A gesture recognition method and system based on right-angle dual-plane capacitive sensing array

Through the right-angle dual-plane capacitance sensing array and subspace judgment model, the problems of low accuracy and strong angle dependence of traditional single-plane capacitance sensors are solved, and higher gesture recognition accuracy and range are achieved, adapting to complex gesture recognition.

CN118642600BActive Publication Date: 2024-11-19ANHUI UNIV
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411116948.3
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-08-15
Publication Date
2024-11-19
Estimated Expiration
2044-08-15

AI Technical Summary

Technical Problem

Traditional single-plane capacitive sensors have problems such as reduced accuracy, strong angle dependence and difficulty in identifying complex gestures when sensing gestures, which limits the application scope of gesture recognition.

Method used

A right-angle dual-planar capacitance sensing array and subspace judgment model are used to collect capacitance values ​​through different sensitive configurations, divide the identification subspace and perform gesture recognition, expanding the recognition range and accuracy.

Benefits of technology

It improves the accuracy and recognition range of gesture recognition, and can obtain gesture features more comprehensively and adapt to target gestures of different angles and distances.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118642600B_ABST
    Figure CN118642600B_ABST
Patent Text Reader

Abstract

The present invention belongs to the field of gesture recognition technology, and specifically relates to a gesture recognition method and system based on a right-angle dual-plane capacitive sensor array, including a sensor module for collecting target gestures through a right-angled capacitive sensor array to obtain a first capacitance value and a second capacitance value containing gesture features; a subspace judgment module for inputting the first capacitance value into a pre-deployed subspace judgment model and outputting a subspace corresponding to the target gesture; a gesture recognition module for inputting the first capacitance value and the second capacitance value into a pre-deployed gesture recognition model corresponding to the subspace for recognition, and outputting a recognition result. The present invention is based on collecting capacitance data containing gesture features at different angles, dividing them into different recognition subspaces, spatially locating the gesture to be recognized, and then performing gesture recognition in the corresponding space, thereby expanding the recognition range, being able to obtain gesture features more comprehensively, and further improving the recognition accuracy.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention belongs to the technical field of gesture recognition, and in particular relates to a gesture recognition method and system based on a right-angle dual-plane capacitive sensing array. Background Art

[0002] In human-computer interaction technology, gesture recognition has become an important way of interaction. With the popularization of wearable devices and Internet of Things technology, gesture recognition will become more and more important. It will become one of the key technologies to connect various devices and systems and achieve seamless interaction. Recognizing gestures through very subtle changes in capacitance of capacitive sensors can provide high-precision gesture recognition. Capacitive sensors can achieve non-contact operation and obtain the required data without the use of special tools or equipment. These advantages make capacitive sensor gesture recognition technology have broad application prospects in the field of human-computer interaction.

[0003] However, the traditional single plane sensor has the following defects when sensing gestures:

[0004] 1. Due to the nonlinear attenuation of the detection distance, the gesture recognition accuracy decreases as the hand distance increases;

[0005] 2. Single-plane sensors can usually only effectively recognize gestures within a specific angle or direction. If the user's gesture deviates from the sensor's optimal sensing angle, the recognition accuracy may drop significantly or even fail to be recognized correctly.

[0006] 3. For complex gestures or subtle changes in movement, a single flat sensor may be difficult to accurately capture and recognize. This limits the application scope of gesture recognition, especially in situations that require high-precision and diverse gesture input. Summary of the invention

[0007] The purpose of the present invention is to provide a gesture recognition method and system based on a right-angle dual-plane capacitive sensing array to solve the problems raised in the background technology.

[0008] The present invention achieves the above-mentioned purpose through the following technical solutions:

[0009] In a first aspect, the present invention provides a gesture recognition system of a right-angle dual-plane capacitive sensing array, the system comprising:

[0010] A sensor module, used to collect a target gesture through a capacitive sensor array in a right-angle configuration, and obtain a first capacitance value and a second capacitance value containing gesture features;

[0011] A subspace determination module, configured to input the first capacitance value into a pre-deployed subspace determination model, and output a subspace corresponding to the target gesture;

[0012] The gesture recognition module is used to input the first capacitance value and the second capacitance value into a pre-deployed gesture recognition model corresponding to the subspace for recognition, and output a recognition result.

[0013] Furthermore, the sensor module comprises:

[0014] A right-angle tooling, wherein the first electrode array plate is placed on the horizontal surface of the tooling to form a first capacitive sensing array, and the second electrode array plate is placed on the vertical surface of the tooling to form a second capacitive sensing array;

[0015] And, a data acquisition device disposed on the back of the right-angle tooling, used to receive the first capacitance value and the second capacitance value and interact with an external device; wherein,

[0016] The data acquisition device switches the working state of the electrodes by driving the analog switches connected to the capacitive sensor array electrodes, so that the capacitive sensor array is in different sensitive configurations, and correspondingly collects the capacitance value containing the target gesture characteristics and sends it to the host computer. The host computer processes the capacitance value and then recognizes the gesture.

[0017] Furthermore, the different sensitive configurations include:

[0018] The first capacitive sensing array adopts The electrodes in the array are switched to form n1 sensitive configurations through analog switches. The second capacitive sensing array uses The array electrodes are switched to different electrodes through analog switches to form n2 sensitive configurations, and each sensitive configuration corresponds to obtaining a capacitance value.

[0019] Furthermore, the system further includes a model training and deployment module for pre-training the subspace judgment model and the gesture recognition model, and the training and deployment module specifically includes:

[0020] A data acquisition unit, configured to acquire a plurality of groups of the first capacitance values ​​and the second capacitance values ​​corresponding to the target gesture generated by a plurality of capacitance sensor arrays with different sensitivity configurations;

[0021] A data processing unit, configured to perform normalization processing on the first capacitance value and the second capacitance value;

[0022] A first training unit, configured to use the normalized first capacitance value as a first eigenvector value, and use the first eigenvector value to train a pre-deployed subspace judgment model;

[0023] The second training unit is used to use the normalized first capacitance value and the second capacitance value as second eigenvector values, train and deploy gesture recognition models in each subspace using the second eigenvector values, and establish a gesture recognition model library.

[0024] Furthermore, each pre-divided subspace is specifically:

[0025] The entire recognition range of the sensor array is divided into a number of independent spaces, each of which is a subspace of the system. A subspace is used as a label, and a group of first capacitance values ​​of each gesture in a subspace is regarded as a group of features under the label.

[0026] In a second aspect, the present invention proposes a gesture recognition method based on a right-angle dual-plane capacitive sensing array, the recognition method is implemented based on any of the above-mentioned recognition systems, and the method includes:

[0027] S1, collecting a target gesture through a capacitive sensor array in a right-angle configuration to obtain a first capacitance value and a second capacitance value containing gesture features;

[0028] S2. Input the first capacitance value into a pre-deployed subspace judgment model, and output the subspace corresponding to the target gesture;

[0029] S3: input the first capacitance value and the second capacitance value into a pre-deployed gesture recognition model corresponding to the subspace for recognition, and output a recognition result.

[0030] Furthermore, in step S2, constructing the subspace judgment model includes:

[0031] S2.1, obtaining a plurality of first capacitance values ​​of the target gesture by using the first capacitance sensing array including a plurality of different sensitive configurations;

[0032] S2.2, normalizing the first capacitance value obtained in step S2.1;

[0033] S2.3. Taking the normalized first capacitance value as a first eigenvector value, and using the first eigenvector value to train a subspace judgment model.

[0034] Furthermore, in step S3, deploying a gesture recognition model corresponding to the subspace includes:

[0035] S3.1, acquiring multiple groups of first capacitance values ​​and second capacitance values ​​of the target gesture by using a first capacitance sensing array and a second capacitance sensing array including multiple different sensitive configurations;

[0036] S3.2, normalizing the first capacitance value and the second capacitance value obtained in step S3.1;

[0037] S3.3. Using the normalized first capacitance value and the second capacitance value as second eigenvector values, and using the second eigenvector values ​​in each of the subspaces to train corresponding gesture recognition models, respectively, to establish a gesture recognition model library.

[0038] Compared with the existing prior art, the beneficial effects of the present invention are:

[0039] The gesture recognition method of the right-angle dual-plane capacitive sensing array in the present invention first collects capacitance data containing gesture features at different angles, divides them into different recognition subspaces, spatially locates the gesture to be recognized, and then performs gesture recognition in the corresponding space, thereby expanding the recognition range, being able to obtain gesture features more comprehensively, and further improving the recognition accuracy.

[0040] When dealing with target gestures at different angles or distances, the gesture recognition system of the present invention preferentially captures the subspace position of the gesture within the recognition range of the sensor array, and then further recognizes the gesture, thereby greatly improving the recognition accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 4 is a system block diagram of the gesture recognition system in the present invention.

[0042] Figure 2 It is a structural diagram of the sensor module in the present invention.

[0043] Figure 3 FIG. 4 is a schematic diagram of a first capacitive sensing array in the present invention.

[0044] Figure 4 FIG. 4 is a schematic diagram of a second capacitive sensing array in the present invention.

[0045] Figure 5 It is a flowchart of the gesture recognition method in the present invention.

[0046] Figure 6 FIG. 1 is a schematic diagram of 18 different sensitive configurations of the first capacitive sensing array in the present invention.

[0047] Figure 7 FIG. 2 is a schematic diagram of 26 different sensitive configurations of the second capacitive sensing array in the present invention.

[0048] Figure 8 This is a schematic diagram of the neutron space division in the present invention.

[0049] Figure 2 1. First electrode integrated board mounting slot; 2. Second electrode integrated board mounting slot; 3. Mounting box; 4. Screw hole; 5. Cable hole. DETAILED DESCRIPTION

[0050] The present application is further described in detail below in conjunction with the accompanying drawings. It is necessary to point out here that the following specific implementation methods are only used to further illustrate the present application and cannot be understood as limiting the scope of protection of the present application. Technical personnel in this field can make some non-essential improvements and adjustments to the present application based on the above application content.

[0051] Example 1

[0052] like Figure 1 As shown, this embodiment proposes a gesture recognition system of a right-angle dual-plane capacitive sensing array, and the system includes:

[0053] A sensor module, used for collecting a target gesture (the gesture to be recognized in the present invention) through a capacitive sensor array in a right-angle configuration, and obtaining a first capacitance value and a second capacitance value containing gesture features;

[0054] A subspace determination module, configured to input the first capacitance value into a pre-deployed subspace determination model, and output a subspace corresponding to the target gesture;

[0055] The gesture recognition module is used to input the first capacitance value and the second capacitance value into a pre-deployed gesture recognition model corresponding to the subspace for recognition, and output a recognition result.

[0056] Further preferably, the sensor module comprises:

[0057] A right-angle tooling, wherein the first electrode array plate is placed on the horizontal surface of the tooling to form a first capacitive sensing array, and the second electrode array plate is placed on the vertical surface of the tooling to form a second capacitive sensing array;

[0058] And, a data acquisition device disposed on the back of the right-angle tooling, used to receive the first capacitance value and the second capacitance value and interact with an external device; wherein,

[0059] The data acquisition device switches the working state of the capacitive sensor array electrodes by driving the analog switches connected to the electrodes, so that the capacitive sensor array is in different sensitive configurations, and correspondingly collects the capacitance value containing the target gesture characteristics and sends it to the host computer. The host computer processes the capacitance value and then recognizes the gesture.

[0060] Combination Figure 2In this embodiment, the capacitive sensor array specifically adopts a dual-plane capacitive sensor array. The two plane capacitive sensor arrays are installed perpendicular to each other to form a right-angle structure. The capacitance value data collected from different angles include both the spatial position characteristics of the gesture to be recognized and the hand characteristics of the gesture to be recognized. All electrodes of the first capacitive sensor array are arranged in an array and integrated on a board to form a first electrode array board. All electrodes of the second capacitive sensor array are arranged in an array and integrated on a board to form a second electrode array board. The sensor is installed in a tooling with a mounting groove. The tooling is a right-angle structure. An electrode array board mounting groove (a first electrode integrated board mounting groove 1 and a second electrode integrated board mounting groove 2) is respectively provided on the two surfaces inside the right angle, for placing the first electrode array board and the second electrode array board, so that the two capacitive sensors are in a right-angle structure, with the first electrode array board in the horizontal direction and the second electrode array board in the vertical direction; screw holes 4 are provided at the four corners of the two mounting grooves for fixing the electrode array boards, and a mounting box 3 and a back cover are also provided on the back of the mounting groove of the first electrode array board. The back cover and the mounting box 3 are fixed by screws and are used to place a data acquisition device for collecting capacitance values. A cable hole 5 is left in the mounting box 3 for easy communication between the system and external equipment.

[0061] Further preferably, the different sensitive configurations include: the first capacitive sensing array adopts The electrodes in the array are switched to form n1 sensitive configurations through analog switches. The second capacitive sensing array uses The array electrodes are switched to different electrodes through analog switches to form n2 sensitive configurations, and each sensitive configuration corresponds to obtaining a capacitance value.

[0062] Example, combined Figure 6 and Figure 7 , which shows 18 different sensitive configurations of the first capacitive sensing array and 26 different sensitive configurations of the second capacitive sensing array. As a further implementation, the first electrode array plate adopts Figure 3 The two rows of comb-tooth electrodes are closely arranged, and the second electrode array plate adopts Figure 4 In The second capacitive sensing array can be formed by Figure 7 Among the 26 switching modes, the first capacitive sensing array can be used Figure 6 There are 18 switching modes in the figure, where T represents the transmitting electrode, R represents the receiving electrode, and N represents no electrical connection. Of course, the present invention is not limited to this configuration.

[0063] Further preferably, the system further includes a model training and deployment module for pre-training the subspace judgment model and the gesture recognition model, and the training and deployment module specifically includes:

[0064] A data acquisition unit, used for acquiring a plurality of groups of first capacitance values ​​and second capacitance values ​​corresponding to a target gesture generated by a plurality of capacitance sensor arrays with different sensitivity configurations;

[0065] A data processing unit, used for normalizing the first capacitance value and the second capacitance value;

[0066] A first training unit, configured to use the normalized first capacitance value as a first eigenvector value, and train a pre-deployed subspace judgment model using the first eigenvector value;

[0067] The second training unit is used to use the normalized first capacitance value and the second capacitance value as second eigenvector values, train and deploy gesture recognition models in each subspace using the second eigenvector values, and establish a gesture recognition model library.

[0068] It can be understood that the gesture recognition system in this embodiment uses two mutually perpendicular capacitive sensor arrays to collect capacitive data containing gesture features at different angles, divide different recognition subspaces, and expand the recognizable range.

[0069] More specifically, the entire recognition range of the sensor array is divided into several independent spaces, each of which is a subspace of the system, and a subspace is used as a label, and a group of first capacitance values ​​of each gesture in a subspace is used as a group of features under the label; when training the subspace judgment model, the groups of first capacitance values ​​generated by all gestures in the space are used as several feature vectors to train the machine learning model to obtain the subspace judgment model; when performing gesture recognition, the first feature vector of the gesture to be recognized is input into the subspace judgment model, and the subspace to which the gesture to be recognized belongs is output. When training the gesture recognition model, the space to which the gesture belongs is determined based on the first capacitance value, and the second capacitance values ​​generated by all gestures in the space are used as several feature vectors to train the machine learning model to obtain the gesture recognition model.

[0070] In this embodiment, by training the data in each space separately, each subspace is not affected by the data in other spaces when the model is deployed, the data fit is better and the accuracy is higher. The accuracy of the edge space that originally had low accuracy and was almost invalid will be significantly improved after separate training, which means that the recognizable range is expanded.

[0071] Based on the same inventive concept, Figure 5 This embodiment also proposes a gesture recognition method corresponding to the above-mentioned gesture recognition system, specifically a gesture recognition method based on a right-angle dual-plane capacitive sensing array. The recognition method is implemented based on the above-mentioned recognition system, and the method includes:

[0072] S1, collecting a target gesture through a capacitive sensor array in a right-angle configuration to obtain a first capacitance value and a second capacitance value containing gesture features;

[0073] S2, inputting the first capacitance value into a pre-deployed subspace judgment model, and outputting a subspace corresponding to the target gesture;

[0074] S3: input the first capacitance value and the second capacitance value into a pre-deployed gesture recognition model corresponding to the subspace for recognition, and output a recognition result.

[0075] Further preferably, in step S2, constructing the subspace judgment model includes:

[0076] S2.1, obtaining a plurality of first capacitance values ​​of the target gesture by using a first capacitance sensing array including a plurality of different sensitive configurations;

[0077] S2.2, normalizing the first capacitance value obtained in step S2.1;

[0078] S2.3. The normalized first capacitance value is used as a first eigenvector value, and the subspace judgment model is trained using the first eigenvector value.

[0079] Combination Figure 8 In this embodiment, the capacitance value of the target gesture is obtained under a plurality of different sensitive configurations of the capacitive sensor array, that is, the capacitance data collected at different angles, and then the recognition range of the capacitive sensor array is divided according to the above capacitance value to obtain a plurality of subspaces of the capacitive sensor array; wherein, the subspace refers to: dividing the entire recognition range of the system into a plurality of independent spaces, each independent space being a subspace of the system.

[0080] Further preferably, in step S3, deploying the gesture recognition model corresponding to the subspace includes:

[0081] S3.1, acquiring multiple groups of first capacitance values ​​and second capacitance values ​​of the target gesture by using a first capacitance sensing array and a second capacitance sensing array including multiple different sensitive configurations;

[0082] S3.2, normalizing the first capacitance value and the second capacitance value obtained in step S3.1;

[0083] S3.3. The normalized first capacitance value and the second capacitance value are used as second eigenvector values, and the second eigenvector values ​​in each subspace are used to train corresponding gesture recognition models respectively, so as to establish a gesture recognition model library.

[0084] According to an embodiment of the present invention, the gesture recognition method of the right-angle dual-plane capacitive sensing array of the present invention first spatially locates the gesture to be recognized, and then performs gesture recognition in the corresponding space, thereby expanding the recognition range, being able to more comprehensively obtain gesture features, and further improving the recognition accuracy.

[0085] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0086] In addition, each functional module in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0087] If the function is implemented in the form of a software function module and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium and includes several instructions for 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 method described in the embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk, etc., various media that can store program codes.

[0088] The above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it. Although the present application has been described in detail with reference to the aforementioned embodiments, a person skilled in the art should understand that the technical solutions described in the aforementioned embodiments can still be modified, or some of the technical features can be replaced by equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A gesture recognition system of a right-angle dual-plane capacitive sensing array, characterized in that: The system comprises: A sensor module, used to collect a target gesture through a capacitive sensor array in a right-angle configuration, and obtain a first capacitance value and a second capacitance value containing gesture features; A subspace determination module, configured to input the first capacitance value into a pre-deployed subspace determination model, and output a subspace corresponding to the target gesture; a gesture recognition module, configured to input the first capacitance value and the second capacitance value into a pre-deployed gesture recognition model corresponding to the subspace for recognition, and output a recognition result; The sensor module comprises: A right-angle tooling, wherein the first electrode array plate is placed on the horizontal surface of the tooling to form a first capacitive sensing array, and the second electrode array plate is placed on the vertical surface of the tooling to form a second capacitive sensing array; And, a data acquisition device disposed on the back of the right-angle tooling, used to receive the first capacitance value and the second capacitance value and interact with an external device; wherein, The data acquisition device switches the working state of the electrodes by driving the analog switches connected to the electrodes of the capacitive sensor array, so that the capacitive sensor array is in different sensitive configurations, and correspondingly collects the capacitance value containing the target gesture feature and sends it to the host computer, which processes the capacitance value and then recognizes the gesture; The different sensitive configurations include: The first capacitive sensing array uses M1*N1 array electrodes to switch different electrodes through analog switches to form n1 sensitive configurations, and the second capacitive sensing array uses M2*N2 array electrodes to switch different electrodes through analog switches to form n2 sensitive configurations, and each sensitive configuration corresponds to obtaining a capacitance value; The system further includes a model training and deployment module for pre-training the subspace judgment model and the gesture recognition model. The training and deployment module specifically includes: A data acquisition unit, configured to acquire a plurality of groups of the first capacitance values ​​and the second capacitance values ​​corresponding to the target gesture generated by a plurality of capacitance sensor arrays with different sensitivity configurations; A data processing unit, configured to perform normalization processing on the first capacitance value and the second capacitance value; A first training unit, configured to use the normalized first capacitance value as a first eigenvector value, and use the first eigenvector value to train a pre-deployed subspace judgment model; a second training unit, configured to use the normalized first capacitance value and the second capacitance value as second eigenvector values, train and deploy gesture recognition models in each pre-divided subspace using the second eigenvector values, and establish a gesture recognition model library; Each pre-divided subspace is specifically: The entire recognition range of the sensor array is divided into a number of independent spaces, each of which is a subspace of the system. A subspace is used as a label, and a group of first capacitance values ​​of each gesture in a subspace is regarded as a group of features under the label.

2. A gesture recognition method based on a right-angle dual-plane capacitive sensing array, the recognition method is implemented based on the recognition system of claim 1, characterized in that: Methods include: S1, collecting a target gesture through a capacitive sensor array in a right-angle configuration to obtain a first capacitance value and a second capacitance value containing gesture features; S2. Input the first capacitance value into a pre-deployed subspace judgment model, and output the subspace corresponding to the target gesture; S3: input the first capacitance value and the second capacitance value into a pre-deployed gesture recognition model corresponding to the subspace for recognition, and output a recognition result.

3. The method for gesture recognition based on a right-angle dual-plane capacitive sensing array according to claim 2, characterized in that: In step S2, constructing the subspace judgment model includes: S2.1, obtaining a plurality of first capacitance values ​​of the target gesture by using the first capacitance sensing array including a plurality of different sensitive configurations; S2.2, normalizing the first capacitance value obtained in step S2.1; S2.

3. Taking the normalized first capacitance value as a first eigenvector value, and using the first eigenvector value to train a subspace judgment model.

4. The method for gesture recognition based on a right-angle dual-plane capacitive sensing array according to claim 3, characterized in that: In step S3, deploying a gesture recognition model corresponding to the subspace includes: S3.1, acquiring multiple groups of first capacitance values ​​and second capacitance values ​​of the target gesture by using a first capacitance sensing array and a second capacitance sensing array including multiple different sensitive configurations; S3.2, normalizing the first capacitance value and the second capacitance value obtained in step S3.1; S3.

3. Using the normalized first capacitance value and the second capacitance value as second eigenvector values, and using the second eigenvector values ​​in each of the subspaces to train corresponding gesture recognition models, respectively, to establish a gesture recognition model library.