Wearable device and material identification method

Abstract

The embodiment of the invention discloses wearable equipment and a material identification method. The wearable device comprises a wearable device body and at least one friction electric signal obtaining module installed on the wearable device body. The signal transmission module is connected with the friction electric signal acquisition module and is used for transmitting the friction electric signal acquired by the friction electric signal acquisition module; the processor is connected with the signal transmission module and is used for acquiring the friction electric signal; according to the friction electric signal, a material identification result is determined, and according to the material identification result, feature extraction is carried out on the friction electric signal to obtain target feature data; and performing material identification on the target feature data to obtain a material identification result based on a corresponding relationship between material types and features learned in advance by a material identification model. According to the embodiment of the invention, the method can achieve the quick and accurate recognition and classification of different substances.

Description

Technical Field

[0001] The present application belongs to the field of computer technology, and in particular, relates to a wearable device and a material identification method. Background Art

[0002] Material identification refers to the process of identifying and confirming the type, properties and purpose of material.

[0003] In related technologies, the material of a substance can be determined through computer vision or chemical composition analysis. However, when using computer vision technology to identify materials, the external environment needs to be strictly controlled, otherwise the identification results will be inaccurate. Chemical composition analysis technology to identify materials often requires the use of expensive and complex large instruments.

[0004] Therefore, how to conveniently and accurately identify the material of an object is a technical problem that needs to be solved urgently by relevant technical personnel. Summary of the invention

[0005] The embodiments of the present application provide a wearable device and a material identification method, which can realize fast, accurate and convenient identification and classification of different substances.

[0006] In a first aspect, an embodiment of the present application provides a wearable device, comprising: a wearable device body and at least one friction electric signal acquisition module installed on the body; a signal transmission module connected to the friction electric signal acquisition module, for transmitting the friction electric signal acquired by the friction electric signal acquisition module; a processor connected to the signal transmission module, for acquiring the friction electric signal; determining a material identification result based on the friction electric signal, wherein the material identification result is obtained by extracting features from the friction electric signal to obtain target feature data; and obtaining a material identification result by performing material identification on the target feature data based on the correspondence between material types and features pre-learned by a material identification model.

[0007] In one embodiment, the wearable device body includes a glove, and the friction electric signal acquisition module is arranged at a material contact position on one or more fingers of the glove.

[0008] In one embodiment, the triboelectric signal acquisition module includes: a triboelectric sensor having a flexible sensing membrane.

[0009] In one embodiment, the signal transmission module includes: a conditioning circuit connected to the friction electric signal acquisition module; an analog-to-digital signal conversion circuit connected to the conditioning circuit; and a control unit connected to the analog-to-digital signal conversion circuit.

[0010] In one embodiment, the wearable device further includes: a display device connected to the processor, and configured to display a material identification result determined by the processor.

[0011] In one embodiment, the processor is also used to extract features from the triboelectric signal to obtain target feature data; based on the correspondence between material types and features pre-learned by the material identification model, the target feature data is used to identify materials to obtain material identification results.

[0012] In a second aspect, an embodiment of the present application provides a material identification method, the method comprising: obtaining a target triboelectric signal; performing feature extraction on the target triboelectric signal to obtain target feature data;

[0013] Based on the correspondence between material types and features pre-learned by the material recognition model, material recognition is performed on the target feature data to obtain a material recognition result.

[0014] In one embodiment, feature extraction is performed on a target triboelectric signal to obtain target feature data, including: feature extraction is performed on a target triboelectric signal to obtain original feature data, root mean square feature data, and charge feature data; and the original feature data, root mean square feature data, and charge feature data are spliced ​​to obtain target feature data.

[0015] In one embodiment, feature extraction is performed on a target triboelectric signal to obtain original feature data, root mean square feature data, and charge quantity feature data, including: sampling and processing the target triboelectric signal at a preset frequency to obtain original feature data; performing root mean square calculation on the original feature data to obtain root mean square feature data; and integrating the original feature data to obtain charge quantity feature data.

[0016] In one embodiment, the material identification model includes an input layer, a hidden layer and a classification layer; based on the correspondence between material types and features pre-learned by the material identification model, material identification is performed on target feature data to obtain a material identification result, including: using the input layer to perform vector mapping on the target feature data to obtain a feature vector; inputting the feature vector into the hidden layer, and using the hidden layer to perform high-dimensional vector extraction on the feature vector to obtain a target vector; using the classification layer to determine the material identification result corresponding to the target vector.

[0017] In one implementation, the classification layer is used to determine the material identification result corresponding to the target vector, including: respectively determining the probability values ​​of the target vector belonging to different material categories; and selecting the material category corresponding to the maximum probability value as the material identification result.

[0018] In one embodiment, before obtaining the friction electric signal, the method also includes: obtaining the original friction electric signal; when it is determined that the original friction electric signal meets the trigger condition, using the original friction electric signal as the target friction electric signal; the trigger condition includes that the original friction electric signal is greater than or equal to a preset signal threshold; or the change rate of the original friction electric signal is greater than or equal to a preset change rate threshold.

[0019] In a third aspect, an embodiment of the present application provides a material identification device, the device comprising:

[0020] An acquisition module, used for acquiring a target triboelectric signal;

[0021] An extraction module is used to extract features of the target triboelectric signal to obtain target feature data;

[0022] The recognition module is used to perform material recognition on the target feature data based on the correspondence between material types and features pre-learned by the material recognition model to obtain a material recognition result.

[0023] In a fourth aspect, an embodiment of the present application provides a material identification system, the system comprising: a wearable device and a data processing device, the data processing device being used to execute the material identification method as in the second aspect or any one of the embodiments of the second aspect.

[0024] In a fifth aspect, an embodiment of the present application provides a material identification device, the device comprising: a processor and a memory storing computer program instructions; when the processor executes the computer program instructions, the material identification method in the second aspect or any one of the embodiments of the second aspect is implemented.

[0025] In a sixth aspect, a computer-readable storage medium stores computer program instructions, and when the computer program instructions are executed by a processor, the material identification method in the second aspect or any one of the embodiments of the second aspect is implemented.

[0026] In the seventh aspect, an embodiment of the present application provides a computer program product. When the instructions in the computer program product are executed by a processor of an electronic device, the electronic device executes the material identification method as described in the second aspect or any one of the embodiments of the second aspect.

[0027] The wearable device and material identification method of the embodiment of the present application obtains the triboelectric signal through one or more triboelectric signal acquisition modules set on the wearable device body, and transmits the triboelectric signal acquired by the triboelectric signal acquisition module to the processor through the signal transmission module, so that the processor can determine the identification result of the material to be identified based on the triboelectric signal. It can be understood that the wearable device can be used to obtain the triboelectric signal for material identification in a convenient and flexible way. Further, the target feature data is obtained by extracting the features of the triboelectric signal; and the material identification result obtained by material identification of the target feature data based on the material identification model, so that the material identification result obtained has a high accuracy, and can realize convenient and accurate identification of the material. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] In order to more clearly illustrate the technical solution of the embodiments of the present application, the following is a brief introduction to the drawings required for use in the embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0029] Figure 1 A structural diagram of a wearable device provided by an embodiment of the present application is shown;

[0030] Figure 2 The figure shows a structural diagram of a flexible sensing film provided by one embodiment of the present application;

[0031] Figure 3 An architecture diagram of a wearable device provided by an embodiment of the present application is shown;

[0032] Figure 4 A structural diagram of a wearable device provided in another embodiment of the present application is shown;

[0033] Figure 5 A structural diagram of a wearable device provided in yet another embodiment of the present application is shown;

[0034] Figure 6 A schematic diagram showing a deployment method of a display device provided by an embodiment of the present application;

[0035] Figure 7 A schematic diagram of a display device provided by an embodiment of the present application is shown;

[0036] Figure 8 A schematic diagram of a material identification method according to an embodiment of the present application is shown;

[0037] Fig. 9 A schematic diagram of the process of a material identification method provided by another embodiment of the present application is shown;

[0038] Fig.10 A schematic diagram of the process of a material identification method provided by another embodiment of the present application is shown;

[0039] Fig.11 A schematic diagram of the process of a material identification method provided in another embodiment of the present application is shown;

[0040] Fig.12 A schematic diagram of a triboelectric signal provided by an embodiment of the present application is shown;

[0041] Fig.13 A schematic diagram showing the accuracy of a material recognition model provided by an embodiment of the present application;

[0042] Fig.14 is a structural schematic diagram of a material identification method and device provided by another embodiment of the present application;

[0043] Fig.15 An architectural diagram of a material identification system provided by an embodiment of the present application is shown;

[0044] Fig.16 It is a structural schematic diagram of a material identification device provided in yet another embodiment of the present application. DETAILED DESCRIPTION

[0045] The features and exemplary embodiments of various aspects of the present application will be described in detail below. In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are only intended to explain the present application, rather than to limit the present application. For those skilled in the art, the present application can be implemented without the need for some of these specific details. The following description of the embodiments is only to provide a better understanding of the present application by illustrating the examples of the present application.

[0046] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the statement "include..." do not exclude the presence of other identical elements in the process, method, article or device including the elements.

[0047] With the development of science and technology, production processes have become more sophisticated, and the types of materials have become more diverse. It is difficult for users to determine the material by feel and appearance. For example, in daily life, consumers cannot accurately distinguish between silk fabrics and acetate fabrics that simulate silk, polyethylene plastics and polyester plastics, and other materials with similar appearance and feel by eyes and touch.

[0048] Therefore, in order to accurately identify materials, in related technologies, the material of a substance can be determined through computer vision or chemical composition analysis. However, when using computer vision technology to identify materials, it is necessary to strictly control the accuracy of the equipment and the external environment, such as light, otherwise the identification results will be inaccurate. Chemical composition analysis technology often requires the use of expensive and complex large instruments to identify materials, and it is difficult to conveniently apply them to different inspection sites.

[0049] In order to solve the problems in the prior art, the embodiments of the present application provide a wearable device and a material identification method. The wearable device provided by the embodiments of the present application is first introduced below.

[0050] Figure 1 FIG. 1 shows a structural diagram of a wearable device provided by an embodiment of the present application. Figure 1 As shown, the wearable device 100 includes:

[0051] A wearable device body 101 and at least one triboelectric signal acquisition module 102 mounted on the body;

[0052] A signal transmission module 103 connected to the triboelectric signal acquisition module 102, and used to transmit the triboelectric signal acquired by the triboelectric signal acquisition module 102;

[0053] The processor 104 is connected to the signal transmission module 103 and is used to obtain the triboelectric signal; determine the material identification result based on the triboelectric signal, wherein the material identification result is obtained by extracting the features of the target triboelectric signal to obtain the target feature data; and obtain the material identification result by performing material identification on the target feature data based on the correspondence between the material type and the feature pre-learned by the material identification model.

[0054] Exemplarily, the wearable device body 101 may be a portable device that can be worn on a user.

[0055] Exemplarily, one or more triboelectric signal acquisition modules 102 may be provided in the wearable device 100. Furthermore, the triboelectric signal acquisition module 102 can generate a triboelectric signal when it contacts the substance to be identified. It is understandable that different types of materials have different intrinsic electron affinities, and therefore, the triboelectric signal acquisition module 102 can generate different triboelectric signals when it contacts different substances to be identified.

[0056] Exemplarily, the triboelectric signal may be a current signal generated by the triboelectric signal acquisition module 102 when it contacts the substance to be identified. Alternatively, the triboelectric signal may be a voltage signal generated by the triboelectric signal acquisition module 102 when it contacts the substance to be identified.

[0057] In some optional embodiments, the triboelectric signal acquisition module 102 may be a triboelectric sensor having a flexible sensing membrane, wherein the flexible sensing membrane may be made of electrostatically spun nanofibers.

[0058] In one example, the flexible sensing film may include a sensing layer and a conductive layer that are overlapped. Figure 2 FIG. 1 shows a schematic diagram of a flexible sensing film provided by an embodiment of the present application. Figure 2 As shown, the flexible sensing film 200 may include a sensing layer 201 and a conductive layer 202 stacked in sequence. The sensing layer may be composed of nanofibers for contacting the substance to be identified and generating an electrical response signal, i.e., a triboelectric signal. The conductive layer may be composed of a nanofiber support and a conductive layer for transmitting the generated triboelectric signal. The conductive layer may be composed of conductive materials such as metal-based and carbon-based. The preparation materials of the sensing layer may also include fluorine-containing polymer nanofiber membranes, polyvinyl alcohol nanofiber membranes, and regenerated silk fibroin nanofiber membranes.

[0059] It can be understood that the structure of the flexible sensor film and the materials of each layer involved in the embodiments of the present application are only used for exemplary description, and the present application does not limit the structure of the flexible sensor film and the materials of each layer.

[0060] In the embodiment of the present application, the friction electric signal acquisition module 102 prepared by the electrospinning process has the characteristics of softness and breathability, which can ensure the comfort of the user when wearing the wearable device 100 and improve the user's experience.

[0061] Further, in some optional embodiments, the wearable device body 101 may include a glove, and the friction electric signal acquisition module 102 may be disposed at a material contact position on one or more fingers of the glove.

[0062] In one example, the gloves may be gloves with better air permeability and higher comfort.

[0063] In another example, Figure 3 FIG. 1 shows an architecture diagram of a wearable device provided by an embodiment of the present application, such as Figure 3 As shown, the triboelectric signal acquisition module 102 may be provided at the fingertips of one or more fingers of the glove 300. Figure 3As shown, a triboelectric signal acquisition module 102 may be provided on each of the index finger 301 , the middle finger 302 , and the ring finger 303 of the glove.

[0064] In an embodiment of the present application, by setting the friction electric signal acquisition module 102 at the material contact position on one or more fingers of the glove, it is ensured that the user can flexibly and conveniently obtain the friction electric signal when using the wearable device 100, thereby ensuring that the material of the substance to be identified can be accurately identified.

[0065] Exemplarily, the signal transmission module 103 is connected to the triboelectric signal acquisition module 102 and is used to transmit the triboelectric signal acquired by the triboelectric signal acquisition module 102 .

[0066] In one example, the signal transmission module 103 may include a fiber-shaped conductor, through which the friction electric signal acquisition module 102 and the processor 104 are connected, so that the processor 104 can receive the friction electric signal collected by the friction electric signal acquisition module 102 .

[0067] Exemplarily, the processor 104 is connected to the signal transmission module 103 to obtain the triboelectric signal and determine the material identification result according to the triboelectric signal.

[0068] Among them, the material identification result is obtained by extracting the features of the friction electric signal to obtain the target feature data; based on the correspondence between the material type and the features pre-learned by the material identification model, the material identification result is obtained by performing material identification on the target feature data.

[0069] Exemplarily, the connection between the processor 104 and the signal transmission module 103 may include an electrical connection or a communication connection. In one example, the electrical connection may be to connect the processor 104 and the signal transmission module 103 through a fiber-shaped conductor; in another example, the communication connection may be to connect the processor 104 and the signal transmission module 103 through a transmission control protocol / Internet protocol (TCP / IP) using a communication device. It is understood that the above-mentioned connection between the processor 104 and the signal transmission module 103 is only for exemplary description, wherein the electrical connection method may also be used to realize the transmission of triboelectric signals through other conductive materials; the communication connection method may also be used to transmit triboelectric signals through the User Datagram Protocol (UDP) and the Bluetooth Protocol Stack, and this application does not limit this.

[0070] In some optional embodiments, the processor 104 in the wearable device 100 may extract features from the received triboelectric signal to obtain target feature data. Further, based on the correspondence between material types and features pre-learned by the material recognition model, the target feature data may be subjected to material recognition to obtain a material recognition result.

[0071] Exemplarily, the processor 104 can also transmit the triboelectric signal to the mobile smart device or platform by interacting with a data processing device (such as a mobile smart device or platform). Furthermore, the mobile smart device or platform extracts features of the received triboelectric signal to obtain target feature data. Then, based on the correspondence between the material type and the feature pre-learned by the material recognition model, the target feature data is subjected to material recognition to obtain a material recognition result.

[0072] In the embodiment of the present application, the material identification structure determined by the material identification model can more accurately determine the material of the substance to be identified.

[0073] In the embodiment of the present application, the triboelectric signal is acquired by one or more triboelectric signal acquisition modules arranged on the wearable device body, and the triboelectric signal acquired by the triboelectric signal acquisition module is transmitted to the processor through the signal transmission module, so that the processor can determine the identification result of the material to be identified according to the triboelectric signal. It can be understood that the wearable device can be used to obtain the triboelectric signal for material identification in a convenient and flexible manner, and further, the target feature data is obtained by extracting the features of the triboelectric signal; and the material identification result obtained by material identification of the target feature data based on the material identification model, so that the material identification result obtained has a high accuracy, and realizes convenient and accurate identification of the material.

[0074] In order to ensure that the friction electric signal acquisition module 102 can be correctly identified by the processor 104, as another implementation method of the present application, the present application also provides another implementation method of the wearable device 100, please refer to the following embodiments for details.

[0075] Figure 4 FIG. 2 shows a structural diagram of a wearable device provided by another embodiment of the present application. Figure 4 As shown, the signal transmission module 103 in the wearable device 100 may include: a conditioning circuit 105 connected to the friction electric signal acquisition module 102; an analog-to-digital signal conversion circuit 106 connected to the conditioning circuit 105; and a control unit 107 connected to the analog-to-digital signal conversion circuit 106.

[0076] Exemplarily, the conditioning circuit 105 is connected to the triboelectric signal acquisition module 102 and is used to filter and amplify the triboelectric signal acquired by the triboelectric signal acquisition module 102 to ensure the stability and anti-interference capability of the triboelectric signal.

[0077] In one example, the conditioning circuit 105 may include a filter circuit and an amplifier circuit. The filter circuit may be composed of a capacitor and a resistor. The amplifier circuit may include a multi-stage amplifier to amplify the triboelectric signal.

[0078] Exemplarily, the analog-to-digital signal conversion circuit 106 is connected to the conditioning circuit 105 and the control unit 107 to convert the analog signal after filtering and amplification into a digital signal to ensure that the processor 104 can recognize it.

[0079] Exemplarily, the control unit 107 is connected to the analog-to-digital signal conversion circuit 106 for converting the digital signal into a data frame format that is convenient for transmission, thereby ensuring that the triboelectric signal can be transmitted accurately and efficiently.

[0080] In one example, technicians can select different data frame formats according to different requirements. For example, when the processor 104 needs to send the triboelectric signal to a platform or a mobile smart device, the data frame format can be set to a data frame format corresponding to TCP / IP.

[0081] It is understandable that when the processor and the signal transmission module are connected through different communication protocols, the triboelectric signal can be converted according to the data frame format supported by the communication protocol. For example, when the processor and the signal transmission module are connected through the Bluetooth protocol, the triboelectric signal can be converted through the data frame format of the Bluetooth basic rate mode to achieve the transmission of the triboelectric signal.

[0082] In the embodiment of the present application, the signal is processed accordingly through a conditioning circuit, an analog-to-digital signal conversion circuit, and a control unit, thereby ensuring that the processor can obtain a triboelectric signal of good quality, laying a good foundation for using triboelectric signals for material identification.

[0083] Furthermore, in order to ensure that the user can obtain the material identification result, as another implementation method of the present application, the present application also provides another implementation method of the wearable device, please refer to the following embodiments for details.

[0084] Figure 5 FIG. 2 shows a structural diagram of a wearable device provided by another embodiment of the present application. Figure 5 As shown, the wearable device 100 may further include:

[0085] The display device 108 is connected to the processor 104 and is used to display the material identification result determined by the processor 104.

[0086] Exemplarily, the display device 108 may be electrically connected to the processor 104, and display the material identification result by receiving a display signal sent by the processor 104. Alternatively, the display device 108 may be connected to the processor 104 by communication, and then receive a display signal sent by the processor 104 to display the material identification result.

[0087] In one example, the display device 108 can be disposed on the wearable device body 101 by means of electrical connection, for example, it can be disposed on the back of the hand in the glove. Figure 6 As shown, Figure 6 A schematic diagram of a deployment method of a display device provided by an embodiment of the present application is shown. The display device 108 may be composed of a light emitting diode (LED) array 601. It is understood that the processor 104 may control the LED array through a corresponding control circuit, so as to display the material identification result in the display device 108. For example, the LED array may be controlled to display numbers, and the user may determine the material of the substance to be identified through a preset corresponding relationship according to the numbers displayed by the LED array.

[0088] In another example, the display device 108 may also be a display in a mobile smart device that is connected to the wearable device through communication. Figure 7 FIG. 1 is a schematic diagram of a display device provided by an embodiment of the present application. Figure 7 As shown, when displaying the material identification result, the display device 108 of the mobile smart device 700 may include a signal display area 701 and a material identification result display area 702. The signal display area may be used to display the triboelectric signal acquired by the triboelectric signal acquisition module 102; the material identification result display area may be used to display the final material identification result, for example, the material identification result display area may display the specific category of the material.

[0089] Exemplarily, the display device 108 may also include an audio playback device to enable the user to obtain the corresponding material identification result. In one example, the material category of the substance to be identified may be played through a speaker.

[0090] In the embodiment of the present application, the display device enables the user to obtain the material identification result intuitively and clearly, ensures the readability of the material identification result, and provides convenience for the user to use the wearable device to identify the material.

[0091] Furthermore, a material identification method is also provided in the embodiment of the present application. The material identification method in the embodiment of the present application is described below.

[0092] Figure 8 FIG. 1 is a flow chart of a material identification method provided by an embodiment of the present application. Figure 8 As shown, the material identification method includes the following steps:

[0093] S810, obtaining a target triboelectric signal.

[0094] Exemplarily, the target triboelectric signal may be a preprocessed signal, wherein the preprocessing may include performing baseline removal and noise reduction on the received triboelectric signal.

[0095] Exemplarily, the target triboelectric signal may be a current signal, or the target triboelectric signal may be a voltage signal. It is understood that the current signal or voltage signal may be a signal generated when the triboelectric signal acquisition module touches the substance to be identified.

[0096] In one example, there may be one or more triboelectric signal acquisition modules that generate triboelectric signals by touching the substance to be identified, and therefore, the target triboelectric signal may be a signal obtained by splicing one or more triboelectric signals.

[0097] S820, extracting features of the target triboelectric signal to obtain target feature data.

[0098] Exemplarily, the target feature data can be obtained by extracting the features of the target triboelectric signal. The target feature data can be data of different dimensions pre-set by the technician. In one example, the different dimensions in the target feature data can be obtained by the technician through different requirements or experiments.

[0099] S830. Based on the correspondence between material types and features pre-learned by the material recognition model, perform material recognition on the target feature data to obtain a material recognition result.

[0100] Exemplarily, the target feature data may be input into a material identification model, and the target feature data may be predicted by the material identification model to obtain a material identification result.

[0101] The material recognition model can be trained based on a large amount of training data in the database to ensure the accuracy and generalization ability of the model. In addition, when the material recognition model is trained, the CrossEntropyLoss method, that is, the following formula (1), is used as the loss function Loss of the material recognition model to achieve the processing of multi-classification tasks:

[0102]

[0103] Among them, i can represent the i-th training sample, yi can represent the material category predicted by the material recognition model for the i-th sample, pi can be the probability value that the i-th sample belongs to the material category, and C can represent the sample capacity.

[0104] It can be understood that in the embodiments of the present application, the above loss function is only used for exemplary description, and the loss function of the material identification model can also be constructed by the Focal Loss method and the Mean Squared Error method. The present application does not limit the loss function.

[0105] Exemplarily, after training, the material recognition model can obtain the relationship between the characteristic data and the material type, so that after the target characteristic data is input into the model, the material recognition of the substance to be recognized can be realized.

[0106] In the embodiment of the present application, the target triboelectric signal is feature extracted to obtain target feature data. Then, the corresponding relationship between the material type and the feature learned in advance by the material recognition model is used to perform material recognition on the target feature data to obtain a material recognition result. It can be understood that the use of the material recognition model can better explore the complex relationship between the triboelectric signal and the material category, thereby significantly improving the accuracy and efficiency of recognition.

[0107] Exemplarily, in order to achieve feature extraction of the target triboelectric signal, as another implementation of the present application, the present application also provides another implementation of the material identification method, see the following embodiments for details.

[0108] Fig. 9 FIG. 2 shows a schematic diagram of a material identification method provided by another embodiment of the present application. Fig. 9 As shown, the material identification method may include the following steps:

[0109] S910, obtaining a target triboelectric signal.

[0110] S920, extracting features of the target triboelectric signal to obtain original feature data, root mean square feature data, and charge feature data.

[0111] S930, concatenating the original characteristic data, the root mean square characteristic data, and the charge characteristic data to obtain target characteristic data.

[0112] S940. Based on the correspondence between material types and features pre-learned by the material recognition model, perform material recognition on the target feature data to obtain a material recognition result.

[0113] Exemplarily, step S910 is consistent with step S810, and step S940 is consistent with step S830, which will not be described in detail here.

[0114] In some embodiments, in S920, feature extraction may be performed on the target triboelectric signal to obtain original feature data, root mean square feature data, and charge feature data.

[0115] Exemplarily, the target friction signal may include a friction electric signal corresponding to the complete process of the friction electric signal acquisition module contacting the substance to be identified.

[0116] In some optional embodiments, the target triboelectric signal may be sampled and processed at a preset frequency to obtain raw characteristic data, perform root mean square calculation on the raw characteristic data to obtain root mean square characteristic data, and perform integration operation on the target triboelectric signal to obtain charge quantity characteristic data.

[0117] For example, the preset frequency can be predetermined by relevant technicians based on experience or needs. After sampling and processing the target triboelectric signal at the preset frequency, a plurality of discrete data points, namely, original characteristic data, can be obtained.

[0118] Exemplarily, the root mean square feature data may be obtained by calculating the average of the square values ​​of a plurality of discrete data points in the original feature data and then performing a square root operation.

[0119] Exemplarily, the charge quantity characteristic data may represent the charge quantity transfer value during the process in which the friction electric signal acquisition module contacts the substance to be identified.

[0120] In one example, the current or voltage in the resistor in parallel with the triboelectric signal acquisition module can be obtained as the target triboelectric signal, and the relative charge transfer value can be obtained by integrating the target triboelectric signal. The integration time can be determined according to the preset frequency. Then, the calculated relative charge transfer value is used as the charge characteristic data.

[0121] In an embodiment of the present application, feature data of different dimensions are obtained by sampling, performing root mean square operations, and performing integration operations on the target friction electric signal. This enables the material identification model to better capture the association between feature data of different dimensions and material categories, thereby enhancing the discriminative power of the material identification model.

[0122] In some embodiments, in S930, the extracted original feature data, root mean square feature data, and charge feature data may be concatenated to obtain target feature data.

[0123] Exemplarily, the original feature data, root mean square feature data and charge feature data can be spliced ​​according to different weights to obtain target feature data, wherein the weights corresponding to the original feature data, root mean square feature data and charge feature data can be determined by a technician.

[0124] In the embodiment of the present application, the feature recognition model realizes feature recognition by utilizing data of multiple dimensions, and can fully explore the complex relationship between the triboelectric signal and the material type, thereby improving the accuracy of the material recognition model.

[0125] Exemplarily, the material recognition model may be a fully connected model, which may include an input layer, a hidden layer, and a classification layer. In order to obtain a material recognition result, as another implementation of the present application, the present application also provides another implementation of a wearable device, which is specifically referred to in the following embodiments.

[0126] Fig.10 FIG. 2 shows a schematic diagram of a material identification method provided by another embodiment of the present application. Fig.10 As shown, the material identification method may include the following steps:

[0127] S1010, obtaining a target triboelectric signal.

[0128] S1020, extracting features of the target triboelectric signal to obtain target feature data.

[0129] S1030. Use the input layer to perform vector mapping on the target feature data to obtain a feature vector.

[0130] S1040, input the feature vector into the hidden layer, use the hidden layer to extract the high-dimensional vector of the feature vector, and obtain the target vector.

[0131] S1050. Determine the material recognition result corresponding to the target vector by using the classification layer.

[0132] Exemplarily, steps S1010-S1020 are consistent with steps S810-S820, and are not described in detail here.

[0133] In some embodiments, in S1030, vector mapping may be performed on the target feature data through an input layer in the material recognition model to obtain a feature vector.

[0134] The process of vector mapping the target feature data may be a process of mapping the discrete target feature data into a dense real-valued vector. That is, it is understood that the target feature data may be represented by a matrix of a preset dimension, and the preset dimension may be pre-set based on the needs or experience of relevant technicians.

[0135] In some embodiments, in S1040, the feature vector obtained through the input layer may be input into the hidden layer, and the hidden layer may be used to perform high-dimensional vector extraction on the feature vector to obtain a target vector.

[0136] In some embodiments, in S1050, a classification layer may be used to determine a material identification result corresponding to the target vector.

[0137] In some optional embodiments, the probability values ​​of the target vector belonging to different material categories can be determined respectively, and the material category corresponding to the maximum probability value is selected as the material recognition result. In this way, the user can intuitively and clearly obtain the material recognition result corresponding to the object to be recognized, thereby improving the user's experience.

[0138] Exemplarily, the classification layer can also determine the material identification result according to a preset classification threshold. In one example, the preset classification threshold can be set according to the probability value and the confidence level, for example, the preset threshold can be set to 0.8. That is, when it is determined that the result output by the classification layer is greater than 0.8, the corresponding material category can be used as the final material identification result.

[0139] Exemplarily, the material recognition model may also include a normalization layer, which improves training stability and accelerates model convergence. In addition, the material recognition model training process may introduce a dropout method to avoid overfitting of the training data and enhance the generalization ability and stability of the model.

[0140] In the embodiment of the present application, the input layer of the feature recognition model can be used to perform vector mapping on the target feature data to obtain a feature vector. The feature vector is input into the hidden layer, and the hidden layer is used to extract the high-dimensional vector of the feature vector to obtain the target vector. Then, the classification layer is used to determine the material recognition result corresponding to the target vector. In this way, the complex relationship between the triboelectric signal and the material category can be better explored, thereby significantly improving the accuracy and efficiency of recognition.

[0141] In order to ensure that the acquired target triboelectric signal can effectively reflect the process of the triboelectric signal acquisition module contacting the object to be identified, as one implementation method of the present application, the present application also provides another implementation method of the wearable device, please refer to the following embodiments for details.

[0142] Fig.11 FIG. 2 shows a schematic diagram of a material identification method provided by another embodiment of the present application. Fig.11 As shown, the material identification method may include the following steps:

[0143] S1110, obtaining original triboelectric signals.

[0144] Exemplarily, the original triboelectric signal may be a triboelectric signal detected by the triboelectric signal acquisition module. Alternatively, the original triboelectric signal may be a signal obtained by performing data processing on the triboelectric signal detected by the triboelectric signal acquisition module, such as filtering, amplification, signal conversion, etc.

[0145] S1120. When it is determined that the original triboelectric signal meets the trigger condition, use the original triboelectric signal as the target triboelectric signal.

[0146] The trigger condition may be that the original triboelectric signal is greater than or equal to a preset signal threshold. Alternatively, the trigger condition may be that the rate of change of the original triboelectric signal is greater than or equal to a preset rate of change threshold.

[0147] In one example, when it is determined that the original friction electric signal meets the trigger condition, the peak value (including the peak value and the valley value) of the original friction signal can be determined first, and then according to the relevant characteristics of the friction electric signal, the friction electric signal of preset lengths can be taken before and after the peak value of the original friction signal, and the friction electric signal of two preset lengths and the friction electric signal between the peak value and the valley value can be used as the target friction electric signal.

[0148] S1130, obtaining a target triboelectric signal.

[0149] S1140, extracting features of the target triboelectric signal to obtain target feature data.

[0150] S1150. Based on the correspondence between material types and features pre-learned by the material recognition model, perform material recognition on the target feature data to obtain a material recognition result.

[0151] Exemplarily, steps S1130-S1150 are consistent with steps S810-S830, and are not described in detail here.

[0152] In an embodiment of the present application, the target triboelectric signal is determined by triggering conditions, thereby ensuring that the triboelectric signal in the input material identification model can characterize the process of the triboelectric signal acquisition module contacting the object to be identified. This can ensure the accuracy of the input data of the material identification model and improve the accuracy of material identification.

[0153] Exemplarily, the material identification method in the application embodiment is described in conjunction with the following examples.

[0154] In one example, the material recognition model can be used to judge various types of fabrics or various types of plastic films. For example, the triboelectric signal acquisition module can be used to contact various types of fabrics respectively to obtain the following: Fig.12 Schematic diagram of the triboelectric signal shown. Fig.12A schematic diagram of a triboelectric signal provided by an embodiment of the present application is shown. The x-axis may be triboelectric signals acquired by different channels, i.e., different triboelectric signal acquisition modules, and the y-axis may be different fabrics, including silk fabric, acetate fabric, and polyester fabric, etc. The z-axis may be the voltage value corresponding to the triboelectric signal.

[0155] Furthermore, by processing the triboelectric signal accordingly and using the material identification model, the corresponding material identification result is obtained. Fig.13 A schematic diagram showing the accuracy of a material recognition model provided by an embodiment of the present application, wherein the horizontal axis represents the material category predicted by the material recognition model, and the vertical axis represents the actual category of the substance to be recognized. Fig.13 It can be seen that the material recognition model can accurately predict the material category of the object to be identified.

[0156] It is understandable that the material identification method in the embodiment of the present application can be applied to the wearable device in the embodiment of the present application. The wearable device realizes material identification of the object to be identified by implementing the material identification method in the above embodiment.

[0157] Based on the material identification method provided in the above embodiment, the present application also provides a specific implementation of a material identification device. Please refer to the following embodiment.

[0158] See first Fig.14 The material identification device provided in the embodiment of the present application includes the following modules:

[0159] An acquisition module 1401 is used to acquire a target triboelectric signal;

[0160] An extraction module 1402 is used to extract features of the target triboelectric signal to obtain target feature data;

[0161] The identification module 1403 is used to perform material identification on the target feature data based on the correspondence between material types and features pre-learned by the material identification model to obtain a material identification result.

[0162] The material identification device of the embodiment of the present application obtains the triboelectric signal through one or more triboelectric signal acquisition modules set on the wearable device body, and transmits the triboelectric signal obtained by the triboelectric signal acquisition module to the processor through the signal transmission module, so that the processor can determine the identification result of the material to be identified based on the triboelectric signal. It can be understood that the wearable device can be used to obtain the triboelectric signal for material identification in a convenient and flexible way. Further, the target feature data is obtained by extracting the features of the triboelectric signal; and the material identification result obtained by material identification of the target feature data based on the material identification model, so that the material identification result obtained has a high accuracy, and can realize convenient and accurate identification of materials.

[0163] As an implementation method of the present application, in order to obtain target characteristic data, 1402 uses the following method to extract features of the target triboelectric signal to obtain target characteristic data: extract features of the target triboelectric signal to obtain original characteristic data, root mean square characteristic data and charge characteristic data; splice the original characteristic data, root mean square characteristic data and charge characteristic data to obtain target characteristic data.

[0164] As an implementation method of the present application, feature extraction is performed on the target triboelectric signal to obtain original feature data, root mean square feature data and charge quantity feature data: the target triboelectric signal is sampled and processed at a preset frequency to obtain original feature data; root mean square calculation is performed on the original feature data to obtain root mean square feature data; integration operation is performed on the original feature data to obtain charge quantity feature data.

[0165] As an implementation method of the present application, the material identification model includes an input layer, a hidden layer and a classification layer; the identification module 1403 adopts the following method to perform material identification on the target feature data based on the correspondence between the material type and the characteristics pre-learned by the material identification model to obtain a material identification result: use the input layer to perform vector mapping on the target feature data to obtain a feature vector; input the feature vector into the hidden layer, and use the hidden layer to extract a high-dimensional vector from the feature vector to obtain a target vector; use the classification layer to determine the material identification result corresponding to the target vector.

[0166] As an implementation method of the present application, the recognition module 1403 uses the classification layer in the following manner to determine the material recognition result corresponding to the target vector: determine the probability values ​​of the target vector belonging to different material categories respectively; and select the material category corresponding to the maximum probability value as the material recognition result.

[0167] As an implementation method of the present application, the acquisition module 1401 is also used to acquire the original friction electric signal; when it is determined that the original friction electric signal meets the trigger condition, the original friction electric signal is used as the target friction electric signal; the trigger condition includes that the original friction electric signal is greater than or equal to a preset signal threshold; or the change rate of the original friction electric signal is greater than or equal to a preset change rate threshold.

[0168] Fig.15 FIG. 1 shows an architecture diagram of a material identification system provided by an embodiment of the present application. Fig.15 As shown, the material identification system 1500 may include:

[0169] Wearable device 100 and data processing device 1510. The data processing device is used to implement any one of the material identification methods in the above embodiments.

[0170] Fig.16 A schematic diagram of the hardware structure of material identification provided in an embodiment of the present application is shown.

[0171] The material identification device may include a processor 104 and a memory 1602 storing computer program instructions.

[0172] Specifically, the processor 104 may include a central processing unit (CPU), or an application specific integrated circuit (ASIC), or may be configured to implement one or more integrated circuits of the embodiments of the present application.

[0173] Memory 1602 may include a large capacity memory for data or instructions. By way of example and not limitation, memory 1602 may include a hard disk drive (HDD), a floppy disk drive, a flash memory, an optical disk, a magneto-optical disk, a magnetic tape, or a universal serial bus (USB) drive or a combination of two or more of these. In appropriate cases, memory 1602 may include removable or non-removable (or fixed) media. In appropriate cases, memory 1602 may be inside or outside the integrated gateway disaster recovery device. In a specific embodiment, memory 1602 is a non-volatile solid-state memory.

[0174] The memory may include read-only memory (ROM), random access memory (RAM), magnetic disk storage media devices, optical storage media devices, flash memory devices, electrical, optical or other physical / tangible memory storage devices. Thus, typically, the memory includes one or more tangible (non-transitory) computer-readable storage media (e.g., memory devices) encoded with software including computer-executable instructions, and when the software is executed (e.g., by one or more processors), it is operable to perform the operations described with reference to the method according to an aspect of the present disclosure.

[0175] The processor 104 implements any one of the material identification methods in the above embodiments by reading and executing the computer program instructions stored in the memory 1602 .

[0176] In one example, the material identification device may further include a communication interface 1603 and a bus 1610. Fig.16 As shown, the processor 104, the memory 1602, and the communication interface 1603 are connected via a bus 1610 and communicate with each other.

[0177] The communication interface 1603 is mainly used to implement communication between various modules, devices, units and / or equipment in the embodiments of the present application.

[0178] Bus 1610 includes hardware, software or both, and the parts of online data flow billing equipment are coupled to each other. For example, but not limitation, bus may include accelerated graphics port (AGP) or other graphics bus, enhanced industrial standard architecture (EISA) bus, front-end bus (FSB), hypertransport (HT) interconnection, industrial standard architecture (ISA) bus, infinite bandwidth interconnection, low pin count (LPC) bus, memory bus, micro channel architecture (MCA) bus, peripheral component interconnection (PCI) bus, PCI-Express (PCI-X) bus, serial advanced technology attachment (SATA) bus, video electronics standard association local (VLB) bus or other suitable bus or two or more of these combinations. In appropriate cases, bus 1610 may include one or more buses. Although the present application embodiment describes and shows a specific bus, the application considers any suitable bus or interconnection.

[0179] The material identification device can perform the material identification method in the embodiment of the present application based on the target triboelectric signal, thereby realizing the combination Figure 8 and Fig.14 A material identification method and apparatus are described.

[0180] In addition, in combination with the material identification method in the above embodiment, the embodiment of the present application can provide a computer storage medium for implementation. The computer storage medium stores computer program instructions; when the computer program instructions are executed by a processor, any one of the material identification methods in the above embodiment is implemented.

[0181] An embodiment of the present application also provides a computer program product, including a computer program, which implements any one of the material identification methods in the above embodiments when the computer program is executed by a processor.

[0182] It should be clear that the present application is not limited to the specific configuration and processing described above and shown in the figures. For the sake of simplicity, a detailed description of the known method is omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of the present application is not limited to the specific steps described and shown, and those skilled in the art can make various changes, modifications and additions, or change the order between the steps after understanding the spirit of the present application.

[0183] The functional blocks shown in the above-described block diagram can be implemented as hardware, software, firmware or a combination thereof. When implemented in hardware, it can be, for example, an electronic circuit, an application specific integrated circuit (ASIC), appropriate firmware, a plug-in, a function card, etc. When implemented in software, the elements of the present application are programs or code segments that are used to perform the required tasks. The program or code segment can be stored in a machine-readable medium, or transmitted on a transmission medium or a communication link by a data signal carried in a carrier wave. "Machine-readable medium" can include any medium capable of storing or transmitting information. Examples of machine-readable media include electronic circuits, semiconductor memory devices, ROM, flash memory, erasable ROM (EROM), floppy disks, CD-ROMs, optical disks, hard disks, optical fiber media, radio frequency (RF) links, etc. The code segment can be downloaded via a computer network such as the Internet, an intranet, etc.

[0184] It should also be noted that the exemplary embodiments mentioned in this application describe some methods or systems based on a series of steps or devices. However, this application is not limited to the order of the above steps, that is, the steps can be performed in the order mentioned in the embodiment, or in a different order from the embodiment, or several steps can be performed simultaneously.

[0185] Aspects of the present disclosure are described above with reference to the flowchart and / or block diagram of the method, device (system) and computer program product according to the embodiment of the present disclosure. It should be understood that each box in the flowchart and / or block diagram and the combination of each box in the flowchart and / or block diagram can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device to produce a machine so that these instructions executed by the processor of the computer or other programmable data processing device enable the implementation of the function / action specified in one or more boxes of the flowchart and / or block diagram. Such a processor can be, but is not limited to, a general-purpose processor, a special-purpose processor, a special application processor, or a field programmable logic circuit. It can also be understood that each box in the block diagram and / or flowchart and the combination of boxes in the block diagram and / or flowchart can also be implemented by dedicated hardware that performs a specified function or action, or can be implemented by a combination of dedicated hardware and computer instructions.

[0186] The above is only a specific implementation of the present application. Those skilled in the art can clearly understand that for the convenience and simplicity of description, the specific working processes of the systems, modules and units described above can refer to the corresponding processes in the aforementioned method embodiments, and will not be repeated here. It should be understood that the protection scope of the present application is not limited to this. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed in this application, and these modifications or replacements should be included in the protection scope of this application.

Claims

1. A wearable device, characterized in that: include: A wearable device body and at least one triboelectric signal acquisition module mounted on the body; A signal transmission module connected to the triboelectric signal acquisition module, used for transmitting the triboelectric signal acquired by the triboelectric signal acquisition module; A processor, connected to the signal transmission module, and configured to obtain the triboelectric signal; Determine a material identification result according to the triboelectric signal, wherein the material identification result is obtained by extracting features from the triboelectric signal to obtain target feature data; Based on the correspondence between material types and features pre-learned by the material recognition model, material recognition is performed on the target feature data to obtain a material recognition result.

2. The wearable device according to claim 1, characterized in that: The wearable device body includes a glove, and the triboelectric signal acquisition module is arranged at a material contact position on one or more fingers of the glove.

3. The wearable device according to claim 1, characterized in that: The signal transmission module comprises: A conditioning circuit connected to the triboelectric signal acquisition module; An analog-to-digital signal conversion circuit connected to the conditioning circuit; A control unit is connected to the analog-to-digital signal conversion circuit.

4. The wearable device according to claim 1, characterized in that: The wearable device further comprises: A display device is connected to the processor and is used to display the material identification result determined by the processor.

5. The wearable device according to claim 1, characterized in that: The processor is also used to extract features from the friction electric signal to obtain target feature data; based on the correspondence between material types and features pre-learned by a material identification model, perform material identification on the target feature data to obtain the material identification result.

6. A material identification method, characterized in that: The method comprises: Acquire target triboelectric signal; Extracting features of the target triboelectric signal to obtain target feature data; Based on the correspondence between material types and features pre-learned by the material recognition model, material recognition is performed on the target feature data to obtain a material recognition result.

7. The method according to claim 6, characterized in that The step of extracting features from the target triboelectric signal to obtain target feature data includes: Extracting features of the target triboelectric signal to obtain original feature data, root mean square feature data, and charge feature data; The original characteristic data, the root mean square characteristic data and the charge quantity characteristic data are spliced ​​to obtain target characteristic data.

8. The method according to claim 7, characterized in that The feature extraction of the target triboelectric signal to obtain original feature data, root mean square feature data and charge feature data includes: Sampling the target triboelectric signal at a preset frequency to obtain the original characteristic data; Performing root mean square calculation on the original feature data to obtain the root mean square feature data; The target friction electric signal is integrated to obtain the charge quantity characteristic data.

9. The method according to claim 6, characterized in that The material recognition model includes an input layer, a hidden layer and a classification layer; The material identification result is obtained by performing material identification on the target feature data based on the correspondence between the material type and the feature pre-learned by the material identification model, including: Use the input layer to perform vector mapping on the target feature data to obtain the feature vector; Inputting the feature vector into the hidden layer, and using the hidden layer to perform high-dimensional vector extraction on the feature vector to obtain a target vector; The classification layer is used to determine a material recognition result corresponding to the target vector.

10. The method according to claim 6, characterized in that Before acquiring the target triboelectric signal, the method further includes: Obtaining raw triboelectric signals; In the case where it is determined that the original triboelectric signal meets the trigger condition, the original triboelectric signal is used as the target triboelectric signal; The trigger conditions include that the original friction electric signal is greater than or equal to a preset signal threshold; or the change rate of the original friction electric signal is greater than or equal to a preset change rate threshold.

Images