Multi-parameter wearable intelligent optical fiber sensing system

Abstract

The invention discloses a multi-parameter wearable intelligent optical fiber sensing system, and the system comprises a fluorescent sensing patch which is used for detecting a target object through a fluorescent reaction, and comprises a biosensor which is used for reacting with the target object and generating a fluorescent signal; the data glove is used for monitoring the bending state of finger joints in real time and achieving gesture recognition and comprises a glove body and a plurality of optical fiber bending sensors attached to the joint portions of the glove body, and the optical fiber bending sensors reflect the finger bending degree in real time through light intensity changes and output multi-dimensional light intensity signals. And the wearable optical fiber detector is connected with the fluorescent sensing patch and the data glove and is used for collecting and analyzing the fluorescent signal and the bending signal. The multi-parameter wearable optical fiber intelligent sensing system has multi-channel and full-automatic intelligent detection and analysis capability, can provide real-time monitoring and analysis services of multiple physical and chemical parameters such as fluorescence, gesture bending and environment temperature and humidity at the same time, and provides powerful technical support for application in various fields.

Description

Technical Field

[0001] The present invention relates to the field of human-computer interaction technology, and in particular to a multi-parameter wearable intelligent optical fiber sensing system. Background Art

[0002] Gestures are a common form of communication between humans. For example, sign language is widely used by people with speech impairments. Automotive gesture recognition can be used for human-computer interaction (HCI), providing a more natural and efficient solution for information exchange between users and devices compared to traditional methods such as a mouse and / or keyboard. Under the influence of the global pandemic, it plays a vital role in practical applications. Wearable devices that enable contactless human-computer interaction (HCI) and pathogen detection have become increasingly important.

[0003] Currently, gesture recognition primarily involves vision-based or sensor-based approaches. Vision-based gesture recognition is susceptible to limitations in both equipment and environmental conditions. For example, if the camera is too bright or too dim during a photo capture, the quality of the captured image will be affected, thereby compromising recognition accuracy. However, sensor-based recognition technology is unaffected by ambient light and background color. It offers stable data collection and simple signal processing, overcoming the aforementioned shortcomings of vision-based recognition.

[0004] Smart gloves equipped with flexible sensors are a viable solution for hand gesture and sign language recognition due to their real-time responsiveness and portability. Furthermore, applications have been demonstrated, including robot-assisted surgery and rehabilitation, human-robot interaction, and gesture recognition for communication and entertainment. Different soft sensor technologies have been used in the design of smart gloves, including resistive and capacitive sensors such as bend sensors for detecting joint bending and / or piezoelectric sensors for detecting finger segment motion, as well as electromyography (EMG).

[0005] However, gesture recognition systems based on electronic sensors are susceptible to electromagnetic interference, and gesture recognition systems based on chemical sensors are complex and expensive. Compared with electronic sensors, optical sensors have the advantages of easy large-scale reuse and resistance to electromagnetic interference. Various methods have been used to achieve bending sensing, among which optical fiber sensors are easy to achieve continuous multi-point curvature measurement. Through extensive research on optical fiber sensors, researchers have proposed a series of design methods that can realize sensing and detection systems for bending, bending curvature, and gestures and movements based on this. However, existing wearable optical fiber sensing detection devices still lack multifunctional sensing functions, cannot achieve multifunctional detection, cannot realize repeated continuous gesture recognition, detect few gestures, and need to improve detection speed and accuracy. Summary of the Invention

[0006] In view of the above problems, the present invention provides a multi-parameter wearable intelligent fiber optic sensing system, which realizes efficient and accurate biosensor detection, pathogenic microorganism detection, etc., realizes human-computer interaction through data gloves and medical equipment, and provides convenient and accurate technical support for on-site detection and human-computer interaction between medical personnel and related equipment.

[0007] In order to achieve the above object, the technical solution adopted by the present invention is as follows:

[0008] The present invention provides a multi-parameter wearable intelligent optical fiber sensing system, comprising:

[0009] A fluorescent sensing patch for detecting a target through a fluorescent reaction, including a biosensor for reacting with the target and generating a fluorescent signal;

[0010] A data glove, used to monitor the bending state of finger joints in real time and realize gesture recognition, includes a glove and multiple fiber optic bending sensors attached to the joints of the glove. The fiber optic bending sensors reflect the degree of finger bending in real time through changes in light intensity and output multi-dimensional light intensity signals;

[0011] A wearable optical fiber detector is connected to the fluorescent sensing patch and the data glove, and is used to collect and analyze fluorescent signals and bending signals.

[0012] In one embodiment of the present invention, the fluorescent sensor patch further comprises:

[0013] an excitation optical fiber, for transmitting excitation light to the biosensor;

[0014] a receiving optical fiber, used to capture the fluorescent signal generated by the biosensor;

[0015] The patch substrate is made of flexible material and integrates the biosensor and optical fiber.

[0016] In one embodiment of the present invention, the biosensor is made by freeze-dried detection reagent on the surface of filter paper. When the target enters the reaction tank of the biosensor, ultraviolet light excitation generates a fluorescent signal. The fluorescent signal is transmitted through the receiving optical fiber and captured by a micro camera and converted into a fluorescent image. The change in image intensity is linearly correlated with the concentration of the target.

[0017] In one embodiment of the present invention, the optical fiber bending sensor is adhered to the metacarpophalangeal joint and the proximal interphalangeal joint of the glove through a transparent silicone rubber material.

[0018] In one embodiment of the present invention, a fiber optic bending sensor is also provided on the wrist of the glove to improve the recognition accuracy of repeated gestures.

[0019] In one embodiment of the present invention, the gesture recognition adopts the following algorithm process:

[0020] Segmenting the collected optical fiber image to obtain the output light intensity signal of each optical fiber bending sensor;

[0021] The forward difference method is used to extract the real-time bending characteristics of the optical fiber bending sensor;

[0022] Eliminate data noise through threshold denoising technology;

[0023] Detect bending features to identify gesture switching points, and segment the start and end times and gesture switching points as key points;

[0024] By retaining the output light intensity signal at the intermediate moment of adjacent key points as the key information of the gesture pattern, compression is performed;

[0025] The compressed gesture pattern is transmitted to the gesture recognition module based on the radial basis function network model for gesture recognition.

[0026] In one embodiment of the present invention, the gesture recognition specifically includes:

[0027] When the system is working, the wearable fiber optic sensing system extracts the R channel pixel intensity from the image acquired by the small camera as the output signal Ic of the fiber optic bending sensor; the initial output signal I0 when the fiber optic bending sensor is not bent; in order to quantize the output signals of the fiber optic bending sensor array to the same scale, the output signals are normalized using the following equation:

[0028] I=k·θ+I b

[0029] Where θ is the bending angle of the optical fiber bending sensor, I = 1–Ic / I0, k = -k′ / I0, k′ is the linear conversion coefficient from the output power of the optical fiber bending sensor to the output light intensity signal, and Ib is the deviation of the linear fitting result;

[0030] I(n) represents the total light intensity signal obtained by sampling the fiber bending sensor array for the nth time, I j (n) represents the light intensity signal obtained by sampling the j-th fiber bend sensor in the fiber bend sensor array for the nth time, I j(n+1) represents the light intensity signal obtained by sampling the jth fiber bend sensor in the fiber bend sensor array for the n+1th time; the gesture switching characteristic curve ΔI(n) is used to represent the degree of change in finger joint movement. If the total light intensity signal change ΔI(n) of the fiber bend sensor array is greater than the threshold, it is considered that the gesture has begun to change; if ΔI(n) is less than the threshold, it is considered that the gesture remains unchanged; during the gesture change process, the finger joint angle vector changes with the hand shape, and ΔI(n) changes accordingly; use formula (2) to process each gesture pattern;

[0031]

[0032] Where N is the total number of fiber optic bend sensors in the data glove. A forward difference is performed on the multidimensional light intensity signal output by the data glove to extract the real-time bending characteristics of the fiber optic bend sensors. An absolute value operation is performed to eliminate the sign information in the data, making the characteristic waveform amplitude values ​​all positive, simplifying the feature extraction and analysis process. An aggregation function is used to achieve dimensionality reduction of the multidimensional bending data through linear addition.

[0033] ΔI(n) is processed by threshold denoising:

[0034]

[0035] Where λ is the light intensity threshold set to determine whether the gesture changes. The gesture switching point is found by maximum value detection. The start and end times and the gesture switching point are used as the key points for continuous gesture segmentation. The data between each two key points is the same oversampled data. The area near each gesture switching point represents the gesture switching process. ΔI h (n) unchanged indicates a stable gesture maintenance process;

[0036] The light intensity vector output by the optical fiber bending sensor at the intermediate moment between adjacent key points is used as the key information of the gesture pattern, and data compression is performed based on this;

[0037] The compressed gesture pattern is transmitted to the gesture recognition module based on the radial basis function network model, and gesture recognition is performed through the trained model.

[0038] In one embodiment of the present invention, the optical fiber probe of the fluorescent sensing patch is chemically etched to remove part of the cladding to enhance the intensity of the excitation light. The specific etching method is: immersing the optical fiber in a 1:1 mixed solution of n-hexane and acetone for 5 seconds, and then cleaning and drying to fix the etching effect.

[0039] In one embodiment of the present invention, a wireless communication module is further included, which is connected to the wearable fiber optic detector and is used to transmit the detection data of the wearable fiber optic detector to a remote server and / or personal device in real time.

[0040] In one embodiment of the present invention, the target is a pathogenic microorganism.

[0041] The beneficial effects achieved by the present invention are:

[0042] The present invention provides a multi-parameter wearable intelligent fiber optic sensing system that achieves highly sensitive detection of pathogenic microorganisms in the air through integrated fluorescent sensor patches. Data gloves enable wireless human-computer interaction between testers and medical equipment. The wearable system design meets the needs of rapid on-site testing. This system not only enables fluorescence detection of pathogenic microorganisms in the air but also innovatively proposes gesture segmentation, compression, and recognition algorithms, enabling efficient recognition of real-time dynamic gestures. Furthermore, the system can also enable remote visualization and sharing of data. In summary, this multi-parameter wearable fiber optic intelligent sensing system possesses multi-channel, fully automatic intelligent detection and analysis capabilities, capable of simultaneously providing real-time monitoring and analysis of multiple physical and chemical parameters, including fluorescence, bending, and ambient temperature and humidity, providing strong technical support for applications in various fields. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the embodiments of the present invention. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the drawings without paying any creative work.

[0044] Figure 1 (a) is the schematic diagram of the multi-parameter wearable intelligent fiber optic sensing system; (b) is a schematic diagram of the wearable fluorescent sensor patch structure; (c) is a schematic diagram of the wearable fiber optic detector structure after etching.

[0045] Figure 2 This is the workflow diagram of the real-time gesture recognition system.

[0046] Figure 3 (a) is a schematic diagram of plastic optical fiber before and after etching; (b) is a schematic diagram of the wearable optical fiber fluorescence sensing system.

[0047] Figure 4 This is a real picture of the data glove.

[0048] Figure 5 (a) is a physical picture of the multi-parameter sensing system; (b) is a wearable picture of the system; and (c) is a mobile phone remote data display picture.

[0049] Figure 6 Schematic diagram of the visualization interface of the multi-parameter wearable intelligent fiber optic sensing system.

[0050] Figure 7 (a) is a schematic diagram of the optical fiber fluorescence image; (b) is a schematic diagram of the camera parameter optimization.

[0051] Figure 8 These are performance test diagrams of the wearable fluorescent sensor patch, where (a) is a comparison of the test results of the device and the commercial device using an enzyme-linked reader; (b) is a stability test diagram.

[0052] Figure 9 This is a dynamic test diagram of pathogenic bacteria.

[0053] Figure 10 (a) is the gesture picture; (b) is the gesture data.

[0054] Figure 11 Figure 2 is a diagram of the static gesture recognition results based on RBFNN, where (a) is a diagram of the confusion matrix of the training set output results; (b) is a diagram of the confusion matrix of the validation set output results; and (c) is a diagram of the gesture recognition accuracy.

[0055] Figure 12 Schematic diagram of the confusion matrix with the maximum eigenvalue of the gesture switching point.

[0056] Figure 13 Figure 3 is a diagram of the real-time gesture recognition data processing process, where (a) is the data glove sampling data diagram that changes with gesture; (b) is the data segmentation signal diagram; (c) is the extracted gesture feature data diagram; and (d) is the gesture recognition diagram based on RBFNN.

[0057] Figure 14 Figure 3 is a diagram of the data processing process for real-time gesture recognition with repeated features, where (a) is the data glove sampling data diagram that changes with gestures; (b) is the data segmentation signal diagram; (c) is the extracted gesture feature data diagram; and (d) is the gesture recognition diagram based on RBFNN. DETAILED DESCRIPTION

[0058] The following embodiments of the technical solution of the present invention will be described in detail with reference to the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present invention and are therefore only examples and are not intended to limit the scope of protection of the present invention.

[0059] Unless otherwise defined, all technical and scientific terms used herein have the same meanings as commonly understood by those skilled in the art to which the present invention belongs; the terms used herein are only for the purpose of describing specific embodiments and are not intended to limit the present invention; the terms "including" and "having" and any variations thereof in the specification and claims of the present invention and the above-mentioned drawings are intended to cover non-exclusive inclusions.

[0060] In the description of the embodiments of the present invention, technical terms such as "first" and "second" are used solely to distinguish between different objects and should not be understood to indicate or imply relative importance, or to implicitly specify the quantity, specific order, or primary and secondary relationship of the technical features indicated. In the description of the embodiments of the present invention, "plurality" means more than two, unless otherwise specifically defined.

[0061] References herein to "embodiments" mean that a particular feature, structure, or characteristic described in connection with the embodiments may be included in at least some embodiments of the present invention. The appearance of this phrase in various places in the specification does not necessarily refer to the same embodiment, nor does it constitute a separate or alternative embodiment that is mutually exclusive of other embodiments. It is understood, both explicitly and implicitly, by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0062] For example, in the medical field, when medical staff wearing protective clothing equipped with fluorescence detection capabilities encounter potentially pathogenic microorganisms, the fluorescence sensor quickly responds, identifying and sounding an alarm through unique fluorescent markers, providing immediate safety warnings. This technology not only improves medical safety but also provides doctors with an intuitive and rapid method for identifying biological risk factors. The application of gesture recognition technology in the interaction between doctors and medical devices further improves work efficiency. Through precise gesture recognition, doctors can easily operate and query the device without being distracted by complex operating interfaces. Especially in emergency situations, gesture recognition technology becomes a critical bridge between doctors and equipment, ensuring rapid response and accurate execution.

[0063] Intelligent fiber optic sensing systems offer broad application prospects thanks to their exceptional multi-signal detection capabilities, including diverse detection capabilities such as fluorescence and bending, as well as their highly integrated and automated recognition capabilities. To further broaden its application, the present invention provides a multi-parameter wearable intelligent fiber optic sensing system that integrates wearable fluorescence detection technology with gesture recognition technology. This integrated multi-parameter wearable intelligent fiber optic sensing system not only accurately identifies biological risk factors but also provides precise analysis of continuous gestures. It is expected to promote the intelligent development of wearable devices, biomedicine, environmental monitoring, and other fields, providing people with a safer, more convenient, and more efficient life experience.

[0064] like Figure 1As shown in (a) of the figure, a multi-parameter wearable intelligent fiber optic sensing system comprises a data glove, a fluorescent sensor patch, and a wearable fiber optic detector. The system can capture the light intensity information output by the data glove in real time, thereby achieving high-precision recognition of gestures. Furthermore, by monitoring the fluorescence response of the biosensor under specific excitation light, the system can effectively detect the presence of pathogens in exposed environments. Using the fiber optic light intensity images captured by the wearable fiber optic detector, the system can provide comprehensive intelligent detection and analysis services for multiple channels and multiple physical and chemical parameters. Notably, all analysis results can be transmitted in real time to mobile devices or personal servers via wireless communication technology, greatly enhancing the convenience and real-time nature of data processing. This system not only detects pathogenic microorganisms such as pathogens and viruses in the air, but also enables dynamic, efficient, and intelligent recognition of gestures, thereby realizing wireless human-computer interaction. The system also enables remote visualization and sharing of data.

[0065] like Figure 1 As shown in (b), the fluorescence sensing patch includes a biosensor, an excitation fiber, a receiving fiber, and a patch substrate. The biosensor reacts with the target and generates a fluorescent signal. The excitation fiber transmits excitation light to the biosensor, and the receiving fiber captures the fluorescent signal generated by the biosensor in response to the target. The patch substrate is made of a flexible material and integrates the biosensor and optical fibers. The biosensor reacts with the pathogenic microorganisms being detected, generating a fluorescent signal through various principles, which is then captured by the optical fibers. The fluorescence sensing patch is powered by an external light source. The fluorescent signal captured by the optical fibers is focused by filters, collimating lenses, and other devices, and a camera captures the fluorescent image for processing and analysis. The biosensor is made by freeze-drying a specific detection reagent onto the surface of filter paper. When a small droplet of pathogenic bacteria in the environment passes through the micropores of the patch and enters the biosensor's reaction chamber, ultraviolet light is emitted through the excitation fiber. The target reacts with the biosensor to generate a fluorescent signal, which is then transmitted through the optical fiber array and ultimately captured by the micro-camera of the fiber optic fluorometer and converted into a fluorescent image. Changes in image intensity are directly correlated with the concentration of the target, providing a reliable basis for pathogen detection.

[0066] like Figure 1As shown in (c) of the figure, the design of the data glove is based on the close integration of fiber optic bend sensors with the glove body. The data glove includes: multiple fiber optic bend sensors, which are bonded to the metacarpophalangeal joints and proximal interphalangeal joints of the glove using transparent silicone rubber material; a fiber optic bend sensor added to the wrist to improve the recognition accuracy of repeated gestures; the sensor reflects the degree of finger bending in real time through changes in light intensity and outputs a multi-dimensional light intensity signal. Soft transparent silicone rubber material is used to precisely bond the fiber optic bend sensors to the joints of the glove, ensuring that the sensors can respond in real time to bending changes caused by flexion or extension of the finger joints. Each sensor independently monitors the bending state of the metacarpophalangeal joints or proximal interphalangeal joints of the five fingers and reflects the degree of finger bending through changes in light intensity. In addition, the additional fiber optic bend sensor added to the wrist further improves the accuracy of repeated gesture recognition.

[0067] like Figure 2 As shown in Figure 2, the workflow of the sign language recognition system primarily consists of three core steps: gesture segmentation, gesture sequence compression, and gesture recognition. During the data preprocessing phase, the system first finely segments the acquired fiber image to obtain the output light intensity signal of each fiber bend sensor. Subsequently, the system uses the forward difference method to extract the real-time bending characteristics of the fiber bend sensor and employs threshold denoising to eliminate potential data noise.

[0068] To accurately segment continuous gesture data, the system detects bending features to identify gesture switching points and segments the start and end moments and gesture switching points as key points. To improve the efficiency and accuracy of gesture recognition, the system introduces an efficient data compression method that retains the output light intensity signal at the intermediate moments between adjacent key points as key information for compression of the gesture pattern. Finally, the compressed gesture pattern is transmitted to the RBFNN gesture recognition module for accurate recognition. The specific process is as follows:

[0069] When the system is working, the wearable fiber optic sensing system extracts the R channel pixel intensity from the image acquired by the small camera as the output signal Ic of the fiber optic bending sensor; the initial output signal I0 when the fiber optic bending sensor is not bent; in order to quantize the output signals of the fiber optic bending sensor array to the same scale, the output signals are normalized using the following equation:

[0070] I=k·θ+I b

[0071] Where θ is the bending angle of the optical fiber bending sensor, I = 1–Ic / I0, k = -k′ / I0, k′ is the linear conversion coefficient from the output power of the optical fiber bending sensor to the output light intensity signal, and Ib is the deviation of the linear fitting result;

[0072] I(n) represents the total light intensity signal obtained by sampling the fiber bending sensor array for the nth time, I j (n) represents the light intensity signal obtained by sampling the j-th fiber bend sensor in the fiber bend sensor array for the nth time, I j (n+1) represents the light intensity signal obtained by sampling the jth fiber bend sensor in the fiber bend sensor array for the n+1th time; the gesture switching characteristic curve ΔI(n) is used to represent the degree of change in finger joint movement. If the total light intensity signal change ΔI(n) of the fiber bend sensor array is greater than the threshold, it is considered that the gesture has begun to change; if ΔI(n) is less than the threshold, it is considered that the gesture remains unchanged; during the gesture change process, the finger joint angle vector changes with the hand shape, and ΔI(n) changes accordingly; use formula (2) to process each gesture pattern;

[0073]

[0074] Where N is the total number of fiber optic bend sensors in the data glove. A forward difference is performed on the multidimensional light intensity signal output by the data glove to extract the real-time bending characteristics of the fiber optic bend sensors. An absolute value operation is performed to eliminate the sign information in the data, making the characteristic waveform amplitude values ​​all positive, simplifying the feature extraction and analysis process. An aggregation function is used to achieve dimensionality reduction of the multidimensional bending data through linear addition.

[0075] ΔI(n) is processed by threshold denoising:

[0076]

[0077] Where λ is the light intensity threshold set to determine whether the gesture changes. The gesture switching point is found by maximum value detection. The start and end times and the gesture switching point are used as the key points for continuous gesture segmentation. The data between each two key points is the same oversampled data. The area near each gesture switching point represents the gesture switching process. ΔI h (n) unchanged indicates a stable gesture maintenance process.

[0078] The light intensity vector output by the fiber optic bending sensor at the intermediate moments between adjacent key points is used as the key information of the gesture pattern, and data compression is performed based on this. Finally, the compressed gesture pattern is transmitted to the RBFNN gesture recognition module, and the trained model is used to accurately recognize the gesture.

[0079] To enhance the intensity of the excitation light and improve detection sensitivity and accuracy, the fiber probe is etched to remove part of the cladding. Chemical etching of plastic optical fibers offers advantages such as simplicity and low cost. A 1:1 solution of n-hexane and acetone can be used to etch the plastic optical fiber sensor probe.

[0080] The specific operation is as follows: immerse the part of the plastic optical fiber that needs to be etched in this solution, ensuring that it is in full contact with the solution so that the etching effect is uniform. Then, soak the plastic optical fiber in the solution for 5 seconds to obtain the desired etching depth. After the immersion is completed, take out the plastic optical fiber immediately and rinse its surface thoroughly with clean water to remove any residual chemical solvents or etching agents. During the rinsing process, make sure to use sufficient water flow and time to avoid any chemical residues. Finally, place the cleaned plastic optical fiber in an oven at a temperature of 55°C for 2 hours. This process helps to ensure that the plastic optical fiber is completely dry and fixes the etching effect, thereby completing the entire etching process. The side images of the plastic optical fiber before and after etching are as follows Figure 3 As shown in (a) in .

[0081] like Figure 3 As shown in (b), the wearable fiber optic fluorescence sensing system includes a plastic optical fiber, a wearable fluorescence sensing patch, and a wearable fiber optic light intensity demodulator. The wearable fluorescence sensing patch is freeze-dried from a biosensor onto filter paper in a reaction tank. Each plastic optical fiber sensor probe is made of four excitation optical fibers and three receiving optical fibers arranged in an alternating pattern and placed in the reaction tank. A 0.5mm diameter optical fiber is used as an excitation fiber to transmit the LED ultraviolet light signal and act on the excitation of the biosensor fluorescence signal. A 0.25mm diameter optical fiber is used as a receiving fiber to transmit the fluorescence signal to the micro camera of the fluorimeter to realize fluorescence signal acquisition.

[0082] Fiber optic bending sensors are attached to knitted gloves to make data gloves for continuous hand motion detection. The corresponding relationship between sensor numbers and finger joints is as follows: Figure 4 shown.

[0083] like Figure 5 As shown in the figure, by integrating a fluorescent patch with a data glove into a wearable fiber-optic light intensity demodulator, the fluorescent sensing patch can be attached to the surface of clothing and used to detect pathogens in the ambient air. This realizes a multi-parameter wearable fiber-optic intelligent sensing system for simultaneous detection of fluorescence and bending signals.

[0084] like Figure 6 As shown in the figure, a web application (wireless communication module) is designed to achieve remote visualization of data from a multi-parameter wearable fiber-optic intelligent sensing system. When the system is working, the data detected by the system can be transmitted to a remote server or personal device via the wireless communication module, and the detection results can be displayed in real time.

[0085] Exposure time t b Select 1s, 2s, 6s, and 10s, and Iso is selected from [100, 200, 300, 400, 500, 600, 700, 800, 900, 1000]. The fiber images collected under the corresponding parameters are as follows: Figure 7As shown in (a), the average pixel intensity extracted is Figure 7 As shown in (b). H is the high fluorescence intensity reagent, and L is the low intensity reagent. The bar graph shows the average pixel difference between the two reagents under the corresponding camera parameters. It can be seen that iso=500, t b =10s, the average pixel intensity is the largest, and the sensitivity of the fiber optic fluorometer is the highest at this time. b = 10s was selected as the camera parameter setting for subsequent experiments.

[0086] like Figure 8 As shown in the figure, the performance of the wearable fluorescent sensing patch was tested. In order to evaluate the fluorescence detection performance of the system, FITC was diluted to 6 different concentrations of 10pM, 100pM, 1nM, 10nM, 100nM, and 1μM for testing and compared with the commercial device microplate reader at excitation and emission wavelengths of 485nm and 525nm respectively. Figure 8 As shown in (a), the horizontal axis represents the fluorescence test results of the microplate reader, and the vertical axis represents the fluorescence intensity measured by the device. The test results of the device and the microplate reader (SynergyH1) show a good linear relationship, showing excellent performance similar to that of the microplate reader. The fluorescence signals of high concentration (1μM) and low concentration (1nM) were measured at intervals of 15 seconds for 1 hour. The change of fluorescence intensity over time is shown in the figure. Figure 8 As shown in (b), it exhibits good stability.

[0087] like Figure 9 As shown in the figure, based on the designed wearable optical fiber fluorescence sensing system, 10 4 The CFU / ml of Salmonella was detected in real time. This verified the dynamic testing function of the designed detection system. The fluorescence intensity gradually increased with the increase of reaction time, and the reaction rates at different concentrations were different, showing strong distinguishability.

[0088] The present invention uses the data gloves made by the system to collect data. Figure 10 As shown in (a) in the figure, 26 English letter gestures from "A" to "Z" and a static gesture "`" were selected as research objects. To ensure the rigor and integrity of the data, each gesture was repeated 5 times, and 60 sample data were collected in each repetition, eventually forming a detailed data set containing 8100 samples. The collected data are as follows Figure 10 As shown in (b) in the figure, the horizontal axis represents different gestures, the vertical axis represents the output response of the multi-channel sensor corresponding to each gesture, and the depth of the color represents the intensity of the light output by the sensor.

[0089] Of the 8100 gesture data samples collected, there are a total of 5670 training set samples and 2430 test set samples. The confusion matrix of the recognition results of all 27 gesture categories in the training set is as follows Figure 11 As shown in (a), the horizontal axis represents the real gesture label, and the vertical axis represents the gesture result output by the model. The blue background data in the figure represents the number of samples corresponding to each gesture that are correctly recognized, and the gray background data represents the number of samples that are misclassified as other gestures. For example, if the real gesture is "A", all 211 training samples are correctly recognized. Among the 200 training samples of gesture "B", 196 samples are correctly recognized, 1 sample is misclassified as "A", and 3 data are misclassified as "C". Similarly, Figure 11 (b) is the confusion matrix of the gesture recognition results of the validation set. The training and validation recognition accuracy of each gesture is as follows Figure 11 As shown in (c) in the training set, the recognition accuracy of eight gestures, including "A," "E," "N," "O," "Q," "R," "V," and "W," was 100%. The recognition accuracy of seven gestures, including "J," "M," "P," "S," "T," "U," and "`," was higher than 99%. The recognition accuracy of eight gestures, including "B," "D," "G," "I," "K," "L," "X," and "Y," was between 96.4% and 99%. The recognition accuracy of four gestures, including "C," "F," "H," and "Z," was between 93.1% and 95%. Among them, the recognition accuracy of "Z" was the lowest, at 93.1%. For the validation set, the recognition accuracy of 13 gestures including "A", "B", "E", "MR", "T", "V", "W", and "Y" is 100%, the recognition accuracy of 9 gestures including "D", "FH", "JL", "S", and "U" is between 95.6% and 99%, the recognition accuracy of 3 gestures including "C", "I", and "X" is between 91.7% and 93.3%, and the recognition accuracy of gesture "Z" is the lowest, at 87.8%.

[0090] Overall, the training accuracy for the 27 gestures was 98.1%, and the test accuracy was 97.8%. This means that the RBFNN model correctly recognized approximately 98.13% of the gestures in the training dataset. This relatively high accuracy indicates that the model learned effective features of the data during training and classified these features well. The test accuracy was 97.8%, indicating that the model correctly recognized approximately 97.78% of the gestures in the test dataset. Although the test accuracy was slightly lower than the training accuracy, it was still a high value, indicating that the model maintained good performance even on unseen data. The combined training and test accuracy indicates that the RBFNN model demonstrated good generalization ability in the gesture recognition task, effectively learning from the training data and applying it to the test data.

[0091] In order to select a suitable threshold, the data of switching between all 27 different gestures are processed to obtain 27*26=702 gesture switching feature curves, and 702 maximum values ​​are extracted as the corresponding gesture switching feature values. The confusion matrix of different gesture switching feature values ​​is as follows: Figure 12 As shown in the figure, the horizontal axis represents the initial gesture, and the vertical axis represents the final gesture after the change. The data points corresponding to the horizontal and vertical axes are the gesture switching feature values ​​from the initial gesture to the final gesture. For example, the feature value for switching from gesture "A" to gesture "B" is 0.98, and the feature value for switching to gesture "C" is 0.8. Experimental results show that when switching between the 27 target gestures, the minimum switching feature value between any two different gestures is 0.3. Therefore, a threshold of 0.25 is selected, with a margin of 0.05.

[0092] The gesture segmentation threshold and static gesture recognition model have been determined, and the real-time gesture recognition algorithm proposed in this invention will be verified.

[0093] like Figure 13 As shown in the figure, the real-time gesture recognition process is divided into four modules: gesture data collection, continuous gesture data segmentation, data compression and gesture recognition. First, select “GESTURE RECOGNITION” as the target gesture for data collection. The data is as follows: Figure 13 As shown in (a) in the figure, as the gesture changes in real time, the output data of the ten optical fiber bending sensors on the data glove corresponding to the finger joints also changes. Figure 13 As shown in (b) of , continuous gesture data segmentation is used to segment continuous gestures, and the segmentation takes 8.3ms. Figure 13 As shown in (c) in the figure, the compression of continuous gestures is achieved by adding the data at the middle moment of the gesture switching point. Figure 13 As shown in (d) of Figure 3, the compressed gesture data is fed into the RBFNN model for gesture recognition, which outputs the recognition result within 11ms. Overall, this system achieves the recognition of the real-time continuous gesture "GESTURE RECOGNITION" within 19.3ms.

[0094] In the actual gesture recognition process, repeated gestures are inevitable. In order to realize the recognition of repeated characteristic gestures, a fiber optic bending sensor is added to the wrist of the data glove. When a gesture is required, the wrist is bent. By detecting the output of the wrist sensor, gesture recognition with repeated characteristics is realized. "I am a bookkeeper" is selected as the target gesture for data collection. The data is as follows Figure 14As shown in (a) of Figure 1, as the gesture changes in real time, the output data of the ten fiber optic bending sensors on the data glove corresponding to the finger joints also changes. Channel 11 represents the wrist sensor. Figure 14 As shown in (b) of Figure 1, if the wrist sensor is not added, the switching feature points of repeated gestures cannot be detected. When the output data of the wrist sensor of channel 11 is also added to the processing of the gesture feature curve, the switching feature points of the repeated gestures "O", "K", and "E" are detected. The proposed method is used to segment continuous gestures, and the segmentation time is 8.3ms. Figure 14 As shown in (c) in the figure, the compression of continuous gestures is achieved by adding the data at the middle moment of the gesture switching point. Figure 14 As shown in (d) of Figure 3, the compressed gesture data is fed into the RBFNN model for gesture recognition, which outputs the recognition result within 10 ms. Overall, this system achieves real-time recognition of the continuous gesture “I am a bookkeeper” with repetitive features within 18.3 ms.

[0095] The proposed continuous gesture recognition algorithm based on a flexible fiber optic bend sensor achieves an accuracy rate of 97.8% when recognizing 26 English letters and space gestures, significantly outperforming similar sensors, which are limited to static gesture recognition. This algorithm not only provides new insights into continuous gesture recognition research but also, with its high accuracy and rapid recognition speed, offers strong technical support for practical applications. The wearable system boasts a simple structure and low maintenance costs, while ensuring a pleasant user experience. Furthermore, the flexibility and sensitivity of the flexible fiber optic sensor, combined with forward difference, threshold segmentation, and RBFNN algorithms, effectively handles complex gestures, demonstrating unique advantages in specific applications.

[0096] In summary, the present invention provides a comprehensive multi-parameter wearable fiber optic intelligent sensing system that integrates key components such as fluorescent sensing patches, data gloves based on fiber optic bending sensors, and wearable fiber optic detectors. The system not only realizes the fluorescence detection of pathogenic bacteria in the air, but also innovatively proposes gesture segmentation, compression and recognition algorithms, thereby realizing efficient recognition of real-time gestures. In order to achieve remote visualization and sharing of data, a wireless communication module is also provided. During the operation of the system, all detection data can be transmitted to a remote server or personal device in real time through the wireless communication module, ensuring that users can grasp the latest test results anytime, anywhere. In terms of the performance of the fluorescent sensing patch, after optimizing the camera parameters and testing with FITC, the results show that it has a good linear relationship with the commercial enzyme labeler, R 2=0.9823, with excellent stability. The system also successfully achieved real-time dynamic detection of Salmonella. In terms of gesture recognition, the system not only achieved static recognition of 27 gesture types with an accuracy rate of 97.78%, but also accurately identified real-time continuous gestures with repetitive features in just 18.3ms, demonstrating extremely high real-time performance and accuracy. Overall, this multi-parameter wearable fiber optic intelligent sensing system possesses multi-channel, fully automatic intelligent detection and analysis capabilities, capable of simultaneously providing real-time monitoring and analysis of multiple physical and chemical parameters such as fluorescence, bending, and ambient temperature and humidity, providing strong technical support for applications in various fields.

[0097] While the present invention has been described with reference to preferred embodiments, various modifications may be made and equivalent components may be substituted without departing from the scope of the present invention. In particular, the various technical features described in the various embodiments may be combined in any manner, provided no structural conflicts exist. The present invention is not limited to the specific embodiments disclosed herein, but encompasses all technical solutions within the scope of the claims.

Claims

1. A multi-parameter wearable intelligent optical fiber sensing system, characterized in that: include: A fluorescent sensing patch for detecting a target through a fluorescent reaction, including a biosensor for reacting with the target and generating a fluorescent signal; A data glove, used to monitor the bending state of finger joints in real time and realize gesture recognition, includes a glove and multiple fiber optic bending sensors attached to the joints of the glove. The fiber optic bending sensors reflect the degree of finger bending in real time through changes in light intensity and output multi-dimensional light intensity signals; A wearable optical fiber detector is connected to the fluorescent sensing patch and the data glove, and is used to collect and analyze fluorescent signals and bending signals.

2. A multi-parameter wearable intelligent optical fiber sensing system according to claim 1, characterized in that: The fluorescent sensor patch further includes: an excitation optical fiber, for transmitting excitation light to the biosensor; a receiving optical fiber, used to capture the fluorescent signal generated by the biosensor; The patch substrate is made of flexible material and integrates the biosensor and optical fiber.

3. A multi-parameter wearable intelligent optical fiber sensing system according to claim 2, characterized in that: The biosensor is made by freeze-dried detection reagent on the surface of filter paper. When the target enters the reaction tank of the biosensor, ultraviolet light excites it to produce a fluorescent signal. The fluorescent signal is transmitted through the receiving optical fiber and captured by a micro camera and converted into a fluorescent image. The change in its image intensity is linearly correlated with the concentration of the target.

4. A multi-parameter wearable intelligent optical fiber sensing system according to claim 3, characterized in that: The optical fiber bending sensor is adhered to the metacarpophalangeal joint and the proximal interphalangeal joint of the glove through a transparent silicone rubber material.

5. A multi-parameter wearable intelligent optical fiber sensing system according to claim 4, characterized in that: The wrist of the glove is also provided with an optical fiber bending sensor to improve the recognition accuracy of repeated gestures.

6. The multi-parameter wearable intelligent optical fiber sensing system according to claim 5, characterized in that: The gesture recognition adopts the following algorithm process: Segmenting the collected optical fiber image to obtain the output light intensity signal of each optical fiber bending sensor; The forward difference method is used to extract the real-time bending characteristics of the optical fiber bending sensor; Eliminate data noise through threshold denoising technology; Detect bending features to identify gesture switching points, and segment the start and end times and gesture switching points as key points; By retaining the output light intensity signal at the intermediate moment of adjacent key points as the key information of the gesture pattern, compression is performed; The compressed gesture pattern is transmitted to the gesture recognition module based on the radial basis function network model for gesture recognition.

7. The multi-parameter wearable intelligent optical fiber sensing system according to claim 6, characterized in that: The gesture recognition specifically includes: When the system is working, the wearable fiber optic sensing system extracts the R channel pixel intensity from the image acquired by the small camera as the output signal Ic of the fiber optic bending sensor; the initial output signal I0 when the fiber optic bending sensor is not bent; in order to quantize the output signals of the fiber optic bending sensor array to the same scale, the output signals are normalized using the following equation: I=k·θ+I b Where θ is the bending angle of the optical fiber bending sensor, I = 1–Ic / I0, k = -k′ / I0, k′ is the linear conversion coefficient from the output power of the optical fiber bending sensor to the output light intensity signal, and Ib is the deviation of the linear fitting result; I(n) represents the total light intensity signal obtained by sampling the fiber bending sensor array for the nth time, I j (n) represents the light intensity signal obtained by sampling the j-th fiber bend sensor in the fiber bend sensor array for the nth time, I j (n+1) represents the light intensity signal obtained by sampling the jth fiber bend sensor in the fiber bend sensor array for the n+1th time; the gesture switching characteristic curve ΔI(n) is used to represent the degree of change in finger joint movement. If the total light intensity signal change ΔI(n) of the fiber bend sensor array is greater than the threshold, it is considered that the gesture has begun to change; if ΔI(n) is less than the threshold, it is considered that the gesture remains unchanged; during the gesture change process, the finger joint angle vector changes with the hand shape, and ΔI(n) changes accordingly; use formula (2) to process each gesture pattern; Where N is the total number of fiber optic bend sensors in the data glove. A forward difference is performed on the multidimensional light intensity signal output by the data glove to extract the real-time bending characteristics of the fiber optic bend sensors. An absolute value operation is performed to eliminate the sign information in the data, making the characteristic waveform amplitude values ​​all positive, simplifying the feature extraction and analysis process. An aggregation function is used to achieve dimensionality reduction of the multidimensional bending data through linear addition. ΔI(n) is processed by threshold denoising: Where λ is the light intensity threshold set to determine whether the gesture changes. The gesture switching point is found by maximum value detection. The start and end times and the gesture switching point are used as the key points for continuous gesture segmentation. The data between each two key points is the same oversampled data. The area near each gesture switching point represents the gesture switching process. ΔI h (n) unchanged indicates a stable gesture maintenance process; The light intensity vector output by the optical fiber bending sensor at the intermediate moment between adjacent key points is used as the key information of the gesture pattern, and data compression is performed based on this; The compressed gesture pattern is transmitted to the gesture recognition module based on the radial basis function network model, and gesture recognition is performed through the trained model.

8. The multi-parameter wearable intelligent optical fiber sensing system according to claim 7, characterized in that: The optical fiber probe of the fluorescent sensing patch is chemically etched to remove part of the cladding to enhance the intensity of the excitation light. The specific etching method is: immersing the optical fiber in a 1:1 mixture of n-hexane and acetone for 5 seconds, cleaning and drying, and then fixing the etching effect.

9. The multi-parameter wearable intelligent optical fiber sensing system according to claim 1, characterized in that: It also includes a wireless communication module, which is connected to the wearable fiber optic detector and is used to transmit the detection data of the wearable fiber optic detector to a remote server and / or personal device in real time.

10. A multi-parameter wearable intelligent optical fiber sensing system according to any one of claims 1 to 9, characterized in that: The target object is a pathogenic microorganism.

Images