression, Bayesian, and the like, and may have or may not have an intercept. In addition, the degree of deviation between the index and any of the multiple types of feature values ​​is also calculated. The "degree of deviation" refers to how much the feature value deviates (diverges) from the regression line (how far it is plotted), how much the feature value deviates from (is not included in) a predetermined confidence interval, etc. There is no limit to the calculation content, such as the specific amount of deviation, the number or percentage of feature values ​​(plots) that deviate by a certain value or more. Then, based on these calculated indices and the degree of deviation, the physical properties that produce the tactile sensation or tactile sensation that changes with the elapsed time after the moving body is brought into contact with the object to which the specified agent is applied are evaluated. The evaluation contents include evaluation based only on the calculated indices, evaluation based on the calculated indices and other information (for example, multiple types of feature amounts of the specified agent, multiple types of feature amounts of other agents, indices calculated from other agents, etc.), evaluation based on the calculated degree of deviation, etc. When the index is a regression line, evaluation is performed based on the slope of the calculated regression line, evaluation based on the length of the regression line, evaluation based on the length from the origin to the regression line (shortest length), etc., drawing and evaluating in a specified coordinate system, evaluation by comparison with a specified reference value, etc. When evaluating based on the regression line and the feature amount, for example, the extent to which the multiple types of feature amounts different from the feature amount used to calculate the regression line deviate from the calculated regression line, the position of the multiple types of feature amounts different from the feature amount used to calculate the regression line are located relative to the calculated regression line, etc. are plotted in a specified coordinate system and evaluated, evaluation is performed by comparison with a specified reference value, etc. In addition, when the index is used as a criterion for determining to what extent two types of feature amounts are included in a predetermined confidence interval, the index is compared with the criterion, and it is evaluated whether or not it is satisfied.

[0023] The correlation between the average and standard deviation calculated from the mechanical physical quantity at each predetermined time will be described with reference to Fig. 5. Here, the regression line is used as an index showing the correlation between the average and the standard deviation, and evaluation is performed based on the degree of deviation from the index. Figure 5(1) shows the correlation between the average and standard deviation when the specified agent applied is Sample 1, Figure 5(2) shows the correlation between the specified agent applied is Sample 2, Figure 5(3) shows the correlation between the specified agent applied is Sample 3, and Figure 5(4) shows the correlation between the average and standard deviation when the specified agent applied is Sample 4. The horizontal axis of each graph is the average, the vertical axis is the standard deviation, and the dotted line is the mean regression line. The coordinate points of each graph are the average and standard deviation calculated at the specified intervals shown in Figure 3. In each sample, the slope is close to that of the mean regression line, and the coefficient of determination is also high in each case, so it can be said that there is a high correlation between the average and standard deviation calculated at the specified intervals. In this manner, the index calculation process calculates a regression line showing the correlation between at least two types of feature amounts among the multiple types of feature amounts, and makes it possible to evaluate the tactile sensation of an object that changes over time or the physical properties that produce that tactile sensation based on any one of the multiple types of feature amounts (e.g., average, standard deviation) and the regression line (e.g., mean regression line).

[0024] Next, a method of calculating a feature amount using coordinate points consisting of the average and standard deviation shown in FIG. 5 and the calculated spike signal amount will be described. FIG. 6 is a diagram showing the correlation between the average and standard deviation when the applied agent is Sample 1, similar to FIG. 5(1). The distance from the origin in FIG. 6 to a coordinate point consisting of the average and standard deviation is calculated. In FIG. 6, "this distance" is shown for one coordinate, but the distance from the origin to the coordinate point consisting of the average and standard deviation is calculated in the same manner for other coordinates. In this embodiment, the distance from the origin to the coordinate point consisting of the average and standard deviation is referred to as "distance from origin" or "x' value."

[0025] FIG. 7 shows a coordinate system (horizontal axis: x' value, vertical axis: y' value) created based on the x' value and y' value. FIG. 7 illustrates an example of a rinsing state in which finger 20 is repeatedly moved while a constant amount of water is being run over the skin of the arm to which detergent 40 as shown in FIG. 1(2) has been applied. Since the x' value (distance from the origin) is the magnitude of vibration in a specified period, if the x' value is smaller than a specified range, the vibration when the moving body is moved is smaller than that of bare skin (state where the specified agent is not applied), and if it is larger than the specified range, the vibration when the moving body is moved is larger than that of bare skin (state where the specified agent is not applied). Also, if the y' value (spike amount) is smaller than a specified range, the moving body slides up when moved compared to bare skin (state where the specified agent is not applied), and if it is larger than the specified range, the moving body does not slide up compared to bare skin (state where the specified agent is not applied). When the x' value and y' value are plotted in such a coordinate system, the upper right of the coordinate system indicates a tighter tactile sensation in the rinsing state, and the lower left of the coordinate system indicates a slimier tactile sensation in the rinsing state. In other words, the upper right of the coordinate system indicates a rinsing sensation that gives a strong feeling of cleaning, the lower left indicates a slimy rinsing sensation, and the center of the coordinate system indicates a rinsing sensation close to that of bare skin. When plotting the x' and y' values ​​in the coordinate system shown in Figure 7, the size of the plot is changed according to the y' value. In other words, the plot makes it possible to grasp the "magnitude of vibration," "amount of spikes," and "non-uniformity of vibration." The subjective evaluation of the given agent (cleaning agent 40) used in FIG. 7 is as follows. Sample 1: Moderate cleaning sensation Sample 2: There is some sluggishness in the first half Sample 3: Strong cleaning sensation Sample 4: Large tactile change

[0026] The results of applying these agents are shown in Figures 8 to 10. The following can be seen from Figures 8 to 10. The first period (Figure 8(1)) is the stage where the detergent is being poured out, and the surface generally feels smooth to the touch. In the fourth period (Figure 8(2)), when all the detergent had been removed, Samples 3 and 4 had a “squeezing” sound. In the fifth period (Figure 9(1)), the “kyuu” becomes stronger in all samples except Sample 2. In the seventh period (Fig. 9(2)), the degree of “tightness” becomes similar to that of bare skin except for Sample 3. During the 9th period (Figure 10), Sample 4 felt smoother than bare skin. In this way, by using highly correlated feature quantities, x' value (distance from the origin), y' value (standard deviation), and y'' value (spike amount), it is possible to grasp the temporal change in the texture of each agent during rinsing. In other words, it is possible to make an evaluation based on the correlation between the magnitude of the mechanical physical quantity (e.g., x' value) and the magnitude of the spike signal amount.

[0027] FIG. 11 shows the characteristics of the rinsing state of each agent shown in FIGS. Sample 1 (Figure 11(1)): The feeling is somewhat tight from the first half to the middle of the rinse, and the feeling of tightness decreases in the second half of the rinse, settling at a medium level (a feeling of rinsing bare skin). Sample 2 (Fig. 11(2)): The vibration intensity is small and the duration of the vibration during rinsing is short, so the rinsing sensation is slimy. Sample 3 (Figure 11(3)): The tight feeling lasts, giving a feeling of cleansing and washing. Sample 4 (Figure 11(4)): The vehicle moves back and forth between the slippery zone and the tight zone. In addition to the "squeak" sensation, when rinsing, you can also feel a "creaking" sensation depending on the magnitude of the vibration.

[0028] Here, we will explain the "creaky" sensation. As mentioned above, the creaky sensation refers to the sensation of rubbing while catching. Using the coordinate system shown in Figure 7, "creaky" can be defined as follows. Creaking is a state in which the vibration is strong and not smooth, and the vibration is not uniform. The opposite tactile sensation to the creaky sensation is the "slippery" sensation. As mentioned above, the slippery sensation refers to a feeling in which there is little catching and the fingers slide easily. Using the coordinate system shown in Figure 7, "slip" can be defined as follows. Slipping is a state in which the vibration pattern is uniform, with strong slippage and weak vibration. Therefore, when the plots shown in Figs. 8 to 10 are located in the upper right corner and are large, it can be said that a strong creaking sensation is felt. As shown in Figures 8 to 10, the change in position of each agent over time plotted in the coordinate system shown in Figure 7 differs depending on the agent, but all move on a straight line. This shows that there is also a high correlation between the x' value (distance from the origin) and the y'' value (spike amount). As such, it can be said that there is a high correlation between the x' value (distance from the origin), y' value (standard deviation), and y'' value (spike amount), and therefore it is possible to display them on a time variation graph with the x' value (distance from the origin) as the representative index.

[0029] A time variation graph representing the x' value (distance from the origin) will be described with reference to FIG. As shown in Figure 12(1), the x value (intensity average) and the y value (standard deviation) are highly correlated, and the x value (distance from the origin), y value (standard deviation), and y value (spike amount) calculated from the x value (intensity average) and y value (standard deviation) are also highly correlated, so by compressing this multidimensional data into one dimension, it is possible to express it as the time dependence of the x value (distance from the origin). Figure 12(2) shows the changes in the mechanical physical quantities of the rinsing state of each agent shown in Figures 8 to 10 as a function of time. From FIG. 12(2), for example, in the case of Sample 1, it can be seen that the time shifts before and after the bare skin zone, and finally stabilizes in the bare skin zone. In this way, it is possible to evaluate the tactile sensation of an object that changes over time or the physical properties that produce that tactile sensation from a graph showing the correlation between multiple types of feature amounts.

[0030] <How to determine tactile values> Next, a method for determining tactile values ​​will be explained. In this method, evaluation is performed taking into consideration that the perceptual meaning of a physical property varies depending on the time of day. Therefore, the tactile value to be determined will take this into consideration. Here, a method for determining a tactile value from multiple types of feature quantities will be explained, but it is also possible to determine a tactile value from a single feature quantity. The agents applied in this embodiment are five types, Sample A, Sample B, Sample C, Sample D, and Sample E. As described above, a plurality of types of feature quantities are calculated, and a graph showing the change in the mechanical physical quantity shown in FIG. 12(2) as a function of time is created. FIG. 13(1) is a graph during washing (sometimes referred to as the "first stage"), and FIG. 13(2) is a graph during rinsing (sometimes referred to as the "second stage"). In this embodiment, the "stage" refers to a stage when operating an operating body that is in contact with an object to which a specific agent has been applied, and there are a first stage, a second stage, and the like, which are periods different from each other in time. By showing the change in the mechanical physical quantity as a function of time as in FIG. 13(1) and FIG. 13(2), it is possible to grasp the tactile sensation that changes over time, for example, when washing Sample D, there is no creaking at the beginning, and the creaking increases as time passes. Also, from the graph in Fig. 13, the horizontal axis can be interpreted as the perceptual meaning of the physical property being different depending on the time period. In other words, the weight of the perceptual meaning can be interpreted as being different. Also, the vertical axis can be interpreted as the distance from the origin being equivalent to the physical property, if it is considered as a feature related to the source signal strength that produces the feature. Taking this into consideration, in this embodiment, a predetermined time weighting value is added to multiple types of feature amounts for each predetermined period according to the time period of the predetermined period, and the feature amount to which the time weighting has been added is used to identify the tactile value of the object.

[0031] Fig. 14(1) is a graph of the second stage (rinsing) shown in Fig. 13(2). Using this graph, a method for determining the time weighting value and the tactile value will be described. As shown in Fig. 14(2) of the graph in Fig. 14(1), the vibration amount at a predetermined time is quantified. In this embodiment, the amount is quantified by determining the area at the predetermined time (corresponding to the trapezoid part marked with 1 in Fig. 14(2)). Fig. 15(1) is a graph of the second stage (rinsing) shown in Fig. 13(2) similarly to Fig. 14(1). As shown in Fig. 15(2), in this embodiment, the vibration amount in each period from 1 to 6 is quantified. Note that, although the periods are from 1 to 6 in this embodiment, the periods are not limited to this, and may be more or less than 6. Fig. 16 shows a method of applying a time weighting factor (sometimes referred to as a "time perception weighting factor") and a perceptual intensity factor to a quantified value (calculated area). As shown in Fig. 16(2), a predetermined time weighting factor (g1 to g6) and a perceptual intensity factor are respectively added to the areas S1 to S6 calculated for each predetermined period, to calculate S1' to S6'.

[0032] The time weighting value will now be described. In this embodiment, the time weighting value may be referred to as a "time weighting coefficient." In the present embodiment, when the mechanical physical quantity is acquired during the repeated movement of the finger 20 and the tactile sensation is evaluated, the degree of influence of the same physical quantity on the tactile sensation differs between immediately after the start of the movement and at the end of the movement. For example, even if the sum of S1 to S6 obtained in FIG. 16(2) is the same, a person will feel differently when the numerical value of S1 is large and when the numerical value of S6 is large. Therefore, by performing a process of adding a time weighting value according to the timing of acquiring the mechanical physical quantity, it becomes possible to specify a tactile sensation value close to the tactile sensation actually felt by a person. In this embodiment, as shown in FIG. 16(2), S1 is multiplied by g1=1, S2 by g2=2, . . . , and S6 by g6=6. In other words, a time weighting value that gradually increases with the passage of time is multiplied.

[0033] Next, we will explain the perceived intensity coefficient. The relationship between the magnitude of a mechanical physical quantity (the magnitude of a stimulus) and the strength with which it is perceived is expressed, for example, by Stevens' power law shown in the following equation (1). (Number 1) R = kS n ...Equation (1) In formula (1), S is the intensity of the stimulus given to a person, and R is the amount of sensation or perception that a person feels through their sense of touch in response to the stimulus intensity S (k is a constant, and n is an exponent determined for each sense). In this way, it is known that even if the intensity of the stimulus given is doubled, the amount of sensation or perception felt by a person does not double, but is proportional to the power of the exponent n. Therefore, in this embodiment, as shown in FIG. 16(2), an exponent n=1 / 3 is added to the determined S1 to S6 to determine a tactile value that is close to the tactile sensation that a person actually feels. Then, as shown in formula (2), the difference between the sum of S1' to S6' calculated in this manner and the reference amount Ss' is used as the tactile value. (Number 2) Index=ΣSi´-ΣSs´...Equation (2) In formula (2), ΣSi' is the sum of the calculated S1' to S6', and ΣSs' is the reference amount. The reference amount is the sum of values ​​calculated in advance from a predetermined agent for which a tactile value is to be obtained and a comparable agent (an agent used as a reference for comparison), obtained by adding a time weighting coefficient and a perceived intensity coefficient to the area at a predetermined time obtained when quantifying the vibration amount. In this way, the mechanical physical quantities are used to determine the tactile value. In addition, a predetermined time weighting value (e.g., g1 to g6) is added to a plurality of feature quantities for each predetermined period (e.g., average, standard deviation, spike amount, distance from the origin, and magnitude of the mechanical physical quantities calculated from these) according to the time zone of the predetermined period, and the feature quantities subjected to the time weighting process are used to determine the tactile value. By using the tactile value calculated in this way for evaluation, it is possible to perform an evaluation close to the tactile sensation that a person actually feels. Therefore, the tactile value determination process not only adds a predetermined time weighting value to multiple types of feature amounts for each predetermined period depending on which predetermined period the time series data of mechanical physical quantities acquired by the multiple types of feature amounts for each predetermined period refers to, but also adds a perception intensity coefficient, which is used when a person touches and perceives an object, to the multiple types of feature amounts for each predetermined period, and it becomes possible to determine the tactile value of the object using the feature amounts to which the perception intensity coefficient has been added, making it possible to perform an evaluation that is closer to the tactile sensation that a person actually feels.

[0034] Next, a method for determining a time weighting value (time perception weighting coefficient) will be described. Note that the method for determining a time weighting value (time perception weighting coefficient) described below is an example, and any method may be adopted, such as machine learning such as deep learning. The skin of a subject is brought into contact with an object to which multiple types of agents (here, Sample A to Sample M) have been applied, and the subject's skin is repeatedly moved to obtain time series data of mechanical physical quantities occurring between the object and the moving body, and multiple types of feature amounts for each predetermined period of time are calculated for each of the multiple types of agents. For each agent, an area S for each predetermined period of time is calculated from the multiple types of feature amounts calculated for each predetermined period of time in the same manner as in the method shown in Figs. 14 and 15. The tactile sensation felt by the subject is obtained as the sensory evaluation result. In this case, it is preferable that the sensory evaluation result satisfies either that the subject is a professional technician knowledgeable about tactile sensation, that the subject is multiple subjects, or that both of these conditions are satisfied. Calculated area for Sample A: S A1 ~S An , time perception weighting coefficient: g1~g n , and the sensory evaluation result: scoreA have the relationship shown in formula (3-A). Similarly, the area calculated for Sample M: M1 ~S Mn , time perception weighting coefficient: g1~g n and the sensory evaluation result: scoreM have the relationship shown in formula (3-M). Note that the exponent 1 / 3 is the perceptual intensity coefficient mentioned above.

number

number

number

number

[0035] In addition, when determining the time weighting value (time perception weighting coefficient), the following is taken into consideration in the case where the change in tactile sensation in a later period due to a stimulus in a previous period, that is, the so-called anchoring effect, is taken into consideration. Calculated area for Sample A: S A1 ~S An , time perception weighting coefficient: g1~g n , and the sensory evaluation result: scoreA have the relationship shown in formula (7-A). Similarly, the area calculated for Sample M: M1 ~S Mn , time perception weighting coefficient: g1~g n and the sensory evaluation result: scoreM have the relationship shown in formula (7-M). Note that the exponent 1 / 3 is the perceptual intensity coefficient described above.

number

number

number

number

[0036] As described above, in this embodiment, the acquiring step involves repeatedly moving the moving body in contact with the object to acquire time series data of mechanical physical quantities occurring between the object and the moving body, the feature amount calculating step involves calculating a plurality of types of feature amounts for each predetermined period from the acquired time series data of the mechanical physical quantities, and the tactile value identifying step involves adding a predetermined time weighting value to the plurality of types of feature amounts depending on which time period in the acquired time series data of the mechanical physical quantities the plurality of types of feature amounts correspond to, and identifying the tactile value of the object using the feature amounts to which the time weighting value has been added; however, it is also possible to add a predetermined time weighting value to the feature amount calculated in the feature amount calculating step or to one of the plurality of feature amounts calculated in the feature amount calculating step depending on which time period in the acquired time series data of the mechanical physical quantities corresponds to, and to identify the tactile value of the object using the feature amount to which the time weighting value has been added. For example, instead of the graph in Fig. 14(1), a predetermined time weighting value may be determined by the above-mentioned method based on a graph in which the horizontal axis represents the time elapsed since the moving object was brought into contact with the object and the moving object was repeatedly moved, and the vertical axis represents the average of the feature values ​​for each predetermined period. The feature values ​​to which the predetermined time weighting value has been added may then be used to specify the tactile value. Even in this way, the tactile value to which the time weighting value has been added can be used for evaluation, making it possible to perform an evaluation close to the tactile sensation that a person actually feels.

[0037] 17 is a graph in which tactile values ​​are determined for each stage and each sample using features to which time weighting values ​​and perceived intensity coefficients have been added by the above-mentioned method, and the determined tactile values ​​are plotted in a coordinate system in which the horizontal axis indicates washing (first stage) and the vertical axis indicates rinsing (second stage). The first tactile value is the tactile value during washing, and the second tactile value is the tactile value during rinsing. By plotting both the first tactile value and the second tactile value in a coordinate system including a first axis (e.g., the horizontal axis) indicating the first tactile value and a second axis (e.g., the vertical axis) indicating the second tactile value, it is possible to evaluate the tactile sensation during a period including washing (first stage) and rinsing (second stage). For example, the following can be said from FIG. Sample A was in a squeaky state during both washing and rinsing, meaning it always felt tight to the touch. Sample B squeaks a little when washing, but when rinsing it feels almost like bare skin. Sample C has almost no squeaking during washing and feels almost like bare skin when rinsing. Sample D had almost no squeaking during washing and rinsing, and had an overall slimy feel. Sample E did not squeak when washing, but had a slightly squeaky, tight texture when rinsing. In this way, the first tactile value during washing and the second tactile value during rinsing can be used to evaluate the tactile feel over a period including washing and rinsing, making it possible to perform an overall tactile evaluation when multiple stages are included.

[0038] <Indices of correlation between multiple types of features, as well as time weighting values ​​and perceptual intensity coefficients> As described above, it is possible to evaluate the tactile sensation of an object that changes over time or the physical properties that produce the tactile sensation based on an index that indicates the correlation between multiple types of feature amounts. When calculating this index, it is possible to perform an evaluation that is closer to the tactile sensation that a person actually feels by taking into consideration the time weighting value and the perceived intensity coefficient. Therefore, for example, the above-mentioned time weighting values ​​and / or perceived intensity coefficients may be added to multiple types of feature quantities calculated at predetermined time intervals from mechanical physical quantities, an index showing a correlation may be calculated using the multiple types of feature quantities that have been added, and the physical properties that produce the tactile sensation or feel of an object that changes over time may be evaluated based on the index and the multiple types of feature quantities that have been added. In this way, the evaluation process evaluates the tactile sensation of an object that changes over time or the physical properties that produce that tactile sensation based on the degree of deviation between any of the feature quantities to which multiple types of time weighting values ​​have been added and an index (e.g., a regression line) and / or any of the feature quantities to which multiple types of time weighting values ​​have been added and an index, thereby making it possible to perform an evaluation that is close to the tactile sensation that a person actually feels.

[0039] <Comparison with multiple types of agents> A method for calculating an index showing a correlation between at least two of a plurality of types of feature amounts calculated from each of a plurality of types of agents, and evaluating the feel of the agent to be evaluated using the index will be described. In this embodiment, the feel of the agent to be evaluated is evaluated based on how much the variation value (standard deviation (y' value)) of the agent to be evaluated deviates from the index (here, the standard regression line is used as the standard index). FIG. 18 is a graph showing coordinate points and standard regression lines consisting of the average and standard deviation shown in FIG. 5. FIG. 18 shows a case where Sample 3 is used as the agent to be evaluated. The "standard regression line" shows the correlation between at least two types of feature amounts (average and standard deviation in the case of FIG. 18) among the multiple types of feature amounts calculated by using multiple types of agents as a population, acquiring time series data of mechanical physical quantities for each agent, and calculating multiple types of feature amounts for each agent. Note that the multiple types of agents included in the population may be any agents, but it is preferable that they have the same purpose (for example, the cleaning target is similar) as the agent to be evaluated. By using a standard regression line calculated from a similar agent, it is possible to appropriately evaluate the tactile sensation of the agent to be evaluated. In addition, it is preferable to use time series data of mechanical physical quantities occurring between an object and a moving body acquired under the same conditions (for example, the site to which the agent is applied is the same, and the environment (method of use) of the agent is the same) except for the different agents. In this way, it is possible to appropriately evaluate that the tactile sensation of the agent to be evaluated is due to the type of agent (due to the physical properties of the agent). As shown in Fig. 18, in this embodiment, the distance in the y-axis direction of the coordinate point consisting of the average and the standard deviation from the standard regression line is calculated. In other words, the distance in the y-axis direction of the coordinate point consisting of the average and the standard deviation from the standard regression line is quantified. In this embodiment, the distance in the y-axis direction between the standard regression line and the coordinate point consisting of the average and the standard deviation is referred to as the "distance in the y-axis direction from the regression line," "deviation from the regression line," or "y value." In this manner, the feature calculation process calculates multiple types of feature amounts for each of the multiple types of agents (e.g., a population of multiple types of agents), the index calculation process further calculates a standard index indicating the correlation between at least two types of feature amounts (e.g., average and standard deviation) among the multiple types of feature amounts of the multiple types of agents, and the degree of deviation between the standard index and any of the feature amounts of the multiple types of agents to be evaluated, and the evaluation process makes it possible to evaluate the feel of an object or the physical properties that create the feel when the agent to be evaluated is applied based on the multiple types of feature amounts (e.g., standard deviation) of the agent to be evaluated (e.g., Sample 3) and the standard index.

[0040] 19 to 21 are graphs showing the relationship between the "distance in the y-axis direction from the standard regression line" calculated by the method shown in FIG. 18 and the "magnitude of vibration" over time (for each specified period). The horizontal axis of FIG. 19 to FIG. 21 is the magnitude of vibration (x' value = distance from the origin), and the vertical axis is the distance in the y-axis direction (y value) from the standard regression line (regression line). The vertical axis indicates non-uniformity by being the distance in the y-axis direction from the standard regression line. Samples 1 to 4 are plotted in each graph. The difference in the size of the plot indicates the difference in the magnitude of the y' value (standard deviation). Note that Samples 1 to 4 are the same as the specified agent used in FIG. 7, and therefore the subjective evaluation is also the same. Here, if the distance in the y-axis direction from the standard regression line is a positive value (deviation in the positive direction), it can be said that the load on the object caused by applying the specified agent is large, and if the distance in the y-axis direction from the standard regression line is a negative value (deviation in the negative direction), it can be said that the load on the object caused by applying the specified agent is small. In other words, by using Figures 19 to 21, it is possible to evaluate the degree of load on the skin when washing and rinsing with a specified agent. As shown in Figures 19 to 21, the following can be seen by comparing certain agents. Sample 1 is somewhat harsh on the skin at the beginning of rinsing, but as time passes the harshness decreases. For Sample 2, the y value was negative for most of the time from the start to the end of rinsing, which indicates that among the products compared, Sample 2 tends to place a lower burden on the skin than the other products. For Sample 3, the y value was positive for most of the time from the start to the end of rinsing, which indicates that among the products compared, Sample 3 tends to place a greater burden on the skin than the other products. In Sample 4, the magnitude of vibration changes from the start to the end of rinsing, but the y-value is negative for most of the time, indicating a tendency for the burden on the skin to be low.

[0041] In this manner, a plurality of types of agents including the agent to be evaluated are each applied to an object, the acquisition process acquires time series data of mechanical physical quantities for each of the plurality of types of agents, the feature calculation process calculates a plurality of types of feature amounts for each of the plurality of types of agents, the index calculation process further calculates a standard index (e.g., a standard regression line) indicating a correlation between at least two types of feature amounts (e.g., average and standard deviation) among the plurality of types of feature amounts of the plurality of types of agents (e.g., Samples 1 to 4) and the degree of deviation between the standard index and any of the feature amounts of the plurality of types of agents to be evaluated (e.g., see FIG. 18), and the evaluation process can evaluate the feel of the object when the agent to be evaluated is applied or the physical properties that produce that feel based on the degree of deviation.

[0042] In addition, by adding a time weighting value and / or a perceived intensity coefficient to multiple types of feature values ​​calculated from each of the multiple types of agents used to calculate the standard index, and using the feature values ​​after the adding process, it is possible to perform an evaluation that takes the anchoring effect into account.

[0043] FIG. 22 shows the process flow of the above-mentioned tactile evaluation method. The step (step S100) is a step of acquiring time series data of mechanical physical quantities. The acquisition method is as described above. The step (step S110) is a step of calculating feature quantities in the acquired time series data of mechanical physical quantities. The calculation method for each feature quantity is as described above. In this embodiment, the average signal strength (x value), standard deviation of signal strength (y' value), spike amount (y'' value), and distance from the origin (x' value) are calculated for each predetermined period (e.g., every 2 seconds). The step (step S120) is a step of specifying a tactile value using the calculated feature quantities. In this embodiment, as described above, each tactile value is specified using a plurality of types of feature quantities, and therefore the method of specifying each tactile value is as described above. In this embodiment, a tactile value that has been subjected to additional processing with a time weighting value and a perceived intensity coefficient is specified. The step (Step S130) is a step of performing an addition process of a time weighting value and / or a perceptual intensity coefficient on the calculated plurality of types of feature quantities, and calculating an index using the plurality of types of feature quantities after the addition process. In the step (step S140), the tactile sensation or the physical property that produces the tactile sensation is evaluated using the specified tactile value. The evaluation method is as described above. For example, the first tactile value and the second tactile value are plotted on the coordinate system shown in FIG. 17 and evaluated.

[0044] <Tactile evaluation device> The tactile evaluation device 200 will be described with reference to FIG. The tactile evaluation device 200 in this embodiment is composed of a moving object 110, an acquisition unit 120, and a feature calculation unit 130. It is preferable that the device further comprises an evaluation unit 150, an index calculation unit 160, and a display unit 170. The device further comprises an information processing terminal 100 capable of executing various processes, and the information processing terminal 100 comprises the feature calculation unit 130, the tactile value determination unit 140, the evaluation unit 150, and the index calculation unit 160. The information processing terminal 100 is a general-purpose personal computer, and comprises input devices such as a keyboard and a pointing device, an arithmetic processing unit (e.g., a CPU (Central Processing Unit), an MPU (Micro Processing Unit), etc.), a storage unit, etc. The information processing terminal 100 further comprises a display unit 170 (display device), but the display unit 170 may be provided outside the information processing terminal 100 and connected via a network.

[0045] The moving object 110 is a finger 20, a palm, or a measurement tool that is brought into contact with the surface of the skin. The acquiring unit 120 is a means (corresponding to an acquiring means) for acquiring, over time, a mechanical physical quantity generated by moving the finger 20, using the sensor 30 attached to the finger 20 shown in Fig. 1. The mechanical physical quantity (electrical signal) acquired by the sensor 30 is configured so that it can be acquired by the information processing terminal 100 via a network line, a medium, etc. The feature amount calculation unit 130 is a unit (corresponding to a feature amount calculation unit) that calculates a feature amount from the time series data of the mechanical physical amount acquired by the acquisition unit 120. The calculation method is the same as that described above. The tactile value determination section 140 is a means for determining a tactile value using the calculated feature amount (corresponding to a tactile value determination means). The method of determination is the same as that described above. The evaluation unit 150 is a means for evaluating the tactile sensation or the physical properties that produce the tactile sensation, using the tactile value determined by the tactile sensation determination unit 140. The evaluation contents are the same as those described above. It is preferable that the evaluation results by the evaluation unit 150 are displayed on the display unit 170 so that the evaluator can easily understand them. The index calculation unit 160 is a means for calculating an index using the calculated feature amount. The calculation method is the same as that described above. The display unit 170 displays the evaluation results by the evaluation unit 150, for example, the graph shown in Fig. 17. Also, for example, by displaying the graphs shown in Fig. 11 and Fig. 13, it is possible to evaluate the tactile sensation of each agent during each period of cleaning or rinsing. Also, for example, by displaying the graphs shown in Fig. 19 to Fig. 21, it is possible to evaluate the tactile sensation of the agent to be evaluated using multiple types of agents. The memory unit of the information processing terminal 100 stores a program for executing the evaluation method described above, in which the program acquires the mechanical physical quantities acquired by the acquisition unit 120, causes the feature calculation unit 130 to calculate the feature quantities from the acquired mechanical physical quantities, causes the tactile value identification unit 140 to identify the tactile value, and causes the evaluation unit 150 to display the evaluation result on the display unit 170.

[0046] As described above, the present invention has been described by showing specific embodiments, but the present invention is not limited to the above-described embodiments and also includes various modifications, improvements, and other aspects as long as the object of the present invention is achieved. <Modification> In this embodiment, a time weighting coefficient that increases linearly with time is used, but this is not limited to this. For example, a time weighting coefficient that increases quadratically may be used. Since the end of a motion has a particularly high contribution to a person's sensation, by using a time weighting coefficient that increases quadratically, the mechanical physical quantity at the end of the motion can be weighted for evaluation, making it possible to perform an evaluation close to the tactile sensation that a person actually feels.

[0047] In addition, in this embodiment, the gradual increase rate of the time weighting coefficient in the first stage and the gradual increase rate of the time weighting coefficient in the second stage are set to be the same, but this is not limited to the above, and the gradual increase rate of the time weighting coefficient in the first stage and the gradual increase rate of the time weighting coefficient in the second stage may be set to different values.

[0048] In addition, in this embodiment, the time weighting coefficient in the first stage and the time weighting coefficient in the second stage are increased with time, but this is not limited to this. For example, since stimuli at the start and end of a motion have a high impact on a person, the coefficients may be gradually increased within a predetermined range from the initial value, and then may not be increased or decreased for a while, and then may be gradually increased. Specifically, for example, g1=1, g2=2, g3=3, g4=3, g5=3, and g6=4 may be used. In this way, when there is an impactful stimulus at the start and end of a motion, it is possible to perform an evaluation close to the tactile sensation that a person actually feels.

[0049] Furthermore, in this embodiment, the first tactile value is a time weighting coefficient in the first stage (e.g., gradually increasing from g1 (=1) to g6 (=6)), and the second tactile value is a time weighting coefficient in the second stage (e.g., gradually increasing from g1 (=1) to g6 (=6)), so the time weighting coefficient when determining the second tactile value is a value associated with the time weighting coefficient of the first stage. The "value associated with the time weighting coefficient of the first stage" means, as in this embodiment, aspect 1) the time weighting coefficient g1 (=1) of the second stage is set to the same time weighting coefficient as the time weighting coefficient g1 (=1) of the first stage, i.e., the initial value (g1) of the time weighting coefficient of the second stage is set to the same as the time weighting coefficient of the first stage and gradually increases at the same rate; aspect 2) the time weighting coefficient g1 of the second stage is set to a time weighting coefficient (e.g., 0.5) smaller than the time weighting coefficient g1 (=1) of the first stage, i.e., the initial value of the time weighting coefficient of the second stage is smaller than the initial value of the time weighting coefficient of the first stage; aspect 3) the time weighting coefficient g1 of the second stage is set to a time weighting coefficient (e.g., g5) before the end of the time weighting coefficient of the first stage. (=5)), i.e., the initial value of the time weighting coefficient of the second stage is the same as the time weighting coefficient before the final value of the time weighting coefficient of the first stage; aspect 4) g1 of the time weighting coefficient of the second stage is a time weighting coefficient (e.g., 5.5) smaller than the final time weighting coefficient (g6 (=6)) of the time weighting coefficient of the first stage, i.e., the initial value of the time weighting coefficient of the second stage is smaller than the final value of the time weighting coefficient of the first stage; aspect 5) g1 of the time weighting coefficient of the second stage is a time weighting coefficient (e.g., 7) greater than or equal to the final time weighting coefficient (g6 (=6)) of the time weighting coefficient of the first stage, i.e., the initial value of the time weighting coefficient of the second stage is greater than the final value of the time weighting coefficient of the first stage. As in aspects 1) to 4), the reason why g1, the time weighting coefficient for the second stage, is a time weighting coefficient smaller than the final time weighting coefficient (g6 (=6)) of the first stage is that the specified agents are different between the first and second stages, the environments of the objects to which the specified agents are applied are different, or the objects are different, and therefore the tactile impression is reset when the second stage is reached. In this way, by specifying the first tactile value and the second tactile value using a value of g1, the time weighting coefficient for the second stage, smaller than the final time weighting coefficient (g6 (=6)) of the time weighting coefficients for the first stage, it is possible to perform an evaluation that is closer to the tactile sensation that a person actually feels. Furthermore, as in aspect 3), when the time weighting coefficient g1 of the second stage is set to a time weighting coefficient before the end of the time weighting coefficient of the first stage (for example, g5 (=5)), there is a high possibility that the tactile impression will not be completely reset when changing from the first stage to the second stage, for example, if the first stage and the second stage are both in a rinsing state, and the first stage is rinsed with water (lower than the surface temperature of the skin) and the second stage is rinsed with lukewarm water. In addition, as in aspect 5), the time weighting coefficient g1 of the second stage is set to a time weighting coefficient (e.g., 7) equal to or greater than the final time weighting coefficient of the first stage (g6 (=6)). For example, when the first and second stages are both in a massage state and the specified agent applied in the first and second stages is different (e.g., oil in the first stage and cream in the second stage), in order to continue the tactile impression of the first stage, the first tactile value and the second tactile value are identified using a time weighting coefficient (e.g., 7) equal to or greater than the final time weighting coefficient of the first stage (g6 (=6)). This makes it possible to perform an evaluation that is closer to the tactile sensation that a person actually feels. In this way, depending on the contents of the first and second stages, it is preferable that the predetermined time weighting value of the second stage is a value related to the predetermined time weighting value of the first stage.

[0050] In this embodiment, the tactile value is determined by adding a time weighting coefficient and a perceived intensity coefficient to the areas S1 to S6 calculated for each predetermined period shown in FIG. 16(2), but this is not limited to the above. For example, the tactile value may be determined without including some values. In addition, the areas obtained in a certain period and the previous period in a certain stage may be compared, and the area value of the period may be biased according to the difference or ratio. For example, if the area in the previous period is smaller than the area in the certain period, the area in the certain period may be multiplied by the ratio to the area in the previous period to determine the tactile value.

[0051] In this embodiment, when determining the tactile value, the amount of vibration at a predetermined time is quantified by finding the area at the predetermined time (corresponding to the trapezoid part marked with 1 in FIG. 14(2)) as shown in FIG. 14(2), but this is not limited to this. For example, the tactile value may be determined using an appropriate basis function such as a rectangular basis function for each predetermined time in the graph shown in FIG. 14(1), a rectangle, a sine, a cosine, or the like.

[0052] In this embodiment, the first stage and the second stage are continuous in time, but this is not limited to the above. For example, the second stage may occur after a predetermined period of time has elapsed after the first stage. Since the first tactile value and the second tactile value are measured for each stage, a series of tactile evaluations can be performed for a period including the first stage and the second stage, even if the second stage occurs after a predetermined period of time has elapsed after the first stage. In addition, the sensory evaluation results for the first stage and the sensory evaluation results for the second stage may be obtained, and the time weighting coefficients for the first stage and the second stage may be calculated using the above-mentioned formulas (3-A) to (6) or (7-A) to (10).

[0053] In this embodiment, the evaluation was performed for the period including the first stage and the second stage, but the number of stages is not limited to this. For example, the state in which a makeup remover is applied as a specific agent is the first stage, the state in which a face wash is applied is the second stage, and the state in which a lotion is applied is the third stage, and it is also possible to evaluate the tactile sensation during the period including the first stage, the second stage, and the third stage. In this case, for example, the first tactile sensation value, the second tactile sensation value, and the third tactile sensation value can be plotted on a three-axis coordinate system to visually grasp the overall characteristics consisting of the characteristics of the three stages, and the tactile sensation or the physical properties that produce the tactile sensation can be evaluated. That is, by plotting each tactile sensation value (e.g., the first tactile sensation value, the second tactile sensation value, and the third tactile sensation value) specified for each of the multiple stages (e.g., the first stage, the second stage, and the third stage) that a specific stage has on a specific coordinate system (e.g., the three-axis coordinate system), it is possible to evaluate the tactile sensation during the period including the multiple stages or the physical properties that produce the tactile sensation.

[0054] In this embodiment, the predetermined threshold for determining a "spike" may be a predetermined percentage of the signal strength of the acquired signal waveform (for example, a predetermined percentage of the maximum signal strength).

[0055] In this embodiment, the perceived intensity coefficient is determined based on Stevens' power law, but in addition to this, it may be determined based on the environment when a person touches and perceives an object. Here, the "environment" may be determined based on, for example, the temperature of water (hot water) used for rinsing, or based on the room temperature and humidity in which the object exists when the person touches the object. In other words, it is preferable to determine the perceived intensity coefficient based on the state (position) of the object and the operating body when the person touches and perceives the object, or the state in which a specific agent is placed when the person touches and perceives the object. In this way, it is possible to perform an evaluation that is closer to the tactile sensation that a person actually feels.

[0056] In the present embodiment, time-series data of mechanical physical quantities is obtained for each of a plurality of types of agents including the agent to be evaluated, a standard index showing the correlation between at least two of a plurality of types of feature quantities for each of the plurality of types of agents is calculated, and the texture of an object or the physical properties that produce the tactile sensation when the agent to be evaluated is applied is evaluated; however, the present invention is not limited to this. For example, the object is a plurality of types of object including an object to be evaluated (e.g., a face), and a predetermined agent is applied to each of the plurality of types of object. For example, the plurality of types of object is a head, a face, an arm, etc. Time series data of mechanical physical quantities is acquired for each of the plurality of types of object, and a plurality of types of feature quantities are calculated for each of the plurality of types of object. The index calculation step calculates a standard index indicating a correlation between at least two types of feature quantities (e.g., an average and a standard deviation) among the plurality of types of feature quantities of the plurality of types of object, and a deviation degree between the standard index and any one of the plurality of types of feature quantities (e.g., a standard deviation), and the evaluation step evaluates the physical properties of the object to be evaluated or the touch that produces the touch when the predetermined agent is applied based on the deviation degree. In this way, by making the population a plurality of types of object, it is possible to evaluate the touch of the object to be evaluated or the physical properties of the touch that produces the touch when the predetermined agent is applied to the object to be evaluated, and for example, it is possible to evaluate the load of the predetermined agent on the object when the predetermined agent (the same agent) is applied to the face. Moreover, when the index is a regression line, the following may be performed. The objects are a plurality of types of objects including an object to be evaluated (e.g., a face), and a predetermined agent is applied to each of the plurality of types of objects. The acquisition step acquires time-series data of mechanical physical quantities for each of the plurality of types of objects, and the feature amount calculation step calculates a plurality of types of feature amounts for each of the plurality of types of objects. The index calculation step calculates a regression line (index) showing a correlation between at least two types of feature amounts among the plurality of types of feature amounts, and further calculates a standard regression line showing a correlation between at least two types of feature amounts (e.g., an average and a standard deviation) among the plurality of types of feature amounts of the plurality of types of objects. The evaluation step evaluates the physical properties of the object to be evaluated or the tactile property that produces the tactile sensation when a predetermined agent is applied based on any one of the plurality of types of feature amounts of the object to be evaluated and the standard regression line. In this way, even when the index is the standard regression line, by using a plurality of types of objects as the population, it is possible to evaluate the feel of the object to be evaluated when a specified agent is applied to the object to be evaluated, or the physical properties of the feel that produces the feel. For example, it is possible to evaluate the load of a specified agent on an object when the specified agent (the same agent) is applied to the face. In addition, by adding a time weighting value and / or a perceived intensity coefficient to multiple types of feature values ​​calculated from each of the multiple types of agents used to calculate the standard index, and using the feature values ​​after the adding process, it is possible to perform an evaluation that takes the anchoring effect into account.

[0057] Also, for example, the object is the environment of multiple types of objects including the environment of the object to be evaluated (for example, rinsing with water), and the predetermined agent is applied in each state of the environment of the multiple types of objects. For example, the environment of the multiple types of objects is rinsed with water, with lukewarm water, with hot water, etc. Time series data of mechanical physical quantities is acquired for each environment of the multiple types of objects, and multiple types of feature quantities are calculated for each environment of the multiple types of objects. The index calculation step calculates a standard index indicating a correlation between at least two types of feature quantities (for example, average and standard deviation) among the multiple types of feature quantities of the environment of the multiple types of objects, and a deviation degree between the standard index and any of the multiple types of feature quantities (for example, standard deviation), and the evaluation step evaluates the tactile feel of the object to be evaluated when the predetermined agent is applied or the physical properties that produce the tactile feel based on the deviation degree. Here, the "environment" refers to the existence state of the object and / or the moving body, such as the temperature of water (hot water) when rinsing, the room temperature and humidity where the object is located, etc. In this way, by making the population into an environment of multiple types of objects, it is possible to evaluate the feel of the object being evaluated or the physical properties that produce the feel when a specified agent is applied to the environment of the object being evaluated. For example, it is possible to evaluate the load of a specified agent on an object when the specified agent (the same agent) is applied to the object and then rinsed with water. Moreover, when the index is a regression line, the following may be performed. There are a plurality of types of object environments including an environment to be evaluated (e.g., rinsed with water), and a predetermined agent is applied in each state of the plurality of types of object environments. The acquisition step acquires time series data of mechanical physical quantities for each of the plurality of types of environments, and the feature amount calculation step calculates a plurality of types of feature amounts for each of the plurality of types of environments. The index calculation step calculates a regression line (index) showing a correlation between at least two types of feature amounts among the plurality of types of feature amounts, and further calculates a standard regression line showing a correlation between at least two types of feature amounts (e.g., average and standard deviation) among the plurality of types of feature amounts for the plurality of types of environments. The evaluation step evaluates the tactile feel of the object or the physical properties that produce the tactile feel when a predetermined agent is applied in the environment to be evaluated based on any one of the plurality of types of feature amounts of the environment to be evaluated and the standard regression line. In this way, even when the index is a standard regression line, by setting the population to be the environments of multiple types of objects, it is possible to evaluate the feel of the object to be evaluated or the physical properties that produce the feel when a specified agent is applied in the environment of the object to be evaluated. For example, it is possible to evaluate the load of the specified agent on the object when a specified agent (the same agent) is applied to the object and then rinsed with water. In addition, by adding a time weighting value and / or a perceived intensity coefficient to multiple types of feature values ​​calculated from each of the multiple types of agents used to calculate the standard index, and using the feature values ​​after the adding process, it is possible to perform an evaluation that takes the anchoring effect into account.

[0058] In this embodiment, the degree of deviation of the variation in the y-axis direction from the standard regression line is quantified, and the tactile sensation of the object when the agent to be evaluated is applied, the load caused by the agent to be evaluated, or the physical properties that generate the tactile sensation are evaluated based on the value, but this is not limited to this. For example, it is also possible to quantify the degree of deviation of the spike signal amount in the y-axis direction, and to evaluate the tactile sensation of the object when the agent to be evaluated is applied, the load caused by the agent to be evaluated, or the physical properties that generate the tactile sensation based on the value. Note that, when quantifying the degree of deviation, it is also possible to focus on the x-axis direction instead of the y-axis direction to quantify.

[0059] In this embodiment, the deviation of the variation (y' value) in the y-axis direction from the standard regression line is quantified, and a graph is created for each predetermined period (e.g., the first period, the fourth period, etc.) as shown in Figs. 19 to 21, and the tactile sensation of the object that changes over time when the agent to be evaluated is applied, the load caused by the agent to be evaluated, or the physical properties that produce the tactile sensation are evaluated, but this is not limited to this. For example, the sum of the quantified values ​​(deviation amounts) for each predetermined period may be calculated, and the total load on the object caused by the agent to be evaluated from the start to the end of the movement of the moving body may be calculated. In addition, by comparing the total load on the object caused by the agent to be evaluated from the start to the end of the movement of the moving body with the deviation amount for each predetermined period, it is possible to evaluate what proportion of the total load the load on the object caused by the agent to be evaluated changes over time.

[0060] In this embodiment, as an example of evaluation based on one of a plurality of types of feature quantities and a regression line, the case of the average regression line and feature quantity (average and standard deviation) and the case of the standard regression line and feature quantities (average and standard deviation) of a plurality of types of agents were described. That is, the evaluation of the texture of the object or the physical properties that produce the texture was performed based on the feature quantity used to create the regression line by regression analysis and the created regression line, but this is not limited thereto. For example, the evaluation of the texture of the object or the physical properties that produce the texture may be performed based on a feature quantity other than the feature quantity used to create the regression line and the regression line. In addition, in this embodiment, the evaluation of the texture of the object or the physical properties that produce the texture was performed based on the position of the feature quantity for each predetermined period (for example, the coordinates of the first period, the coordinates of the fourth period, etc., described as "time series plot") relative to the regression line, but this is not limited thereto. For example, the distance that the time series plot has moved on the regression line (total distance moved on the regression line), the initial position of the time series plot on the regression line, the final position of the time series plot on the regression line, the direction in which the time series plot starts moving from the initial position on the regression line (starting direction), the direction from which the time series plot has moved to the final position on the regression line, the sum of the deviations (or the sum of squares of the deviations) between the standard regression line and the regression line to be evaluated (individual regression line), etc. These values ​​may be used individually for evaluation, or a combination of these values ​​may be used for evaluation. [Explanation of symbols]

[0061] 10 Arithmetic unit 20 fingers 30 Sensors 40 Cleaning Agent 100 Information processing terminal 110 Action 120 Acquisition Department 130 Feature Calculation Unit 140 Tactile value determination unit 150 Evaluation Department 160 Indicator calculation section 170 Display section 200 Tactile evaluation device

Claims

1. An acquisition step involves repeatedly moving a moving body that is in contact with an object to acquire time-series data of mechanical physical quantities generated between the object and the moving body, A feature calculation step which calculates feature quantities at predetermined intervals from the time-series data of the acquired mechanical physical quantities, A tactile evaluation method comprising: a step of adding a predetermined time weighting value to the feature quantity according to which time period in the acquired time series data of the mechanical physical quantity the feature quantity corresponds to; and a step of identifying the tactile value of the object using the feature quantity to which the time weighting value has been added.

2. The tactile evaluation method according to claim 1, wherein the feature quantity is one of the mechanical physical quantity, the statistical quantity of the mechanical physical quantity, the spike signal quantity which is greater than or equal to a predetermined threshold value of the mechanical physical quantity or the statistical quantity of the mechanical physical quantity, and the average value or variability which is the statistical quantity of the mechanical physical quantity.

3. The feature calculation step involves calculating multiple types of features from the acquired time-series data of the mechanical physical quantities at predetermined intervals. A step to calculate an index that shows the correlation between at least two types of features from among the features that have been processed with the aforementioned multiple types of time-weighted values, A method for evaluating tactile sensation according to claim 1 or 2, further comprising an evaluation step of evaluating the tactile sensation of an object or the physical properties that produce such a tactile sensation, which change over time based on the calculated index.

4. The aforementioned index calculation step calculates the degree of discrepancy between the index and one of the multiple types of feature quantities. The tactile sensation evaluation method according to claim 3, wherein the evaluation step evaluates the tactile sensation of the object or the physical properties that produce the tactile sensation, which change over time based on the degree of displacement.

5. The tactile sensation evaluation method according to claim 3, wherein the evaluation step evaluates the tactile sensation of an object or the physical properties that produce such a tactile sensation, which change over time, based on any of the feature quantities obtained by adding the multiple types of time-weighted values ​​and the index.

6. The aforementioned multiple types of feature quantities include at least one of the mechanical physical quantity, the statistic of the mechanical physical quantity, the spike signal quantity which is a value greater than or equal to a predetermined threshold of the mechanical physical quantity or the statistic of the mechanical physical quantity, and a feature quantity calculated from the mean or variability which is the statistic of the mechanical physical quantity. The tactile sensation evaluation method according to claim 3, wherein the evaluation step evaluates the tactile sensation of an object or the physical properties that produce said tactile sensation based on the correlation between the magnitude of the mechanical physical quantity and the magnitude of the spike signal quantity.

7. The tactile sensation evaluation method according to claim 1 or 2, wherein the tactile sensation value identification step involves adding a perceptual intensity coefficient to the feature quantity for each predetermined period, and using the feature quantity to which the perceptual intensity coefficient has been added to identify the tactile sensation value of the object.

8. The tactile evaluation method according to claim 7, wherein the perceptual intensity coefficient is determined based on the environment in which the person touches and perceives the object.

9. Multiple types of agents, including the agent to be evaluated, are applied to the aforementioned object. The acquisition step involves acquiring time-series data of the mechanical physical quantities for each of the multiple types of agents. The feature calculation step involves calculating the multiple types of feature quantities for each of the multiple types of agents, The aforementioned index calculation process is as follows: A standard index that shows the correlation between at least two of the feature quantities of the aforementioned multiple types of agents, Further calculate the degree of discrepancy between the standard index and any of the multiple types of characteristic quantities of the agent being evaluated. The tactile sensation evaluation method according to claim 3, wherein the evaluation step evaluates the tactile sensation of the object or the physical properties that produce the tactile sensation when the agent to be evaluated is applied based on the degree of deviation.

10. The subject's skin is repeatedly moved while in contact with an object to which multiple types of agents have been applied, and time-series data of the mechanical physical quantities generated between the object and the moving body are obtained. From the time-series data of the mechanical and physical quantities for each of the multiple types of agents obtained, feature quantities are calculated at predetermined intervals. The sensory evaluation results obtained when the subject repeatedly performs the aforementioned skin movements are acquired. Based on the aforementioned feature quantities and the sensory evaluation results, the predetermined time-weighted values ​​are calculated. The tactile evaluation method according to claim 1 or 2, characterized in that the tactile value identification step uses the calculated time-weighted value to perform the additional processing.

11. The tactile evaluation method according to claim 10, characterized in that the predetermined time weighting value is calculated based on the feature quantity and the sensory evaluation result, as well as which time period in the time series data of the acquired mechanical physical quantity the feature quantity corresponds to, and the feature quantity corresponding to a time period earlier than that time period.

12. An acquisition means for repeatedly moving a moving body that is in contact with an object to acquire time-series data of mechanical physical quantities generated between the object and the moving body, A feature calculation means that calculates feature quantities at predetermined intervals from the time-series data of the acquired mechanical physical quantities, A tactile evaluation device comprising: a means for adding a predetermined time weighting value to the feature quantity according to which time period in the acquired time series data of the mechanical physical quantity the feature quantity corresponds to; and a means for identifying the tactile value of the object using the feature quantity to which the time weighting value has been added.

Images