Tactile evaluating system and method

Abstract

A sensor assembly for evaluating a property of a surface, whilst the sensor assembly is in motion relative to the surface, wherein the sensor assembly comprises: a) a housing for: b) an elastomeric sensor, which comprises i) a stem, ii) a domed head, c) a source of illumination, d) a camera; wherein the housing comprises a means to retain the elastomeric sensor in contact with the source of illumination, a system further comprising e) a surface; f) a data processing unit, g) a means to move the sensor assembly along the surface; wherein the surface comprises multiple elements that can move relative to each other when contacted with the sensor; and a method of evaluating a property of a surface using the system.

Description

[0001] TACTILE EVALUATING SYSTEM AND METHOD

[0002] Field of the Invention

[0003] The present invention relates to a system comprising a device for evaluation of tactile sensations of surfaces and a method of evaluating a tactile sensation on a surface using the device. The invention has particular application in the field of hair and skin.

[0004] Background of the Invention

[0005] The invention provides a system comprising a tactile sensation evaluating device capable of acquiring optical data during physical contact of a sensor with a surface, that objectively relate a sensory evaluation of the surface and of an optional treatment composition applied to the surface by comparing the sensory evaluation with measured data. The evaluation is performed during relative motion between the sensor and the surface.

[0006] Sensory evaluation devices that can differentiate a variety of properties of surfaces are known. The following are concerned with devices for assessing the condition of hair and / or skin and converting a signal from a sensor:

[0007] JP020-197481 A (Univ Hiroshima) discloses a tactile sense evaluation system, method and program. The system comprises: (i) an image data acquisition unit that acquires image data of a contact surface between an analyte and a pseudo finger when the analyte or the pseudo finger is operated in a state where the analyte is contacted with the pseudo finger being an elastic body; (ii) a vibration information acquisition unit that acquires vibration information; and (iii) an evaluation unit that analyzes the eccentricity apparent within the image data combined with the vibration information to evaluate a tactile sense. Tactile sensation is evaluated by dividing into a composite confidence interval to be performed and a vibration confidence interval to be evaluated based on the vibration information. The surface may be leather, rubber or paper.

[0008] JP 2020-180934 A (Kao) discloses a tactile sensation evaluating device for evaluating a tactile sensation pertaining to application of a preparation, by means of tactile sensation evaluation information representing the tactile sensation, provided with: (i) an application surface to which a preparation under evaluation is applied by an evaluator; (ii) a measuring unit for measuring a physical quantity representing a physical effect occurring in the preparation when the evaluator applies the preparation to the application surface; (iii) a tactile sensation evaluation information input unit for accepting the tactile sensation evaluation information input by the evaluator while applying the preparation to the application surface; and (iv) an information recording unit for recording, in association with one another, measurement information pertaining to the physical quantity measured by the measuring unit, and the tactile sensation evaluation information input at the timing at which the measurement information was measured. Here, physical data (friction) is recorded at the same time that the evaluator records tactile sensory data.

[0009] WO 19 / 039466 (Shiseido) discloses a method for evaluating skin feel of a cosmetic comprising: a vibration detection step for chronologically detecting vibrations generated by moving a movement body which is kept in contact with a cosmetic applied over a target application surface; a data generation step for generating data related to chronological changes in the frequency spectrum of the vibrations which were detected; and an evaluation step for evaluating the skin feel of the cosmetic using the generated data.

[0010] JP 2017-009489 A (Kao) discloses a method of measuring haptic sense of skin or providing haptic sensation on skin using a device that generates vibration, where the measurement is made after applying a composition containing a moisturizing ingredient on skin to be measured.

[0011] Despite the prior art, a need remains for richer information sets to enable production of more accurate models. Devices and methods that are capable of more accurate measurements and distinguishing finer differences between surfaces, as well as changes in surfaces following different treatments, will enable the delivery of superior products.

[0012] Most assessment of tactile properties of a surface will rely on a force measurement, such as friction or ease of combing. We have now found that an imaging sensor combined with a data processing system, such as a 3D convolutional neural network, provides an improved assessment of tactile properties by capturing information whilst in tactile contact and relative motion with the surface. Further, the evaluation is capable of collecting information on multicomponent surfaces comprising individual elements that are free to move relative to each other. This provides an evaluation that is much more realistic and more closely resembles in vivo surface evaluation.

[0013] Consumers use touch or feel to evaluate characteristics of surfaces, for example, softness, friction (smoothness or roughness), stiffness, abrasion, lubricity, damage and cleanliness. The surface used in the present invention comprises multiple elements that can move relative to each other when contacted with the sensor. The present invention enables accurate replication of such consumer evaluation by generating information from the collective individual motion of the multiple elements of the surface. This information is used to build models of tactile assessments based on known consumer responses.

[0014] Summary of the Invention

[0015] In a first aspect, the invention provides a sensor assembly for evaluating a property of a surface, wherein the sensor assembly comprises: a) a housing for: b) an elastomeric sensor, which comprises i) a stem ii) a domed head c) a source of illumination d) a camera; wherein the housing comprises a means to retain the elastomeric sensor in contact with the source of illumination; and wherein the sensor is configured to move relative to the surface; and wherein the sensor does not comprise a reflective coating.

[0016] In a second aspect, the invention provides a system comprising the sensor assembly of the first aspect and additionally comprising: e) a surface; f) a data processing unit, preferably employing a 3D convolutional neural network; and g) a means to move the sensor assembly along the surface; wherein the surface comprises multiple elements that can move relative to each other when contacted with the sensor.

[0017] A third aspect of the invention provides a method of evaluating a property of a surface, comprising the steps of:

[0018] (i) providing a system as defined in the second aspect;

[0019] (ii) contacting the surface with the domed head;

[0020] (iii) causing relative movement between the surface and the sensor head;

[0021] (iv) capturing image data using the camera; (v) inputting the image data from the camera into the data processing unit and

[0022] (vi) using the image processing unit to; to produce a model enabling predictions of consumer tactile response; wherein the surface is a substrate comprising multiple elements that can move relative to each other when contacted with the sensor.

[0023] Where the surface is, for example, an array of imperfectly aligned elements, for example fibres, the elements move across each other and change position under the influence of contact with the sensor. The fibres can move from side to side as well as up and down within the depth of the array. Fibres may also deform by bending or straightening. Fibres touch each other or become separated from each other throughout the measurement. In the evaluation using the present invention, physical contact is made and moved along the surface to create motion on and within the surface. This provides information that is not possible using a single surface and / or conventional non-contact imaging.

[0024] The relative movement between the surface and the probing device can be, for example, stroking, rubbing, sweeping, , combing or pulling, preferably stroking, rubbing, sweeping or combing, most preferably stroking, rubbing or combing. Preferably, the relative movement in is linear in nature. Where the surface is hair, the relative movement is preferably in the direction of the cuticles (from root to tip).

[0025] The relative movement between the surface and the probing device is preferably a sliding motion

[0026] The sliding movement is preferably a single directional force with no simultaneous force from another direction.

[0027] During testing of the surface, the sensor is moved across the surface.

[0028] The method can be used to evaluate the quality of a surface, with or without a cosmetic treatment, for example, stickiness, silkiness, softness, friction, smoothness, roughness, stiffness, abrasion, lubricity, damage, or clean feel. Embodiments of the invention will now be described with reference to the following non-limiting drawings in which:

[0029] Figure 1 is a perspective view of a system (01) and the sensor assembly (1) showing the housing (2) and the domed head (3) contacting a surface that comprises multiple elements (4), which is a switch of hair held at one end by a clip (5). A mechanical moveable arm (6) moves the sensor assembly (1) along the surface (4) in the direction indicated by the arrow and data are fed into the processing unit (7).

[0030] Figure 2 is a section view of the elastomeric sensor assembly (1) in a housing (2), said sensor assembly comprising a stem (12), a domed head (3), source of illumination (8) and a camera (9), wherein the housing (2) comprises multiple parts that are joined together with bayonet fittings (10). One part (2A) houses the camera, one part (2B) the illumination source and one part (2C) retains the elastomeric sensor in position relative to the other components. The illumination sources sit on an illumination band (11).

[0031] Detailed Description of the Invention

[0032] The surface

[0033] The surface comprises multiple elements that can move relative to each other when contacted with a probe such as the sensor described herein.

[0034] Multiple elements are present in many surfaces, for example in an array of hair fibres, or other fibres, textiles and fabrics, woven or non-woven, keratinous fibres, yarns and threads. These all have some degree of freedom of element movement relative to each other.

[0035] The surface may be a fibre array comprising multiple elements of the same or mixed types. The elements may be natural or synthetic, for examples fibres used in textiles and fabrics.

[0036] The surface is preferably selected from hair and hairy skin (for example on the body (such as underarm), scalp or face (such as beard, moustache or eyebrows), most preferably hair. Hair is preferably used as a bundle of hair fibres, such as a switch, where the hair is secured at one end, for example by tying or gluing, or in a clip or band, preferably a metal band. Preferred switches have a weight of from 2 to 20 g; preferably, the length of the switches is from 10 to 40 cm, for example a 5 g, 25 cm switch.

[0037] For skin, in vivo measurements are preferred.

[0038] The housing

[0039] The housing houses the other components and or affixes these to one another. Preferably, the housing comprises multiple parts. Preferably, one part houses the camera, another the illumination source and yet another retains the elastomeric sensor in position relative to the other components.

[0040] A preferred illumination housing comprises an illumination band, such that sources of illumination may be positioned around the edges of the elastomeric sensor which acts as a waveguide for the illumination, directing it onto the surface to be evaluated. Sufficient space is available to enable connections to be made to power the illumination and camera. A coupling between the camera housing and the illumination housing can be provided by a coupling means, for example a bayonet mount design, incorporated into the housing parts. A retainer for the elastomeric sensor can also be coupled to the opposite end of the illumination housing by a coupling means, for example a bayonet mount design incorporated into the housing parts.

[0041] Preferably the parts are formed by 3D printing.

[0042] The elastomeric sensor

[0043] The elastomeric sensor comprises a stem and a domed head. Preferably, the domed head and stem are formed as a single, contiguous piece.

[0044] The stem is preferably cylindrical.

[0045] Preferably, the domed head and stem are cast, using an elastomeric material. Preferably, the elastomeric sensor is cast within 3D printed moulds.

[0046] Preferably the elastomeric sensor is transparent. The elastomeric sensor (the stem and the domed head) is made from an elastomeric material. Preferably, the elastomeric material is transparent. Preferably the elastomeric material is silicone based.

[0047] The elastomeric material is obtainable by a platinum catalysed addition reaction between a two part mixture of a base and an activator, in the presence of an elastomer softening agent.

[0048] In a highly preferred embodiment, the elastomer is obtained using a ratio of base: activator: softener of 1:22:22.

[0049] A suitable two part mixture is XP-565 (from Silicones, Inc. High Point, NC, USA). A suitable elastomer softening agent is Slacker™ (from Smooth-On, Inc. Macungie, PA, USA).

[0050] Advantageously, the domed head has texture on the surface in order to closely replicate the touch of a contact surface on the surface. For example, the texture on the surface is in the pattern of skin texture, hairy skin texture, cloth, garments or hair human fingerprints. Most preferably the texture is in the pattern of human fingerprints.

[0051] The optional coating on the surface of the domed head

[0052] The domed head optionally has a coating. Preferably, the coating is opaque.

[0053] Preferably the opaque coating has the same elastic properties as the domed head itself and is fabricated using the same elastomeric material.

[0054] Preferably the coating comprises the elastomeric material, a solvent, for example Methyl Ethylene Ketone (MEK), and a pigment.

[0055] A preferred coating comprises a ratio of elastomer (obtained using a ratio of base:activator:softener of 1:22:22): Methyl Ethylene Ketone (MEK) solventpigment is 1:1:1.

[0056] Suitable pigments include metallic pigment, aluminium powder, Cast Magic™ from Smooth-On, Silver Bullet™ pigments from Silver Rocket Metallic Pigments Co. Ltd, Jinan City, China, or metallic liquid pigment.

[0057] The coating is carried out by directly dipping the domed head in the coating mixture and allowing to dry. The source of illumination

[0058] Preferred illumination sources are white light LEDs and Red Green Blue (RGB) LEDs.

[0059] In one embodiment, the illumination band of the housing is designed to host 18 x 3 mm LEDs, each soldered to one another. Resistors can be added to balance the illumination intensity.

[0060] The camera

[0061] Any suitable camera may be used. An example is a Logitech C310 webcam. In one example this was configured to capture images with 640 x 480 pixels and used in conjunction with a domed sensor head of 5 mm radius to give a pixel size of 10 to 20 microns, suitable for imaging hair fibres. To ensure the captured tactile images are in focus, the camera lens is manually adjusted while pointing the sensor against a visual pattern placed at the same distance as the membrane surface (without having the LEDs and membrane installed on the sensor).

[0062] The

[0063] The data processing unit

[0064] The system comprises a means to produce and optimise a model enabling predictions of consumer assessed scores for subjective quality attributes to be made for test subjects, comprising a surface or treated surface, wherein the surface has a multiplicity of elements. The treated surface has been treated with, for example a cosmetic product, a styling product, for example a combing creme, a gel, a mousse, a lotion, a fixative spray, a cleaning composition, a conditioner, a leave in product, a softener, a deodorant, an anti-dandruff product, an antiperspirant and a skin product. Preferably cleaning compositions are selected from a hair shampoo, a body wash, a hand wash and a face cleanser. Most preferably, the cleansing composition is a shampoo or a body wash.

[0065] A 3D convolutional neural network may for example be used for this purpose, for example a ResNet 3D network comprising multiple 3D convolution layers may be used.

[0066] The optional moveable arm

[0067] The system may comprise an optional moveable arm configured to move the sensor assembly along a surface. The sensor assembly can be moved along the surface manually, by hand. Alternatively, a mechanical arm can be used. The mechanical moveable arm can be, for example, a robot arm.

[0068] An example of a suitable robot arm is a Dobot Magician, available from Dobot Robotics China, which has three degrees of freedom.

[0069] Optionally a robot arm may, by means of a feedback loop, maintain a constant vertical position or normal contact load relative to the test subject via monitoring for example the contact force, by means of suitable optional force sensors.

[0070] A robotic arm may optionally be wired to a feedback loop to enable a constant vertical position or constant normal load and / or contact force.

[0071] Optional other sensors

[0072] The system may optionally comprise other sensors. For example, the system may optionally comprise a means to quantify the intensity of light reflected from the test subject, for example a photodiode circuit may be used, suitably positioned with respect to the camera, within the housing of the sensor. A suitable design of circuit would be apparent to one skilled in the art based on other aspects of the particular instance of the invention such as the amount of reflected light. Optionally, other types of electromagnetic radiation may be detected for using suitable sensors. Examples include infra-red and or ultra-violet light sensors, suitably positioned with respect to the camera, preferably within the housing of the sensor.

[0073] The Method

[0074] The method is used to evaluate a property of a surface.

[0075] Advantageously, the method is carried out in a dark environment so as to minimize background lighting, preferably to zero.

[0076] The surface is contacted with the domed head of the sensor. The sensor is moved along the surface and video data are simultaneously acquired by the camera, with the aid of illumination passing through the sensor head. Optionally data from optional other sensors are also collected. Sequences of still image data may be extracted from the collected video sequence and optionally annotated prior to feeding into the data processing unit.

[0077] The data from the camera (and any optional other sensors) is fed into the processing unit (preferably a 3D convolutional neural network). The processing unit uses the data about the surfaces to produce models of objective quality. Models are preferably categorical or regression-based models.

[0078] A preferred method comprises the additional steps of treating the surface with a treatment composition to produce a treated surface and then repeating steps (ii) to (vi) and comparing the output of step (vi) of the surface and the treated surface.

[0079] Three different surfaces finishes (metallic liquid pigment, aluminium powder and transparent membrane) of the elastomeric sensor head were tested and used to generate model predictions. RGB illumination was used.

[0080] The model was first trained to recognise differently treated hair (from high conditioning to low conditioning). Then for these surface finish tests, more data was generated with the given surface finish to see how accurate the predictions from the model were for a newly generated dataset.

[0081] Packet quality and tangent force parameters were extracted from the dataset and used to arrive at mean square error (MSE) scores. Similarly, formula class was used to determine an accuracy score.

[0082] It is desirable that accuracy should be high and MSE as low as possible. The results are reported in the following table.

[0083] The “best” outcomes in each case are in bold text in the table.

[0084] It will be seen that the transparent sensor, with no reflective coating, gives the highest accuracy outcomes when training the model on classification tasks. The transparent sensor also produced the lowest MSE results.

Claims

Claims1. A sensor assembly (1) for evaluating a property of a surface (4), wherein the sensor assembly comprises: a) a housing (2) for: b) an elastomeric sensor, which comprises i) a stem (12), and ii) a domed head (3); c) a source of illumination (8); d) a camera (9); wherein the housing comprises a means (2C) to retain the elastomeric sensor in contact with the source of illumination; and wherein the sensor is configured to move relative to the surface; and wherein the sensor does not comprise a reflective coating.

2. A sensor assembly as claimed in claim 1 , wherein the sensor assembly is configured to move relative to the surface in a sliding motion.

3. A sensor assembly as claimed in claim 1 or claim 2, wherein the sensor assembly is transparent.

4. A sensor assembly as claimed in any preceding claim, wherein the elastomeric sensor is made from a transparent, silicone based elastomeric material.

5. A sensor assembly as claimed in any preceding claim, wherein the domed head has texture on the surface in the pattern of human fingerprints, skin texture, hairy skin texture, cloth, garments or hair.

6. A sensor assembly as claimed in any preceding claim, wherein the illumination sources are selected from white light LEDs and Red Green Blue (RGB) LEDs.

7. A sensor assembly as claimed in any preceding claim, the housing comprises an illumination band (11), where the sources of illumination are positioned around the elastomeric sensor.

8. A system (01) comprising the sensor assembly of any one of claims 1 to 7 and additionally comprising: e) a surface; f) a data processing unit (7), g) a means to move the sensor assembly along the surface; wherein the surface comprises multiple elements that can move relative to each other when contacted with the sensor.

9. A system as claimed in claim 8, wherein the surface is selected from hair; hairy skin including on the body, including underarm skin; scalp; or on the face including beard, moustache or eyebrows, preferably hair.

10. A system as claimed in claim 9, wherein the data processing unit employs a 3D convolutional neural network.

11. A system as claimed in claim 9 or claim 10, which further comprises a moveable arm (6) configured to move the sensor assembly along a surface.

12. A system as claimed in any one of claims 9 to 11 , wherein the surface has been treated with a composition selected from a cosmetic product, a styling product, a leave in product, a cleaning composition, a conditioner, a softener, a deodorant, an anti-dandruff product, an antiperspirant and a skin product.

13. A system as claimed in any one of claim 8 to 12, which comprises a means to quantify the intensity of light reflected from the test subject, preferably a photodiode circuit.

14. A method of evaluating a property of a surface, comprising the steps of:(i) providing a system as defined in any one of claims 7 to 13;(ii) contacting the surface with the domed head;(iii) causing relative movement between the surface and the sensor head;(iv) capturing image data using the camera;(v) inputting the image data from the camera into the data processing unit and(vi) using the image processing unit to; to produce a model enabling predictions of consumer tactile response;wherein the surface is a substrate comprising multiple elements that can move relative to each other when contacted with the sensor.

15. A method as claimed in claim 14, which comprises the additional steps of treating the surface with a treatment composition to produce a treated surface and then repeating steps (ii) to (vi) and comparing the output of step (vi) of the surface and the treated surface.

Images