Preparation method of hydrogel fiber crack strain sensor and sign language recognition system

Abstract

The invention discloses a preparation method of a hydrogel fiber crack strain sensor and a sign language recognition system. The method comprises the following steps: preparing hydrogel fibers; binding a silver wire serving as an electrode on the hydrogel fiber, and placing the silver wire on a substrate; connecting the two ends of the hydrogel fiber through conductive paste so as to enable the hydrogel fiber to form a closed loop; a packaging layer is covered; and creating a crack between the two electrodes, and coating conductive paste on the fracture surface of the hydrogel fiber at the crack to obtain the hydrogel fiber crack strain sensor. According to the embodiment of the invention, the crack is created in the hydrogel fiber, the conductive path is changed, the performance of the sensor is improved, the sensitivity is improved, and the response / recovery time is shortened; and the sign language recognition precision is improved. The method can be widely applied to the technical field of sensor detection.

Description

Technical Field

[0001] The present invention relates to the technical field of detection sensors, and in particular to a preparation method of a hydrogel fiber crack strain sensor and a sign language recognition system. Background Art

[0002] Sign language is a non-verbal communication method based on human gestures and body postures. It has been widely used in the deaf community, but its abstractness and regional diversity may cause communication barriers for those who do not understand it. Currently, various technologies for recognizing human gestures, including visual image processing, electromyography, and wearable strain sensors, play an important role in improving communication in the deaf community. Conductive hydrogels have the characteristics of soft tissue-like hydrated structures, with advantages such as adjustable softness, inherent stretchability, interesting electromechanical responsiveness, and biocompatibility; they are widely used in sign language signal processing. In related technologies, the inherent limitations of traditional bulk hydrogels in the strain sensing process, such as the purely geometric effect working mechanism and insufficient resilience, lead to the shortcomings of hydrogel-based strain sensing, such as low sensitivity, insufficient detection limit, long response / recovery time, and poor cyclability. For example, the low sensitivity and insufficient detection limit of traditional bulk hydrogels in strain monitoring cannot effectively capture subtle strain changes and cannot meet application scenarios with high precision requirements. Summary of the invention

[0003] The purpose of the present invention is to solve one of the technical problems existing in the prior art to at least a certain extent.

[0004] To this end, the purpose of the present invention is to provide a method for preparing a highly sensitive hydrogel fiber crack strain sensor and a sign language recognition system.

[0005] In order to achieve the above technical objectives, one aspect of an embodiment of the present invention provides a method for preparing a hydrogel fiber crack strain sensor, comprising the following steps: preparing a hydrogel fiber; tying a silver wire as an electrode to the hydrogel fiber and placing it on a substrate; connecting the two ends of the hydrogel fiber through a conductive paste so that the hydrogel fiber forms a closed loop; and covering the packaging layer; creating a crack between the two electrodes, and applying a conductive paste on the cross section of the hydrogel fiber at the crack to obtain a hydrogel fiber crack strain sensor. The embodiment of the present application improves the performance of the sensor by creating a crack in the hydrogel fiber, changing the conductive path, and thereby improving the sensitivity and reducing the response / recovery time; thereby improving the accuracy of sign language recognition.

[0006] In some embodiments, in the method for preparing the hydrogel fiber crack strain sensor of the embodiment of the present invention, the substrate includes an elastomer, and the substrate is prepared by the following steps:

[0007] The elastomer is obtained by a salt template method;

[0008] Alternatively, the elastomer is obtained by spin coating.

[0009] In some embodiments, in one embodiment of the present invention, the substrate comprises an Ecoflex substrate, and the Ecoflex substrate is prepared by the following steps:

[0010] mixing NaCl and Ecoflex prepolymer to obtain a NaCl / Ecoflex composite material;

[0011] coating the NaCl / Ecoflex composite material on a metal substrate at a preset rotation speed to obtain a NaCl / Ecoflex composite film;

[0012] The NaCl / Ecoflex composite film is immersed, ultrasonicated and dried to obtain an Ecoflex substrate.

[0013] In some embodiments, in one embodiment of the present invention, the preparation of hydrogel fiber comprises:

[0014] The hydrogel precursor solution is heated to obtain first fibers; and the first fibers are soaked to obtain hydrogel fibers.

[0015] In some embodiments, in one embodiment of the present invention, the step of heating the hydrogel precursor solution to obtain the first fiber comprises:

[0016] The capillary glass tube containing the hydrogel precursor solution is subjected to a heating treatment to form a first network, and the fiber containing the first network is taken out from the capillary glass tube to obtain a first fiber;

[0017] The step of soaking the first fiber to obtain the hydrogel fiber comprises:

[0018] Soaking the first fiber in a first solution of a preset concentration for a preset time to form a second network to obtain a hydrogel fiber;

[0019] The hydrogel fiber is immersed in a second solution with a preset mass proportion for a preset time period to optimize the hydrogel fiber.

[0020] In some embodiments, in one embodiment of the present invention, the method further comprises:

[0021] The ratio of long to short fiber segments is fixed, and the strain threshold causing the sudden change in sensor response is reduced by increasing the length of the crack;

[0022] Alternatively, the length of the crack is fixed, and the amplitude and sensitivity of the sudden change in sensor response are improved by increasing the ratio of long to short fiber segments; the ratio of long to short fiber segments is the ratio of the first segment to the second segment, and the hydrogel fiber closed loop is divided into the first segment and the second segment by the electrode, and the length of the first segment is greater than that of the second segment.

[0023] In some embodiments, in one embodiment of the present invention, before the step of covering the encapsulation layer, the method further comprises:

[0024] Ecoflex prepolymer was poured onto the substrate to fix the hydrogel fibers.

[0025] On the other hand, an embodiment of the present invention provides a sign language recognition system, including:

[0026] A hardware module, a software module and a wearable module; the wearable module is connected to the hardware module, and the hardware module is connected to the software module;

[0027] The wearable module includes: a sensor obtained by the preparation method of the hydrogel fiber crack strain sensor; the sensor is used to: be placed at the knuckle of the target finger to be identified, and collect a first signal related to the sign language of the target finger; the hardware module is used to process the first signal; and the software module is used to determine a prediction result of the sign language meaning based on the first signal.

[0028] In one embodiment of the present invention, the hardware module includes:

[0029] A digital synthesizer and a voltage follower; the second signal generated by the digital synthesizer is applied to the sensor through the voltage follower;

[0030] A transimpedance amplifier, a microprocessor, and a wireless communication unit; the first signal is transmitted to the microprocessor through the transimpedance amplifier; the microprocessor processes the first signal and transmits it to the software module through the wireless communication unit.

[0031] In one embodiment of the present invention, the software module includes a neural network unit, which is used to: obtain a first signal sample; input the first signal sample into the neural network unit for prediction to obtain a sample prediction result; update the parameters of the neural network unit according to the relationship between the sample prediction result and the sample true result to obtain the trained neural network unit, and perform sign language recognition through the trained neural network unit.

[0032] The embodiments of the present application include at least the following beneficial effects: preparing hydrogel fibers; tying silver wires as electrodes to the hydrogel fibers and placing them on a substrate; connecting the two ends of the hydrogel fibers through a conductive paste so that the hydrogel fibers form a closed loop; and covering the encapsulation layer; creating a crack between the two electrodes, and applying a conductive paste on the cross section of the hydrogel fibers at the crack to obtain a hydrogel fiber crack strain sensor. The embodiments of the present application improve the performance of the sensor by creating cracks in the hydrogel fibers, changing the conductive path, and thereby improving the sensitivity and reducing the response / recovery time; thereby improving the accuracy of sign language recognition. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the embodiments of the present invention or the drawings of related technical solutions in the prior art are introduced below. It should be understood that the drawings introduced below are only for the convenience of clearly describing some embodiments of the technical solutions of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0034] Figure 1 A schematic flow chart of an embodiment of a method for preparing a strain sensor provided in an embodiment of the present application;

[0035] Figure 2 A schematic diagram of a device of an embodiment of a strain sensor provided in an embodiment of the present application;

[0036] FIG3( a ) is a comparison of the mass loss of deionized water in an experiment between the sensor provided in an embodiment of the present application and the sensor in the related art;

[0037] FIG3( b ) is a comparison of the sensitivity of the sensor provided in the embodiment of the present application and the sensor in the related art in an experiment;

[0038] FIG3( c ) is a comparison of the resistance response of the sensor provided in the embodiment of the present application and the sensor in the related art as the strain increases in the experiment;

[0039] FIG3( d ) is a comparison of the cyclic resistance response of the sensor provided in the embodiment of the present application and the sensor in the related art to weak strain in the experiment;

[0040] FIG3( e ) is a comparison of the response stability of the sensor provided in the embodiment of the present application and the sensor in the related art during loading and unloading in an experiment;

[0041] FIG3( f ) is a comparison of the response / recovery time of the sensor provided in the embodiment of the present application and the sensor in the related art in an experiment;

[0042] Figure 4A schematic diagram of an embodiment of a response signal of a strain sensor provided in an embodiment of the present application during sign language recognition;

[0043] FIG5( a ) is a schematic diagram of an embodiment of a sign language recognition system based on a strain sensor provided in an embodiment of the present application;

[0044] FIG5( b ) is a schematic diagram of a sign language recognition scenario based on a strain sensor provided in an embodiment of the present application;

[0045] FIG5( c ) is a schematic diagram of an interface for displaying sign language recognition based on a strain sensor provided in an embodiment of the present application;

[0046] FIG5( d ) is a schematic diagram of the probability distribution of sign language recognition results based on strain sensors provided in an embodiment of the present application. DETAILED DESCRIPTION

[0047] The embodiments of the present invention are described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and are not to be construed as limitations of the present invention. For the step numbers in the following embodiments, they are only provided for the convenience of explanation, and the order between the steps is not limited in any way, and the execution order of each step in the embodiment can be adaptively adjusted according to the understanding of those skilled in the art.

[0048] The flexible strain sensor in the embodiment of the present application is a device that can detect and convert strain changes into electrical signal changes, making it practical in human motion detection and medical health applications. The embodiment provided in the present application can be used for flexible sensor devices for human motion detection, intelligent sign language recognition and translation systems.

[0049] The deaf community is a unique group that faces considerable barriers to effective communication and interaction with hearing people due to inherent limitations in their hearing and speaking abilities in daily life. Sign language, a nonverbal communication method based on human gestures and body postures, is widely used in the deaf community, but its abstractness and geographical diversity can create communication barriers for those who do not understand it. Currently, various technologies for recognizing human gestures, including visual image processing, electromyography, and wearable strain sensors, play an important role in improving communication in the deaf community. However, the accuracy of sign language recognition is limited by the ambient background and light intensity in visual image processing, and by electromagnetic noise and strict positioning requirements in electromyography.

[0050] Fortunately, compared with rigid sensors, stretchable and wearable strain sensors can be conformally adhered or fixed to various key positions of the human body, facilitating the conversion of various physical and mechanical strain information in the human body into different electrical signals (resistance, capacitance, voltage, etc.) for further processing and analysis, thereby providing a more convenient and effective communication tool for the deaf and dumb. Conductive hydrogels have the characteristics of soft tissue-like hydrated structures, with advantages such as adjustable softness, inherent stretchability, interesting electromechanical responsiveness, and biocompatibility. With this in mind, they are widely used as strain-sensitive layers for real-time monitoring of body movements, providing basic data samples required for accurate sign language recognition. For example, one embodiment is a three-dimensional (3D) printed wearable strain sensor based on an injectable and adhesive agarose double network (DN) hydrogel, with a strain gauge factor (GF) of 4.33, which can accurately detect a variety of gestures. One embodiment prepares a highly stretchable, fast self-healing conductive hydrogel containing polyaniline nanoparticles for the manufacture of a flexible strain sensor with a GF of 1.46 and a low detection limit (0.5% strain) within a strain range of 0% to 50%, thereby achieving reliable monitoring of various external stimuli and bending of human fingers.

[0051] In order to obtain high-quality sign language information signals, the need to simultaneously detect fine and large-scale gestures requires strain sensors with high sensitivity and a wide working range. However, the sensing mechanism of most hydrogel-based strain sensors relies on geometric effects, namely, the simultaneous elongation of the hydrogel and the reduction of the cross-sectional area with tensile strain lead to limited resistance change. Therefore, traditional bulk hydrogels suffer from low strain sensitivity (usually GF less than 10) and insufficient detection limit. In addition, the inherent limited elasticity of hydrogels leads to prolonged response / recovery time for strain sensing. Therefore, it remains a huge challenge to design stretchable strain sensors with high performance (e.g., high sensitivity, wide detection range, fast response / recovery speed, stability) to capture complex sign language gestures through simple and cost-effective manufacturing techniques. In addition, the importance of breathability in wearable devices is often overlooked, leading to skin moisture accumulation and discomfort when worn for a long time.

[0052] At present, the inherent limitations of traditional bulk hydrogels in the strain sensing process, such as the purely geometric effect working mechanism and insufficient resilience, lead to the shortcomings of hydrogel-based strain sensing, such as low sensitivity, insufficient detection limit, long response / recovery time, and poor cyclability. For example, the low sensitivity and insufficient detection limit of traditional bulk hydrogels in strain monitoring cannot effectively capture subtle strain changes and cannot meet application scenarios with high precision requirements. The long response / recovery time limits its application effectiveness in real-time monitoring that requires rapid response, while poor cyclability may cause its performance to gradually decline or fail in long-term use. Therefore, it is necessary to comprehensively consider the innovation of materials and structural design to improve the overall sensing performance of hydrogel-based flexible strain sensors to meet the increasingly complex and diverse application needs.

[0053] In addition, the breathability of wearable devices is often overlooked in the design and selection process, but its importance for long-term wear cannot be underestimated. Lack of breathability may cause the wearer's skin to become moist during use, which may cause the sensor to be disturbed by the surrounding environment, thereby affecting the accuracy and stability of the data. At the same time, non-breathable wearable devices will limit the ventilation of the skin, which may cause discomfort and potential skin problems (eczema, itching or other skin problems) to the wearer, especially for people with sensitive skin. Therefore, when developing and applying wearable strain sensors, full consideration of breathability is one of the key factors to ensure the long-term stable operation of the device and the health of the wearer.

[0054] The following describes in detail a method and system for preparing a hydrogel fiber crack strain sensor according to an embodiment of the present invention with reference to the accompanying drawings. First, the method for preparing a hydrogel fiber crack strain sensor according to an embodiment of the present invention will be described with reference to the accompanying drawings.

[0055] An embodiment of the present invention provides a method for preparing a hydrogel fiber crack strain sensor. The method for preparing the hydrogel fiber crack strain sensor in the embodiment of the present invention mainly comprises the following steps:

[0056] S100: preparing hydrogel fibers;

[0057] S200: tying silver wire as an electrode to the hydrogel fiber and placing it on a substrate;

[0058] S300: connecting two ends of the hydrogel fiber through a conductive paste so that the hydrogel fiber forms a closed loop; and covering with a packaging layer;

[0059] S400: creating a crack between two electrodes, and applying a conductive paste on the cross section of the hydrogel fiber at the crack to obtain a hydrogel fiber crack strain sensor;

[0060] In some possible implementations, the hydrogel fiber in the embodiments of the present application may include two networks, and the present application does not limit the specific preparation process of the hydrogel fiber. For example, Figure 1 This is a preparation flow chart of an embodiment of a crack strain sensor based on hydrogel fibers. The specific process is as follows: First, a capillary glass tube containing a hydrogel precursor solution is placed at 70°C and heated for 1 hour to form a first covalently cross-linked polyacrylamide (PAM) network (i.e., the first network in the embodiment of the present application). Subsequently, the fiber is taken out of the glass capillary and immersed in a 1M CaCl2 solution for 20 minutes to obtain a second ion-crosslinked alginate network (i.e., the second network in the embodiment of the present application). The fiber is then immersed in a 50wt% LiBr solution for 20 minutes to prepare a PAM / Ca-Alg double network hydrogel fiber that is resistant to drying and freezing. The silver wire is tied to the middle of the polyacrylamide / calcium alginate (PAM / Ca-Alg) hydrogel fiber as an electrode, and the distance between the electrodes is usually controlled to be about 1 cm. The fiber with silver wire is placed on a porous Ecoflex elastomer substrate, and a conductive silver paste is applied to the ends of the two fibers to achieve complete connection and form a closed loop, thereby forming a good and stable electrical contact. Then, a small amount of Ecoflex prepolymer is poured on the Ecoflex film containing the fiber to fix the hydrogel fiber and quickly covered with another layer of the same porous Ecoflex elastomer encapsulation film. After Ecoflex is fully cured, scissors are used to produce a crack of custom length at the midpoint of the short fiber between the two electrodes. The substrate at the crack needs to be completely cracked and penetrated. Finally, a conductive silver paste is applied to the fiber cross section at the crack to strengthen the contact of the fiber cross section, thereby forming the designed crack strain sensor. It is understandable that the above-mentioned use of experimental parameters and experimental consumables are all illustrative examples, and this application does not limit the parameters and related consumables in the experimental process. The cracks in the embodiments of the present application run through the substrate and the encapsulation layer. The crack is a closed-loop local crack perpendicular to the hydrogel fiber, and the crack length will affect the sensing performance of the device (such as the response jump threshold and range). The components of the hydrogel fiber can be other system materials. In some embodiments, the substrate and the encapsulation layer can both be porous Ecoflex elastomer films.

[0061] The hydrogel fiber-based crack strain sensor of the present invention has a multi-layer stacking structure. Figure 1 and Figure 2They are respectively the preparation flow chart and device schematic diagram of the crack strain sensor. Among them, hydrogel fibers 202 are fixed in the porous Ecoflex substrate and packaging film 201, silver wires 204 are arranged on the hydrogel fibers, and cracks 203 are arranged on the sensor between the two connection points of the silver wires and the hydrogel fibers. It can be understood that in the structure of the sensor, the cyclic polyacrylamide (PAM) / calcium alginate (Ca-Alg) double network hydrogel fibers connected at both ends are strain sensitive layers, the silver wires tied to the hydrogel fibers are electrodes at both ends, and the porous Ecoflex elastomer packaging membrane (PEEM) is the flexible substrate and packaging layer of the device. Finally, an innovative adjustable crack structure is introduced by cutting the hydrogel fibers to improve the performance of the sensor.

[0062] Optionally, in one embodiment of the present invention, the substrate comprises an elastomer, and the substrate is prepared by the following steps:

[0063] The elastomer is obtained by a salt template method;

[0064] Alternatively, the elastomer is obtained by spin coating.

[0065] In some possible embodiments, both the substrate and the encapsulation layer in the embodiments of the present application may include an elastomer. Specifically, a porous elastomer is obtained by a salt template method; or, a non-porous elastomer is obtained by a spin coating method. Of course, the embodiments of the present application do not limit the specific type of elastomer. For porous elastomers, they can be prepared by a salt template method. Specifically, a salt template is determined; the salt template is mixed with a prepolymer to obtain a composite material; the composite material and the salt template are subjected to spin coating to obtain a composite film; the composite film is immersed, ultrasonicated, and dried to obtain a porous elastomer. It can be understood that the salt template usually selects a salt with a specific particle size distribution, which can be sodium chloride, potassium chloride, etc.

[0066] The porous Ecoflex elastomer encapsulation membrane prepared by the salt template method is used as the substrate and encapsulation layer of the device, which gives the crack strain sensor good air permeability. As shown in Figure 3(a), the mass loss of deionized water in two beakers is compared under the coverage of the sensor based on the porous / non-porous Ecoflex encapsulation membrane. The results show that the water loss of the sensor based on the porous Ecoflex elastomer encapsulation membrane (PEEM) is about three times that of the non-porous Ecoflex elastomer encapsulation membrane (NEEM), indicating that the sensor based on PEEM has better air permeability.

[0067] The working mechanism of the crack sensor designed in the present invention relies on the resistance change caused by the sharp switching of the conductive path triggered by the closing / opening of the pre-cut crack and the geometric deformation effect of the hydrogel fiber under external strain loading / unloading, which gives the hydrogel-based strain sensor excellent comprehensive sensing performance. Figure 3(b) to Figure 3(f) The sensing performance of the breathable crack strain sensor based on the porous Ecoflex elastomer packaging film. Figure 3(b) is a comparison of the sensitivity of the sensor provided in the embodiment of the present application and the sensor in the related art in the experiment; Figure 3(c) is a comparison of the resistance response of the sensor provided in the embodiment of the present application and the sensor in the related art with the increase of strain in the experiment; Figure 3(d) is a comparison of the cyclic resistance response of the sensor provided in the embodiment of the present application and the sensor in the related art to weak strain in the experiment; Figure 3(e) is a comparison of the response stability of the sensor provided in the embodiment of the present application and the sensor in the related art during loading and unloading in the experiment; Figure 3(f) is a comparison of the response / recovery time of the sensor provided in the embodiment of the present application and the sensor in the related art in the experiment. It can be seen that the permeable strain sensor based on porous Ecoflex elastomer encapsulation membrane exhibits impressive sensing performance, including ultrahigh sensitivity (1188.5), wide sensing range (0.02%-80%), low detection limit (0.02%), short response / recovery time (78 / 52ms) and good repeatability (5000 cycles), which is superior to the traditional block hydrogel flexible strain sensor in terms of sensitivity, detection limit and response / recovery time.

[0068] Optionally, in one embodiment of the present invention, the substrate comprises an Ecoflex substrate, and the Ecoflex substrate is prepared by the following steps:

[0069] mixing NaCl and Ecoflex prepolymer to obtain a NaCl / Ecoflex composite material;

[0070] coating the NaCl / Ecoflex composite material on a metal substrate at a preset rotation speed to obtain a NaCl / Ecoflex composite film;

[0071] The NaCl / Ecoflex composite film is immersed, ultrasonicated and dried to obtain an Ecoflex substrate.

[0072] In some possible embodiments, the porous Ecoflex elastomer substrate and the encapsulation membrane are prepared based on the salt template method, and NaCl and Ecoflex prepolymer are uniformly mixed in different mass ratios (1: 1) to obtain a NaCl / Ecoflex composite material, which is then spin-coated on an aluminum substrate at a speed of 600 rpm. The obtained NaCl / Ecoflex composite membrane is subjected to subsequent curing and stripping steps, immersed in deionized water, ultrasonically treated for 12 hours to completely dissolve the NaCl particles, and then dried at 60 ° C for 2 hours to obtain a porous Ecoflex elastomer encapsulation membrane (PEEM) substrate. Of course, it is understandable that a feasible substrate can also be obtained by appropriately adjusting the above-mentioned experimental parameters and experimental consumables. The encapsulation layer in the embodiment of the present application can also be prepared by the same steps.

[0073] It should be noted that the porous elastomer in the present application may also be polydimethylsiloxane (PDMS), based on which, the substrate used in the embodiments of the present application is a PDMS substrate. The preparation process of the PDMS substrate is substantially the same as that of the Ecoflex substrate, and those skilled in the art can adjust the mixing ratio of NaCl and PDMS prepolymer according to actual needs to prepare a PDMS substrate that meets the requirements. Of course, the elastomer in the present application may also be made of other materials, and the present application does not limit the specific material of the elastomer.

[0074] The air permeability test in the embodiment of the present application: two identical beakers filled with hot water are covered and sealed using crack strain sensors based on non-porous and porous Ecoflex packaging films, respectively, and then the water mass change is measured in an environment of 25°C and 60% relative humidity for the same time to evaluate the air permeability of the device.

[0075] Strain sensing test: The prepared crack strain sensor was subjected to quantitative tensile strain on a homemade tensile device, and an LCR digital bridge tester with set parameters (voltage of 1 V and frequency of 1 kHz) was used to record the resistance change of the sensor during the stretching process.

[0076] Optionally, in one embodiment of the present invention, the preparation of the hydrogel fiber comprises:

[0077] The hydrogel precursor solution is heated to obtain first fibers; and the first fibers are soaked to obtain hydrogel fibers.

[0078] Optionally, in one embodiment of the present invention, the step of heating the hydrogel precursor solution to obtain the first fiber comprises:

[0079] The capillary glass tube containing the hydrogel precursor solution is subjected to a heating treatment to form a first network, and the fiber containing the first network is taken out from the capillary glass tube to obtain a first fiber;

[0080] The step of soaking the first fiber to obtain the hydrogel fiber comprises:

[0081] Soaking the first fiber in a first solution of a preset concentration for a preset time to form a second network to obtain a hydrogel fiber;

[0082] The hydrogel fiber is immersed in a second solution with a preset mass proportion for a preset time period to optimize the hydrogel fiber.

[0083] The first solution in the embodiment of the present application may be a solution containing Ca ions, and the second solution may be a solution containing Li ions. Therefore, the soaking treatment of the first fiber to obtain the hydrogel fiber includes: soaking the first fiber in a CaCl2 solution of a preset concentration for a preset time to obtain the hydrogel fiber; soaking the hydrogel fiber in a LiBr solution of a preset mass percentage for a preset time to optimize the hydrogel fiber.

[0084] Optionally, in one embodiment of the present invention, the method further includes:

[0085] The ratio of long to short fiber segments is fixed, and the strain threshold causing the sudden change in sensor response is reduced by increasing the length of the crack;

[0086] Alternatively, the length of the crack is fixed, and the amplitude and sensitivity of the sudden change in sensor response are improved by increasing the ratio of long to short fiber segments; the ratio of long to short fiber segments is the ratio of the first segment to the second segment, and the hydrogel fiber closed loop is divided into the first segment and the second segment by the electrode, and the length of the first segment is greater than that of the second segment.

[0087] In some possible implementations, when the ratio of the long / short fiber lengths is the same, the longer the crack length, the smaller the strain threshold causing the sensor response mutation; when the crack lengths are the same, the larger the ratio of the long / short fiber lengths, the higher the amplitude and sensitivity causing the sensor response mutation;

[0088] There are two fiber segments between the two silver wire electrodes in the hydrogel fiber loop, the long fiber segment is the long fiber segment, and the short fiber segment is the short fiber segment. The distance between the electrodes is usually controlled at about 1 cm, that is, the length of the short fiber segment is 1 cm.

[0089] Optionally, in one embodiment of the present invention, before the step of covering the encapsulation layer, the method further comprises:

[0090] Ecoflex prepolymer was poured onto the substrate to fix the hydrogel fibers.

[0091] In some possible embodiments, the steps of pouring silver paste and Ecoflex have no effect on the composition of the raw materials. The conductive silver paste is applied to achieve complete connection between the two ends of the fiber and form a closed loop. A small amount of Ecoflex prepolymer is poured onto the fiber to fix the position of the hydrogel fiber.

[0092] In summary, the above method prepares hydrogel fibers; ties silver wires as electrodes to the hydrogel fibers and places them on a substrate; connects the two ends of the hydrogel fibers through a conductive paste so that the hydrogel fibers form a closed loop; and covers the encapsulation layer; creates a crack between the two electrodes, and applies a conductive paste on the cross section of the hydrogel fibers at the crack to obtain a hydrogel fiber crack strain sensor. The embodiment of the present application improves the performance of the sensor by creating cracks in the hydrogel fibers, changing the conductive path, and thereby improving the sensitivity and reducing the response / recovery time; thereby improving the accuracy of sign language recognition.

[0093] Secondly, a sign language recognition system proposed in an embodiment of the present invention specifically includes:

[0094] A hardware module, a software module and a wearable module; the wearable module is connected to the hardware module, and the hardware module is connected to the software module;

[0095] The wearable module includes: a sensor obtained by the preparation method of the hydrogel fiber crack strain sensor as described above; the sensor is used to: be placed at the knuckle of the target finger to be identified, and collect a first signal related to the sign language of the target finger; the hardware module is used to process the first signal; and the software module is used to determine a prediction result of the sign language meaning based on the first signal.

[0096] The present invention fixes five crack strain sensors on the knuckles of five fingers on the human hand to form a five-channel sensing module, and monitors the resistance change of the device in real time to conduct recognition experiments of different sign language movements. Figure 4 As shown in FIG. 5 , when the human body makes different gestures according to different numbers and letters and common phrases in American Sign Language, different response patterns are generated in the five sensing channels corresponding to the five fingers, forming distinguishable characteristic combination signals corresponding to the meaning of each sign language. Further, as shown in FIG. 5 , an integrated wireless sign language recognition system is constructed by combining a five-channel strain sensing module, a wireless transmission circuit and a machine learning algorithm (i.e., a hardware module and a software module), which realizes wireless transmission and remote real-time display of different sign languages, and realizes high-precision recognition of different sign languages ​​(recognition accuracy rate is as high as 98.1%), providing a highly potential way to promote barrier-free communication for the deaf and mute.

[0097] Optionally, in one embodiment of the present invention, the hardware module includes:

[0098] A digital synthesizer and a voltage follower; the second signal generated by the digital synthesizer is applied to the sensor through the voltage follower;

[0099] A transimpedance amplifier, a microprocessor, and a wireless communication unit; the first signal is transmitted to the microprocessor through the transimpedance amplifier; the microprocessor processes the first signal and transmits it to the software module through the wireless communication unit.

[0100] Optionally, in one embodiment of the present invention, the software module includes a neural network unit, and the neural network unit is used to: obtain a first signal sample; input the first signal sample into the neural network unit for prediction to obtain a sample prediction result; update the parameters of the neural network unit according to the relationship between the sample prediction result and the sample true result to obtain the trained neural network unit, and perform sign language recognition through the trained neural network unit.

[0101] The following is a detailed description of the sign language recognition system provided by the present application using a specific embodiment, specifically:

[0102] Refer to Figure 5(a), which is a circuit diagram of the signal flow of the data wireless transmission module; Figure 5(b) is a schematic diagram of sign language communication between deaf-mute people through a wireless sign language recognition system; Figure 5(c) is a real-time monitoring of various sign language gestures on a smartphone interface using a crack strain sensor and a wireless transmission system; Figure 5(d) is a heat map of the sign language recognition probability distribution predicted by the decoding five-channel sensing module.

[0103] The signal transmission process is as follows: First, the direct digital synthesizer (DDS) module generates a sine wave signal with a frequency of 200Hz and an amplitude of 600mV (i.e., the second signal in the embodiment of the present application). The signal is adjusted by a voltage follower and then applied to the prepared crack strain sensor. The current signal (i.e., the first signal in the embodiment of the present application) output in real time from the sensor is converted and amplified into a voltage signal by a transimpedance amplifier (TIA). After signal processing, it is collected by the analog-to-digital converter (ADC) of the STM32 microcontroller and then wirelessly transmitted to the terminal device via a Bluetooth module. Finally, as shown in Figure 5 (c), the waveform signals corresponding to different sign language gestures (1, 2, 9, 4, and S) can be wirelessly transmitted to a smartphone and displayed in real time on a self-programming APP.

[0104] The training and recognition process of the neural network unit is as follows: First, the prepared crack strain sensor is installed on the five finger joints of the hand to form a five-channel sensing module, which realizes the real-time recording of the resistance change caused by different sign language movements. We systematically collected a large number of sign language samples by selecting 6 different gestures (4, 1, 2, Y, 9 and S) to obtain a dataset of about 300 samples (each gesture is repeated 50 times) as training samples and test samples. The collected data (training / test samples) are processed using a feature parameter extraction program to extract several feature parameters, which are used as inputs to the neural network classification algorithm for identifying different sign language gestures. The neural network classification algorithm consists of an input layer, three hidden layers and an output layer. The number of neurons in the input layer and the three hidden layers are 10, 1024, 512 and 256 respectively, and the activation function is the tanh function. The number of neurons in the output layer is 6, and the activation function is the ReLU function deployed in the output layer, which outputs the predicted probability of each sign language type. It is worth noting that before the formal sign language decoding, 10 signals of each sign language type were selected to form a 60×10 training sample so that the multi-label neural network could be trained. The training was iterated 1000 times, the learning rate was 0.02, and the initial condition was set to random. The weight parameter file was obtained after the training process and loaded into the multi-label neural network classification algorithm. The pre-trained neural network classification algorithm was used to evaluate the recognition accuracy of the test samples for different sign languages, output the predicted probability of each sign language type, and use the sign language meaning corresponding to the highest predicted probability as the final prediction.

[0105] This application proposes a joint effect of crack opening / closing effect and hydrogel fiber geometric deformation effect under applied strain, which gives the sensor excellent comprehensive sensing performance, including high sensitivity (GF=1188.5), wide detection range (from 0.02% to 80% strain), short response / recovery time (78 / 52ms) and good repeatability (5000 cycles), which is particularly superior to traditional hydrogel-based flexible strain sensors. In addition, the response mutation range and sensitivity of the sensor can be customized by adjusting the crack length in the hydrogel fiber ring and the ratio of long / short fiber segment length, making the sensing performance programmable.

[0106] The present application adopts the porous Ecoflex elastomer packaging film prepared by the salt template method as the substrate and packaging layer of the device, which gives the crack strain sensor good air permeability, reduces the user's feeling of moisture, and thus improves the user's comfort.

[0107] This application is based on the ability of the prepared sensor to accurately detect various strains in various human body movements. By organically integrating multi-channel hydrogel crack strain sensors, wireless circuit modules and artificial intelligence technology, we have successfully developed a wearable sign language recognition system equipped with wireless data transmission function for real-time tracking and accurate recognition of gesture movements.

[0108] Compared with the reported hydrogel-based flexible strain sensors, the present invention organically combines crack engineering with the geometric deformation effect of hydrogel, introduces adjustable local cracks in the closed-loop connected hydrogel fibers wrapped by porous elastomer membranes, and proposes a high-performance and breathable hydrogel crack strain sensor. The overall sensing performance of the crack sensor is better than that of the hydrogel-based strain sensors reported in other literature, and because the device has good breathability, it improves the comfort of the user. At the same time, under the synergistic effect of artificial intelligence algorithms and hardware circuits, the wearable sign language recognition system based on multi-channel crack strain sensors can achieve real-time tracking and accurate decoding and recognition of gesture movements.

[0109] It is understandable that the microstructure design of hydrogels through preparation techniques such as laser cutting and perforation, spinning technology, etc., can make flexible strain sensors have certain air permeability and improve sensing sensitivity, but these technologies have limited improvement in the overall performance of the device and the preparation cost is relatively high and the preparation process is complicated. In addition, non-hydrogel substrate material systems such as elastomers and textile materials are used to prepare high-performance strain sensors, but many non-hydrogel materials do not match the modulus of biological tissues, and their biocompatibility needs to be improved, which may lead to mechanical mismatch and lack of comfort when in contact with biological tissues such as skin. Hydrogels have similar mechanical properties to biological tissues and can be better compatible with skin or other biological tissues, making hydrogel-based strain sensors more comfortable when in contact with the skin and more suitable for long-term wear.

[0110] In some selectable embodiments, the function / operation mentioned in the block diagram may not occur in the order mentioned in the operation diagram. For example, depending on the function / operation involved, the two boxes shown in succession can actually be executed substantially simultaneously or the boxes can sometimes be executed in reverse order. In addition, the embodiment presented and described in the flow chart of the present invention is provided by way of example, for the purpose of providing a more comprehensive understanding of technology. The disclosed method is not limited to the operation and logic flow presented herein. Selectable embodiments are expected, wherein the order of various operations is changed and the sub-operation of a part for which is described as a larger operation is performed independently.

[0111] In addition, although the present invention is described in the context of functional modules, it should be understood that, unless otherwise specified, one or more of the functions and / or features can be integrated into a single physical device and / or software module, or one or more functions and / or features can be implemented in a separate physical device or software module. It is also understood that a detailed discussion of the actual implementation of each module is unnecessary for understanding the present invention. More specifically, in view of the properties, functions and internal relationships of the various functional modules in the device disclosed herein, the actual implementation of the module will be understood within the conventional skills of the engineer. Therefore, those skilled in the art can implement the present invention set forth in the claims without excessive experimentation using ordinary techniques. It is also understood that the specific concepts disclosed are merely illustrative and are not intended to limit the scope of the present invention, which is determined by the full scope of the appended claims and their equivalents.

[0112] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several programs to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc., which can store program codes.

[0113] The logic and / or steps represented in the flowchart or otherwise described herein, for example, may be considered as an ordered list of executable programs for implementing the logical functions, and may be embodied in any computer-readable medium for use by a program execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch and execute a program from a program execution system, device or apparatus), or in conjunction with such program execution systems, devices or apparatuses. For purposes of this specification, a "computer-readable medium" may be any device that can contain, store, communicate, propagate or transmit a program for use by a program execution system, device or apparatus, or in conjunction with such program execution systems, devices or apparatuses.

[0114] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk case (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering or, if necessary, processing in another suitable manner, and then stored in a computer memory.

[0115] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, a plurality of steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable program execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0116] In the above description of this specification, the description with reference to the terms "one embodiment / example", "another embodiment / example" or "certain embodiments / examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0117] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the claims and their equivalents.

[0118] The above is a specific description of the preferred implementation of the present invention, but the present invention is not limited to the embodiments. Those skilled in the art may make various equivalent modifications or substitutions without violating the spirit of the present invention. These equivalent modifications or substitutions are all included in the scope defined by the claims of the present invention.

Claims

1. A method for preparing a hydrogel fiber crack strain sensor, characterized in that: The following steps are involved: preparing hydrogel fibers; Binding silver wires as electrodes to the hydrogel fibers and placing them on a substrate; Connecting two ends of the hydrogel fiber through a conductive paste so that the hydrogel fiber forms a closed loop; and covering the encapsulation layer; A crack is created between two electrodes, and a conductive paste is applied to the cross section of the hydrogel fiber at the crack to obtain a hydrogel fiber crack strain sensor.

2. The method for preparing the hydrogel fiber crack strain sensor according to claim 1, characterized in that: The substrate comprises an elastomer, and the substrate is prepared by the following steps: The elastomer is obtained by a salt template method; Alternatively, the elastomer is obtained by spin coating.

3. The method for preparing the hydrogel fiber crack strain sensor according to claim 2, characterized in that: The substrate comprises an Ecoflex substrate, and the Ecoflex substrate is prepared by the following steps: mixing NaCl and Ecoflex prepolymer to obtain a NaCl / Ecoflex composite material; coating the NaCl / Ecoflex composite material on a metal substrate at a preset rotation speed to obtain a NaCl / Ecoflex composite film; The NaCl / Ecoflex composite film is immersed, ultrasonicated and dried to obtain an Ecoflex substrate.

4. The method for preparing the hydrogel fiber crack strain sensor according to claim 1, characterized in that: The method for preparing the hydrogel fiber comprises: The hydrogel precursor solution is heated to obtain first fibers; and the first fibers are soaked to obtain hydrogel fibers.

5. The method for preparing the hydrogel fiber crack strain sensor according to claim 4, characterized in that: The step of heating the hydrogel precursor solution to obtain the first fiber comprises: The capillary glass tube containing the hydrogel precursor solution is subjected to a heating treatment to form a first network, and the fiber containing the first network is taken out from the capillary glass tube to obtain a first fiber; The step of soaking the first fiber to obtain the hydrogel fiber comprises: Soaking the first fiber in a first solution of a preset concentration for a preset time to form a second network to obtain a hydrogel fiber; The hydrogel fiber is immersed in a second solution with a preset mass proportion for a preset time period to optimize the hydrogel fiber.

6. The method for preparing a hydrogel fiber crack strain sensor according to claim 1, characterized in that: The method further comprises: The ratio of long to short fiber segments is fixed, and the strain threshold causing the sudden change in sensor response is reduced by increasing the length of the crack; Alternatively, the length of the crack is fixed, and the amplitude and sensitivity of the sudden change in sensor response are improved by increasing the ratio of long to short fiber segments; the ratio of the long to short fiber segments is the ratio of the first segment to the second segment, and the hydrogel fiber closed loop is divided into the first segment and the second segment by the electrode, and the length of the first segment is greater than that of the second segment.

7. The method for preparing a hydrogel fiber crack strain sensor according to claim 1, characterized in that: Before the step of covering the encapsulation layer, the method further comprises: Ecoflex prepolymer was poured onto the substrate to fix the hydrogel fibers.

8. A sign language recognition system, characterized in that: include: Hardware modules, software modules and wearable modules; The wearable module is connected to the hardware module, and the hardware module is connected to the software module; The wearable module includes: a sensor obtained by the preparation method of the hydrogel fiber crack strain sensor according to any one of claims 1 to 7; the sensor is used to: be placed at the knuckle of the target finger to be identified, and collect a first signal related to the sign language of the target finger; the hardware module is used to process the first signal; and the software module is used to determine a prediction result of the sign language meaning based on the first signal.

9. The sign language recognition system according to claim 8, characterized in that: The hardware module includes: A digital synthesizer and a voltage follower; the second signal generated by the digital synthesizer is applied to the sensor through the voltage follower; A transimpedance amplifier, a microprocessor, and a wireless communication unit; the first signal is transmitted to the microprocessor through the transimpedance amplifier; the microprocessor processes the first signal and transmits it to the software module through the wireless communication unit.

10. The sign language recognition system according to claim 8, characterized in that: The software module includes a neural network unit, which is used to: obtain a first signal sample; input the first signal sample into the neural network unit for prediction to obtain a sample prediction result; update the parameters of the neural network unit according to the relationship between the sample prediction result and the sample true result to obtain the trained neural network unit, and perform sign language recognition through the trained neural network unit.

Images