A method, apparatus, device and storage medium for material identification

Abstract

The present application provides a material recognition method, device, equipment and storage medium, which relates to the field of computer tactile technology. When a tactile sensor array contacts a material to be recognized, a plurality of electrical signals output by the tactile sensor array are obtained. The tactile sensor array includes a plurality of single tactile sensors, and each tactile sensor outputs an electrical signal correspondingly. According to the plurality of electrical signals and the corresponding relationship between each electrical signal and the tactile sensor, the electrical signal feature corresponding to the material to be recognized is determined. According to the electrical signal feature and the corresponding relationship between each material category and the preset electrical signal feature, the material to be recognized is determined. In the present application, the electrical signal feature obtained by the multi-array tactile sensor is more accurate than the electrical signal feature obtained by the tactile sensor formed by the single triboelectric nanogenerator technology, and is not affected by contact conditions such as contact pressure, temperature or humidity and the external environment, so that the recognition result is more accurate.

Description

Technical Field

[0001] This application relates to the field of computer haptic technology, and particularly to a material recognition method, device, equipment, and storage medium. Background Art

[0002] With the rapid development of technical fields such as intelligent robots and artificial intelligence, traditional intelligent sensors have been difficult to meet the application requirements. Moreover, the triboelectric tactile sensors in the prior art based on triboelectrification and electrostatic induction effects are easily affected by contact conditions and the external environment when in use. For example, when external conditions such as contact pressure, environmental temperature, and environmental humidity change, their use effects are not good, and they cannot accurately identify the currently touched material.

[0003] Therefore, it is imperative to implement a material recognition method that can be unaffected by external conditions such as contact pressure, environmental temperature, and environmental humidity. Summary of the Invention

[0004] To solve the above problems in the prior art, the embodiments of this application provide a material recognition method, which can achieve that the result and accuracy of material recognition are not affected when the contact pressure, environmental temperature, and environmental humidity change.

[0005] In a first aspect, the embodiments of this application provide a material recognition method, and the method includes:

[0006] Obtain a plurality of electrical signals output by a tactile sensor array when it comes into contact with a material to be recognized; the tactile sensor array includes a plurality of tactile sensors, and each tactile sensor outputs one electrical signal correspondingly;

[0007] Determine the electrical signal characteristics corresponding to the material to be recognized according to the relationship between the plurality of electrical signals and the corresponding relationship between each electrical signal and the tactile sensor;

[0008] Determine the category of the material to be identified according to the electrical signal characteristics and the corresponding relationship between each material category and the preset electrical signal characteristics. An embodiment of the present application provides a material identification method. When a tactile sensor array contacts a material to be identified, obtain a plurality of electrical signals output by the tactile sensor array. The tactile sensor array includes a plurality of tactile sensors, and each tactile sensor outputs an electrical signal correspondingly. According to the plurality of electrical signals and the corresponding relationship between each electrical signal and the tactile sensor, determine the electrical signal characteristics corresponding to the material to be identified. According to the electrical signal characteristics and the corresponding relationship between each material category and the preset electrical signal characteristics, determine the category of the material to be identified. In the present application, the electrical signal characteristics obtained by the multi-array tactile sensor are more accurate than the electrical signal characteristics obtained by the tactile sensor formed by the single triboelectric nanogenerator technology, and are not affected by contact conditions such as contact pressure, temperature, or humidity and the external environment, making the identification result more accurate. In a possible implementation manner, the plurality of tactile sensors are arranged in sequence, and each tactile sensor is made of a different material.

[0009] In a possible implementation manner, the tactile sensor array includes three tactile sensors. The manufacturing materials of the three tactile sensors respectively include: first polyacrylonitrile powder, polymethyl methacrylate powder, and second polyacrylonitrile powder; wherein, the molecular weights of the first polyacrylonitrile powder and the second polyacrylonitrile powder are different.

[0010] In a possible implementation manner, the electrical signal characteristics include one or more of the following characteristics: voltage value characteristic, voltage ratio characteristic, voltage change trend characteristic;

[0011] Among them, the voltage value characteristic is obtained by sorting the voltage peaks of the electrical signals corresponding to each tactile sensor;

[0012] The voltage ratio characteristic is determined according to the ratio between the voltage peaks of the electrical signals corresponding to each tactile sensor;

[0013] The voltage change trend characteristic is determined according to the change trend of the electrical signals corresponding to each tactile sensor.

[0014] In the above method, the electrical signal characteristics include one or more characteristics. Multiple electrical signal characteristics are determined according to the voltage peaks of the plurality of electrical signals, the proportional relationship between the voltage peaks, and the change trend of the electrical signals. For example, the voltage value characteristic obtained by sorting the voltage peaks between the plurality of electrical signals does not change with the change of the external contact pressure. For example, the change trend of the electrical signal does not change with the change of temperature or humidity. The electrical signal characteristics determined in this way can still determine the category of the material to be identified when the contact pressure, ambient temperature, and ambient humidity change.

[0015] In a possible implementation manner, the voltage change trend feature includes a peak-valley voltage feature and a valley-peak voltage feature; the voltage change trend feature is determined in the following manner:

[0016] If the voltage values of the electrical signals corresponding to each tactile sensor first increase and then decrease, the voltage change trend feature is the peak-valley voltage feature;

[0017] If the voltage values of the electrical signals corresponding to each tactile sensor first decrease and then increase, the voltage change trend feature is the valley-peak voltage feature.

[0018] In a possible implementation manner, the correspondence between each material category and the preset electrical signal feature is determined in the following manner:

[0019] For each material category, the following operations are respectively performed: obtaining a plurality of electrical signals output when the tactile sensor array is in contact with the material of the material category; determining the preset electrical signal feature corresponding to the material category according to the plurality of electrical signals and the correspondence between each electrical signal and the tactile sensor;

[0020] Based on the preset electrical signal features corresponding to each material category, the correspondence between each material category and the preset electrical signal feature is obtained.

[0021] In a possible implementation manner, at least one of the electrical signal features is compared with the preset electrical signal features corresponding to each material category to determine the target preset electrical signal feature that is the same as at least one of the electrical signal features;

[0022] The material category corresponding to the target preset electrical signal feature is used as the category of the material to be recognized.

[0023] In a second aspect, the present application provides a material recognition device, and the device includes:

[0024] An acquisition unit, configured to acquire a plurality of electrical signals output by the tactile sensor array when the tactile sensor array is in contact with the material to be recognized; the tactile sensor array includes a plurality of tactile sensors, and each tactile sensor outputs an electrical signal correspondingly;

[0025] A feature extraction unit, configured to determine the electrical signal feature corresponding to the material to be recognized according to the relationship between the plurality of electrical signals and the correspondence between each electrical signal and the tactile sensor;

[0026] A determination unit, configured to determine the category of the material to be recognized according to the electrical signal feature and the correspondence between each material category and the preset electrical signal feature.

[0027] In a possible implementation manner, the voltage change trend feature includes a peak-valley voltage feature and a valley-peak voltage feature;

[0028] The obtaining unit is further configured to determine the voltage change trend feature in the following manner:

[0029] If the voltage value of the electrical signal corresponding to each tactile sensor first increases and then decreases, the voltage change trend feature is the peak-valley voltage feature;

[0030] If the voltage value of the electrical signal corresponding to each tactile sensor first decreases and then increases, the voltage change trend feature is the valley-peak voltage feature.

[0031] In a possible implementation manner, the determining unit is further configured to perform the following operations for each material category: obtain a plurality of electrical signals output when the tactile sensor array contacts the material of the material category; determine a preset electrical signal feature corresponding to the material category according to the plurality of electrical signals and the corresponding relationship between each electrical signal and the tactile sensor;

[0032] Based on the preset electrical signal features corresponding to each material category, obtain the corresponding relationship between each material category and the preset electrical signal features.

[0033] In a possible implementation manner, the determining unit is further configured to compare at least one of the electrical signal features with the preset electrical signal features corresponding to each material category, and determine a target preset electrical signal feature that is the same as at least one of the electrical signal features;

[0034] Use the material category corresponding to the target preset electrical signal feature as the category of the material to be recognized.

[0035] In a third aspect, an embodiment of the present application provides a material identification device, including a memory and a processor, and a computer program that can run on the processor is stored on the memory. When the computer program is executed by the processor, the steps of any one of the material identification methods in the first aspect are implemented.

[0036] In a possible implementation manner, the device further includes a tactile sensor array, the tactile sensor array includes a plurality of tactile sensors, the plurality of tactile sensors are arranged in sequence, and adjacent two tactile sensors are made of different materials.

[0037] Fourthly, an embodiment of the present application provides a tactile sensor array applied to the material identification method described in the first aspect. The tactile sensor array includes a plurality of tactile sensors. When the tactile sensor array contacts the material to be identified, each tactile sensor outputs an electrical signal correspondingly; the plurality of tactile sensors are arranged in sequence, and two adjacent tactile sensors are made of different materials.

[0038] Fifthly, an embodiment of the present application provides a computer-readable storage medium. A computer program is stored in the computer-readable storage medium. When the computer program is executed by a processor, the steps of any one of the material identification methods in the first aspect are implemented.

[0039] The technical effects achieved by the second to fifth aspects provided by the embodiments of the present application are the same as those of the material identification method provided in the first aspect, and will not be elaborated here. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following briefly introduces the drawings required for description in the embodiments. Obviously, the following drawings are only some embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0041] Figure 1 It is a schematic flow chart of a material identification method provided by an embodiment of the present application;

[0042] Figure 2 It is a schematic diagram of a tactile sensor provided by an embodiment of the present application;

[0043] Figure 3 It is an application schematic diagram of a tactile sensor array provided by an embodiment of the present application;

[0044] Figure 4 It is a schematic diagram of the voltage signal obtained by the tactile sensor array provided by an embodiment of the present application;

[0045] Figure 5 It is a schematic diagram of the voltage signal when the contact pressure changes provided by an embodiment of the present application;

[0046] Figure 6 It is a schematic diagram of the voltage peak when the ambient temperature changes provided by an embodiment of the present application;

[0047] Figure 7 It is a schematic diagram of the voltage peak when the ambient humidity changes provided by an embodiment of the present application;

[0048] Figure 8Schematic structural diagram of a material recognition device provided by an embodiment of the present application;

[0049] Figure 9 Schematic structural diagram of a material recognition device provided by an embodiment of the present application. Detailed implementation manners

[0050] In order to make the objectives, technical solutions, and advantages of the present application clearer, the present application will be further described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in the present application without creative efforts shall fall within the protection scope of the present application.

[0051] It should be noted that the terms "including" and "having" and their variations involved in the documents of the present application are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device including a series of steps or units does not necessarily limit to those clearly listed steps or units, but may include other steps or units not clearly listed or inherent to these processes, methods, products, or devices.

[0052] To solve the problem of inaccurate recognition of existing tactile sensors due to changes in external contact pressure, environmental temperature, or environmental humidity, the present application provides a material recognition method based on triboelectric nanogenerator technology. When a tactile sensor array contacts a material to be recognized, multiple electrical signals output by the tactile sensor array are obtained. The tactile sensor array includes multiple tactile sensors, and each tactile sensor outputs an electrical signal correspondingly. According to the multiple electrical signals and the corresponding relationship between each electrical signal and the tactile sensor, the electrical signal characteristics corresponding to the material to be recognized are determined. According to the electrical signal characteristics and the corresponding relationship between each material category and the preset electrical signal characteristics, the category of the material to be recognized is determined. In the present application, the electrical signal characteristics obtained by the multi-array tactile sensors are more accurate than those obtained by the tactile sensors formed by the single triboelectric nanogenerator technology, and are not affected by contact conditions such as contact pressure, temperature, or humidity and the external environment, making the recognition result more accurate.

[0053] Secondly, the triboelectric nanogenerator technology introduced in the material recognition method provided by the present application uses the electrical signals obtained by friction to extract multiple electrical signal characteristics to realize the recognition of the final material category, and can recognize the material category without devices such as a power supply, exceeding the method of recognizing the material category based on skin tactile perception of traditional tactile sensors. Figure 1 Shows a schematic flowchart of a material recognition method provided by an embodiment of the present application, which is applied to a material recognition device, and the material recognition device may be an intelligent robot. AsFigure 1 As shown in the figure, the material recognition method provided by the embodiment of the present application includes the following steps:

[0054] Step S101: Obtain a plurality of electrical signals output by the tactile sensor array when the tactile sensor array is in contact with the material to be recognized.

[0055] Among them, the tactile sensor array includes a plurality of tactile sensors, and each tactile sensor correspondingly outputs an electrical signal.

[0056] The tactile sensor array in the embodiment of the present application includes a plurality of tactile sensors, and different numbers and different combinations of materials can be selected as the main body of the tactile sensor. Exemplarily, the tactile sensor array can be made into a circular shape, and each tactile sensor therein is made into a fan shape, occupying a part of the circle; or each tactile sensor in the tactile sensor array can be placed in parallel, and a plurality of tactile sensors are arranged in sequence, and adjacent two tactile sensors are made of different materials. Hereinafter, a tactile sensor array including three tactile sensors placed in parallel will be used as an example for illustration.

[0057] If three tactile sensors are placed in parallel, the material of each tactile sensor also needs to be selected. In order to achieve flexibility and stretchability and still maintain the original state of the device and realize the sensing function after long-term multiple presses, the tactile sensor array of the embodiment of the present application can use a flexible transparent gel material as the substrate, such as silicone solution.

[0058] If the material of a certain tactile sensor in the tactile sensor array is the same as the material to be recognized, during the contact process between this tactile sensor and the material to be recognized, the potential will mutate due to the contact of the same material, and effective features cannot be extracted, resulting in ineffective recognition. In order to avoid the above situation during the use of the tactile sensor array, the present application has improved the materials selected for each tactile sensor in the tactile sensor array. For example: doping different components or powder materials with different molecular weights, such as polymer powder, in the silicone solution to ensure that the obtained material is the only existing material in daily life and prepare for the subsequent accurate material recognition.

[0059] After selecting the materials and arrangement methods for making the tactile sensors, the surface topography of the tactile sensor array can also be selected. The surface topography also affects the tactile sensing function of the device. The present application selects smooth and convex tactile sensors to form the tactile sensor array. Figure 2 Shows a schematic diagram of a tactile sensor. As Figure 2 shown, 21 is the main body of the tactile sensor with silicone as the substrate; 22 is the detailed enlarged view of the convex structure on the surface of the tactile sensor; 23 is the bifurcated conductive cotton thread.

[0060] Exemplarily, the tactile sensor array is composed of multiple tactile sensors placed in parallel. The placement distance between individual tactile sensors will also have different effects on the final result. The selected placement pitch in this application is 1 mm.

[0061] After all selections are completed, since the materials of each tactile sensor in the tactile sensor array are different, the electrical signals generated after each tactile sensor made of different materials comes into contact with the material to be recognized are also different. Therefore, multiple distinguishable electrical signals will be output.

[0062] More specifically, three tactile sensors are selected. In each tactile sensor, a flexible transparent gel material is used as the substrate and doped with different materials for production.

[0063] The doping materials for the production of the three tactile sensors respectively include: first polyacrylonitrile powder, polymethyl methacrylate powder, and second polyacrylonitrile powder; among them, the molecular weights of the first polyacrylonitrile powder and the second polyacrylonitrile powder are different. The first polyacrylonitrile powder can use polyacrylonitrile powder with a molecular weight of 1 million, and the second polyacrylonitrile powder can use polyacrylonitrile powder with a molecular weight of 150,000. Specific molecular weights can be selected with different values. As Figure 3 shown, Figure 3 FIG. is a schematic diagram of three tactile sensors placed in parallel coming into contact with a key made of metal. Among them, 31 is the key made of metal material; 32 is the array of three tactile sensors placed in parallel; 33 is polyacrylonitrile with a molecular weight of 1 million; 34 is a forked conductive cotton thread, 35 is polymethyl methacrylate; 36 is polyacrylonitrile with a molecular weight of 150,000.

[0064] That is to say, the production material of the first tactile sensor is made of silica gel as the substrate and doped with polyacrylonitrile with a molecular weight of 1 million; the production material of the second tactile sensor is made of silica gel as the substrate and doped with polymethyl methacrylate; the production material of the third tactile sensor is made of silica gel as the substrate and doped with polyacrylonitrile with a molecular weight of 150,000.

[0065] Step S102: Determine the electrical signal characteristics corresponding to the material to be recognized according to the relationship between multiple electrical signals and the corresponding relationship between each electrical signal and the tactile sensor.

[0066] In a possible embodiment, after coming into contact with the material to be recognized, a plurality of electrical signals can be obtained. Each electrical signal has a corresponding relationship with a tactile sensor in the tactile sensor array. A wire needs to be led out inside each tactile sensor, and each tactile sensor outputs the corresponding electrical signal, and the electrical signal can also be transmitted in real time. In the embodiment of the present application, in order to ensure the overall softness and flexibility of the device, the wire selects wire cotton thread as the lead wire, and a forked conductive cotton thread is selected. The electrical output of the forked conductive cotton thread is higher than that of the non-forked conductive cotton thread. Therefore, the forked conductive cotton thread in the embodiment of the present application can output better electrical signals and can better achieve the purpose of identifying the category of the material.

[0067] According to the obtained plurality of electrical signals and the corresponding relationship between each electrical signal and the tactile sensor, the electrical signal characteristics corresponding to the material to be recognized can be determined. The electrical signal characteristics can include one or more, and can include voltage value characteristics, voltage ratio characteristics, and voltage change trend characteristics.

[0068] Different characteristics are determined according to different rules. For example, the voltage value characteristic is obtained by sorting the voltage peaks of the electrical signals corresponding to each tactile sensor. The voltage ratio characteristic is determined according to the ratio between the voltage peaks of the electrical signals corresponding to each tactile sensor. The voltage change trend characteristic is determined according to the change trend of the electrical signals corresponding to each tactile sensor.

[0069] Among them, the voltage change trend characteristic is determined according to the change trend of the electrical signal. Therefore, under different change trends, there are different characteristics. For example, the peak-valley voltage characteristic and the valley-peak voltage characteristic. After obtaining the electrical signals of each tactile sensor, if the voltage values of the electrical signals first increase and then decrease, such a voltage change trend characteristic is the peak-valley voltage characteristic. If the voltage values of the electrical signals first decrease and then increase, the voltage change trend characteristic is the valley-peak voltage characteristic.

[0070] After obtaining the electrical signal characteristics corresponding to the material to be recognized, step S103 can be performed.

[0071] Step S103: Determine the category of the material to be recognized according to the electrical signal characteristics and the corresponding relationship between each material category and the preset electrical signal characteristics.

[0072] In a possible embodiment, the correspondence between each material category and the preset electrical signal characteristics needs to be determined in advance, and the determination method is as follows: for each material category, the following operations are performed separately: obtain multiple electrical signals output when the tactile sensor array is in contact with the material of the material category, and based on the multiple electrical signals and the correspondence between each electrical signal and the tactile sensor, determine the preset electrical signal characteristics corresponding to the material category, and based on the preset electrical signal characteristics corresponding to each material category, obtain the correspondence between each material category and the preset electrical signal characteristics.

[0073] After obtaining the correspondence between each material category and the preset electrical signal characteristics, at least one of the electrical signal characteristics can be compared with the preset electrical signal characteristics corresponding to each material category to determine the target preset electrical signal characteristics that are the same as at least one of the electrical signal characteristics, and the material category corresponding to the target preset electrical signal characteristics is used as the category of the material to be identified.

[0074] Exemplarily, when obtaining the correspondence between each material category and the preset electrical signal characteristics, the material categories used can include: plastics, metals, fabrics, papers, glasses, woods, and leathers, etc. In addition, various soft materials such as skins and furs can also be part of the material library. Contacting the tactile sensor array with the above materials can obtain the electrical signals output by the tactile sensor array corresponding to each material. Here, the electrical signal is taken as an example of a voltage signal, as Figure 4 shown. In the process of selecting materials, plastics, as common materials in daily life, can be selected in three types to enrich the material library. Plastic 1 can be selected as polytetrafluoroethylene, and it is only necessary that plastics 2 and 3 have different materials from plastic 1.

[0075] In the process of extracting the characteristics of the voltage signals output by various materials, the characteristics are extracted according to different rules. For example, the voltage values of glass and wood are different, the voltage change trends between plastic 1 and metal are different, and the arrangement order of the voltage values between plastic 2 and fabric is different. According to the above rules for feature extraction, the features extracted from the electrical signals corresponding to different materials can be used as the unique electrical signal characteristics of this material, and the preset electrical signal characteristics can be obtained. Then, establish a correspondence between each material and the preset electrical signal characteristics, and the correspondence between each material category and the preset electrical signal characteristics can be obtained.

[0076] In the case of the same contact material, the magnitude of the contact pressure, the temperature of the external environment, and the humidity of the external environment will also have more or less influence on the triboelectric tactile sensor. The contact pressure of a single sensor is positively correlated with the output voltage of the device, while high temperature and high humidity will both reduce the output performance of the triboelectric device, thus affecting the final judgment. When extracting the preset electrical signal characteristics, these factors' influence on the preset electrical signal characteristics also needs to be considered.

[0077] Exemplarily, when the temperature, humidity, and contact materials for the test are the same, different contact pressures will generate different electrical signals. For example, Figure 5 as shown, when the temperature is 25°C, the humidity is 20%, and the material is plastic 1 - polytetrafluoroethylene material, the greater the contact pressure, the increasing trend of the generated voltage signal is presented. However, when extracting features from the voltage signal in this application, it can be determined according to the ratio between the voltage peaks of the electrical signal. The extracted feature is named the voltage ratio feature. The method of this kind of extraction can use the normalized feature extraction method. Taking Figure 5 as an example, regardless of the contact pressure, the voltage peak value output by the first tactile sensor is taken as unit 1, and the voltage peak values output by the other two tactile sensors are scaled up or down in the same proportion with the voltage peak value of the first tactile sensor as unit 1 to establish the proportional relationship between the voltage peaks. It can be found that under any contact pressure, for the same material, the same temperature, and the same humidity, the ratio of the voltage peaks obtained remains unchanged.

[0078] Under this kind of material, the contact pressure value can be first determined as 0.1N, the environmental humidity value remains unchanged, and only the environmental temperature value is changed. For example, when the contact pressure is 0.1N, the temperature is 30°C, and the humidity is 20%, when extracting the voltage ratio feature as the voltage ratio feature at this temperature, it can be denoted as temperature 30 - voltage ratio feature. When the contact pressure is 0.1N, the temperature is 35°C, and the humidity is 20%, the voltage ratio feature in this case is extracted and can be denoted as temperature 35 - voltage ratio feature. As Figure 6 shown, a schematic diagram of the voltage peak value of plastic 1 - polytetrafluoroethylene material is given when the contact pressure is 0.1N and the environmental humidity is 20%. The voltage ratio feature can be extracted according to the Figure 6 voltage peak value shown by using the normalized feature extraction method. It can be found that under the same contact pressure and the same humidity, the voltage ratio feature of this kind of material may change, but it can be found that the voltage value feature obtained according to Figure 6 does not change. The maximum voltage peak value is always the voltage peak value output by the third tactile sensor, and the minimum voltage peak value is always the voltage peak value output by the second tactile sensor.

[0079] Still under this kind of material, the contact pressure value can be determined as 0.1N, the environmental temperature value remains unchanged, and only the environmental humidity value is changed. For example, when the contact pressure is 0.1N, the temperature is 25°C, and the humidity is 30%, when extracting the voltage ratio feature as the voltage ratio feature at this humidity, it can be denoted as humidity 30 - voltage ratio feature. When the contact pressure is 0.1N, the temperature is 25°C, and the humidity is 40%, the voltage ratio feature in this case is extracted and can be denoted as humidity 40 - voltage ratio feature. AsFigure 7 As shown, a schematic diagram of the peak voltage of the plastic 1 - polytetrafluoroethylene material is given when the contact pressure is 0.1 N and the ambient temperature is 25°C. According to Figure 7 the peak voltage shown, the voltage ratio feature can be extracted using the normalized feature extraction method. It is found that the voltage ratio feature has also changed, but regardless of the humidity value, the voltage value feature has never changed.

[0080] For all possible temperatures, humidities, and contact pressures, voltage ratio features are established, and voltage value features and voltage change trend features are also extracted simultaneously. It should be noted that Figure 4 only one material among the materials shows the valley - peak voltage feature, but there are many other materials among all possible materials that may have the valley - peak voltage feature. Here, it is intended to illustrate that since there are materials with this valley - peak voltage feature, the preset electrical signal features should include the valley - peak voltage feature and the peak - valley voltage feature in the voltage change trend feature.

[0081] After extracting the electrical signal features, the features can be repeatedly extracted multiple times to verify whether the extracted electrical signal features have conventional properties rather than being obtained by coincidence in a single instance. That is, through multiple training processes, if the electrical signal features obtained from multiple training processes are all consistent, these electrical signal features can be used as the preset electrical signal features.

[0082] During use, when the output electrical signal is obtained, the voltage ratio feature is extracted from the electrical signal and compared with the preset electrical signal features. If there is only one match, the target preset electrical signal feature can be obtained, and then the material category corresponding to the target preset electrical signal feature can be used as the material to be identified to complete the identification task.

[0083] Taking plastic 1 as an example, it is possible that the voltage ratio feature of a certain material is determined to be the same as that of plastic 1, but the peak voltage of each electrical signal is either smaller or larger than that of plastic 1. Therefore, when comparing the voltage value features, the preset electrical signal feature corresponding to the electrical signal feature can also be determined. The process of comparing the obtained electrical signal features with the preset electrical signal features has no order. By directly comparing at least one of the electrical signal features with the preset electrical signal features, a unique and definite correspondence relationship between the electrical signal features and the preset electrical signal features can be obtained.

[0084] Exemplarily, if the material to be identified is paper, the voltage ratio feature obtained may be approximately the same as the preset electrical signal features corresponding to paper and glass. However, since the voltage value features corresponding to paper and glass are completely different, the category of the material to be identified can also be determined as paper through the voltage value feature.

[0085] In a possible implementation, after extracting the preset electrical signal features, a deep learning model can be used to perform deep learning on the preset electrical signal features, and finally a material recognition database is obtained, which includes the corresponding relationship between each material category and the preset material depth features. If this is done, when using it, the deep learning model that performs deep learning on the preset electrical signal features needs to be used synchronously on the device equipped with the tactile sensor array, so that the deep learning model can also perform deep learning on the electrical signal features obtained during the use process, convert them into a material depth feature in the material recognition database, find the target preset material depth feature corresponding to the material depth feature in the material recognition database, and then find the corresponding material category to complete the recognition task.

[0086] The material recognition method in the embodiments of the present application can be used in various application scenarios. For example, in the process of visually impaired people understanding the world, such a method can tell the visually impaired people what they touch, and certain voice descriptions can be added for this material, which can effectively help the visually impaired people recognize and be familiar with the world and train their initial cognition of objects.

[0087] The material recognition method provided in the embodiments of the present application is not affected by changes in external contact pressure, environmental temperature, and environmental humidity. Even when the temperature and humidity are constantly changing when contacting the material to be recognized, as long as the detected output electrical signal is obtained, the electrical signal features can be obtained based on the electrical signal, and at least one of the electrical signal features is compared with the preset electrical signal features corresponding to each material category to determine the target preset electrical signal feature that is the same as at least one of the electrical signal features, and then the material category corresponding to the target preset electrical signal feature is obtained as the category of the material to be recognized, and the recognition task is completed.

[0088] The embodiments of the present application also provide a material recognition device. Figure 8 It is a schematic structural diagram of a material recognition device provided in the embodiments of the present application; as Figure 8 shown, the material recognition device includes:

[0089] An acquisition unit 801, configured to acquire a plurality of electrical signals output by the tactile sensor array when the tactile sensor array contacts the material to be recognized; the tactile sensor array includes a plurality of tactile sensors, and each tactile sensor outputs an electrical signal correspondingly;

[0090] A feature extraction unit 802, configured to determine the electrical signal features corresponding to the material to be recognized according to the relationship between the plurality of electrical signals and the corresponding relationship between each electrical signal and the tactile sensor;

[0091] A determination unit 803, configured to determine the category of the material to be recognized according to the electrical signal features and the corresponding relationship between each material category and the preset electrical signal features.

[0092] In a possible implementation, the voltage change trend feature includes a peak-valley voltage feature and a valley-peak voltage feature;

[0093] The obtaining unit 801 is further configured to determine the voltage change trend feature in the following manner:

[0094] If the voltage value of the electrical signal corresponding to each tactile sensor first increases and then decreases, the voltage change trend feature is the peak-valley voltage feature;

[0095] If the voltage value of the electrical signal corresponding to each tactile sensor first decreases and then increases, the voltage change trend feature is the valley-peak voltage feature.

[0096] In a possible implementation, the determining unit 803 is further configured to perform the following operations for each material category: obtain a plurality of electrical signals output when the tactile sensor array contacts the material of the material category; determine the preset electrical signal feature corresponding to the material category according to the plurality of electrical signals and the corresponding relationship between each electrical signal and the tactile sensor;

[0097] Based on the preset electrical signal features corresponding to each material category, obtain the corresponding relationship between each material category and the preset electrical signal features.

[0098] In a possible implementation, the determining unit 803 is further configured to compare at least one of the electrical signal features with the preset electrical signal features corresponding to each material category, and determine the target preset electrical signal feature that is the same as at least one of the electrical signal features;

[0099] Use the material category corresponding to the target preset electrical signal feature as the category of the material to be recognized.

[0100] The embodiment of the present application further provides a material recognition device, which at least includes a memory for storing data and a processor. Among them, for the processor for data processing, when performing processing, it can be implemented by a microprocessor, a CPU, a GPU (Graphics Processing Unit), a DSP or an FPGA. For the memory, operation instructions are stored in the memory, and the operation instructions can be computer-executable code, and each step in the process of the material recognition method of the embodiment of the present application is implemented through the operation instructions. A program of a deep learning model can also be stored in the memory, where the deep learning model can be used to perform deep learning on the preset electrical signal features and the electrical signal features to obtain a material recognition database.

[0101] Figure 9 It is a schematic structural diagram of a material recognition device provided by an embodiment of the present application. AsFigure 9 As shown in Figure 9 , the material identification device 900 includes a memory 901, a processor 902, a data acquisition module 903, and a bus 904. The memory 901, the processor 902, and the data acquisition module 903 are all connected via the bus 904, and the bus 904 is used for data transmission between the memory 901, the processor 902, and the data acquisition module 903.

[0102] Among them, the memory 901 can be used to store software programs and modules. The processor 902 executes various functional applications and data processing of the material identification device 900 by running the software programs and modules stored in the memory 901, such as the material identification method provided in the embodiments of the present application. The memory 901 mainly includes a program storage area and a data storage area. Among them, the program storage area can store an operating system, application programs of at least one application, etc.; the data storage area can store data created according to the use of the material identification device 900. In addition, the memory 901 can include a high-speed random access memory, and can also include a non-volatile memory, such as at least one magnetic disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0103] The processor 902 is the control center of the material identification device 900, connects all parts of the entire material identification device 900 using the bus 904 and various interfaces and lines, and executes various functions of the material identification device 900 and processes data by running or executing the software programs and / or modules stored in the memory 901, and calling the data stored in the memory 901. Optionally, the processor 902 can include one or more processing units, such as a CPU, a GPU (Graphics Processing Unit, graphics processing unit), a digital processing unit, etc.

[0104] The data acquisition module 903 in the material identification device 900 can include a tactile sensor array. When the tactile sensor array comes into contact with the material to be identified, an electrical signal output by the tactile sensor array can be obtained. The data acquisition module 903 can also include a detection module for detecting the ambient temperature, ambient humidity, and the contact pressure when the tactile sensor array comes into contact with the material to be identified, so as to make the result of identifying the material to be identified more accurate.

[0105] The material identification device 900 can also include a voice output module. The voice output module can be used to perform a voice broadcast on the category of the material identified by the tactile sensor array. Descriptions for each material category can be stored in the memory. When the voice output module broadcasts the material category, some descriptions of this material can be made, so that visually impaired people can better understand the material they touch.

[0106] The embodiment of the present application also provides a tactile sensor array applied to a material recognition method. The tactile sensor array includes a plurality of tactile sensors. When the tactile sensor array contacts the material to be recognized, each tactile sensor correspondingly outputs an electrical signal. The plurality of tactile sensors are arranged in sequence, and two adjacent tactile sensors are made of different materials.

[0107] In a possible embodiment, the tactile sensor array can obtain electrical signals corresponding to a plurality of tactile sensors by using tactile sensors made of different arrangements, different quantities, and different materials. The electrical signal features obtained by using different electrical signals can more accurately determine the category of the material to be recognized. Exemplarily, Figure 3 A schematic diagram of a tactile sensor array with three tactile sensors placed in parallel is shown.

[0108] The embodiment of the present application also provides a computer-readable storage medium. Computer-executable instructions are stored in the computer storage medium. When the computer program is executed by a processor, it can be used to implement the material recognition method described in any embodiment of the present application.

[0109] In some possible implementation manners, each aspect of the material recognition method provided by the present application can also be implemented in the form of a program product, which includes program code. When the program product runs on a computer device, the program code is used to cause the computer device to execute the steps of the material recognition method according to various exemplary embodiments of the present application described above in this specification. For example, the computer device can execute the process of the material recognition method of steps S101 to S103 as Figure 1 shown.

[0110] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk memories, CD-ROMs, optical memories, etc.) containing computer-usable program codes.

[0111] The present application is described with reference to the flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to the present application. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and the combination of flows and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate for implementation in the processFigure 1 a process or processes and / or blocks Figure 1 means for the functions specified in a block or blocks.

[0112] These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means that implement the functions in the process Figure 1 a process or processes and / or blocks Figure 1 specified in a block or blocks.

[0113] These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process, so that the instructions executed on the computer or other programmable apparatus provide steps for implementing the functions in the process Figure 1 a process or processes and / or blocks Figure 1 specified in a block or blocks.

[0114] Obviously, those skilled in the art can make various changes and modifications to this application without departing from the spirit and scope of this application. Thus, if these modifications and variations of this application fall within the scope of the claims of this application and their equivalent technologies, this application is also intended to include these changes and modifications.

Claims

1. A material identification method, characterized in that, The method includes: Obtaining a plurality of electrical signals output by the tactile sensor array when it contacts the material to be recognized; the tactile sensor array includes a plurality of tactile sensors, and each tactile sensor outputs an electrical signal correspondingly; Determining the electrical signal features corresponding to the material to be recognized according to the relationship between the plurality of electrical signals and the corresponding relationship between each electrical signal and the tactile sensor; Performing deep learning processing on the electrical signal features using a pre-trained deep learning model to convert them into material depth features, matching the material depth features corresponding to the target preset material depth features in the material recognition database, and determining the category of the material to be recognized corresponding to the target preset material depth features according to the corresponding relationship between each material category and the preset material depth features in the material recognition database; The electrical signal features include one or more of the following features: voltage value feature, voltage ratio feature, voltage change trend feature; Among them, the voltage value feature is obtained by sorting the voltage peaks of the electrical signals corresponding to each tactile sensor; The voltage ratio feature is determined according to the ratio between the voltage peaks of the electrical signals corresponding to each tactile sensor; The voltage change trend feature includes peak-valley voltage feature and valley-peak voltage feature; the voltage change trend feature is determined in the following way: if the voltage values of the electrical signals corresponding to each tactile sensor first increase and then decrease, the voltage change trend feature is the peak-valley voltage feature; if the voltage values of the electrical signals corresponding to each tactile sensor first decrease and then increase, the voltage change trend feature is the valley-peak voltage feature; The corresponding relationship between each material category and the preset material depth features is determined in the following way: For each material category, the following operations are respectively performed: obtaining a plurality of electrical signals output by the tactile sensor array when it contacts the material of the material category; determining the preset electrical signal features corresponding to the material category according to the plurality of electrical signals and the corresponding relationship between each electrical signal and the tactile sensor; Using a deep learning model to perform deep learning on the preset electrical signal features corresponding to each material category to obtain the material recognition database, and the material recognition database includes the corresponding relationship between each material category and the preset material depth features.

2. The method according to claim 1, wherein The plurality of tactile sensors are arranged in sequence, and two adjacent tactile sensors are made of different materials.

3. The method according to claim 2, characterized in that, The tactile sensor array includes three tactile sensors, and the manufacturing materials of the three tactile sensors respectively include: first polyacrylonitrile powder, polymethyl methacrylate powder, and second polyacrylonitrile powder; among them, the molecular weights of the first polyacrylonitrile powder and the second polyacrylonitrile powder are different.

4. A material identification device, characterized in that, The device includes: An acquisition unit, configured to obtain a plurality of electrical signals output by the tactile sensor array when it contacts the material to be recognized; the tactile sensor array includes a plurality of tactile sensors, and each tactile sensor outputs an electrical signal correspondingly; A feature extraction unit, configured to determine the electrical signal features corresponding to the material to be identified according to the relationships between the multiple electrical signals and the corresponding relationships between each electrical signal and the tactile sensors; A determination unit, configured to perform deep learning processing on the electrical signal features using a pre-learned deep learning model to convert them into material depth features, match the material depth features with the target preset material depth features in a material recognition database, and determine the category of the material to be identified corresponding to the target preset material depth features according to the corresponding relationships between the material categories and the preset material depth features in the material recognition database; The electrical signal features include one or more of the following features: voltage value feature, voltage ratio feature, voltage change trend feature; Among them, the voltage value feature is obtained by sorting the voltage peaks of the electrical signals corresponding to each tactile sensor; The voltage ratio feature is determined according to the ratios between the voltage peaks of the electrical signals corresponding to each tactile sensor; The voltage change trend feature includes a peak-valley voltage feature and a valley-peak voltage feature; The acquisition unit is further configured to determine the voltage change trend feature in the following manner: if the voltage values of the electrical signals corresponding to each tactile sensor first increase and then decrease, the voltage change trend feature is the peak-valley voltage feature; if the voltage values of the electrical signals corresponding to each tactile sensor first decrease and then increase, the voltage change trend feature is the valley-peak voltage feature; The determination unit is further configured to perform the following operations for each material category respectively: acquire the multiple electrical signals output when the tactile sensor array contacts the material of the material category; determine the preset electrical signal features corresponding to the material category according to the multiple electrical signals and the corresponding relationships between each electrical signal and the tactile sensors; perform deep learning on the preset electrical signal features corresponding to each material category using a deep learning model to obtain the material recognition database, where the material recognition database includes the corresponding relationships between the material categories and the preset material depth features.

5. A material identification device, characterized in that, It includes a memory and a processor, and a computer program is stored on the memory and can run on the processor. When the computer program is executed by the processor, the method according to any one of claims 1 to 3 is implemented.

6. The device according to claim 5, characterized in that, The device further includes a tactile sensor array, where the tactile sensor array includes multiple tactile sensors, the multiple tactile sensors are arranged in sequence, and adjacent two tactile sensors are made of different materials.

7. A tactile sensor array applied to the material identification method according to claim 1, characterized in that, The tactile sensor array includes multiple tactile sensors. When the tactile sensor array contacts the material to be identified, each tactile sensor outputs an electrical signal correspondingly; The multiple tactile sensors are arranged in sequence, and adjacent two tactile sensors are made of different materials.

8. A computer-readable storage medium storing a computer program therein, characterized in that: When the computer program is executed by the processor, the method according to any one of claims 1 to 3 is implemented.

Images