Surface electromyography signal recognition method and system

Abstract

The present invention provides a surface electromyographic signal recognition method and system, wherein the method includes the following steps: establishing electromyographic control relationships between each monomer finger and each mechanical finger, and obtaining each mechanical finger sleeve for sleeved mechanical finger; performing morphological acquisition based on each different monomer finger, and obtaining each test finger sleeve for sleeved each monomer finger; establishing each color association relationship between each different mechanical color and each different test color in one-to-one correspondence based on each electromyographic control relationship; in response to the first detection end detecting finger activity of any monomer finger, triggering the second detection end to detect the mechanical finger, and obtaining the mechanical activity range corresponding to the mechanical finger; in response to the mechanical activity range being different from the monomer activity range, obtaining the initial electromyographic signal, and updating the initial electromyographic signal based on the monomer activity range to obtain the current electromyographic signal. The present invention at least reduces the difficulty and cost of adjustment.

Description

Technical Field

[0001] The present invention relates to data processing technology, and in particular to a surface electromyography signal recognition method and system. Background Art

[0002] In the current myoelectric control manipulator technology, although it has been used in assisted rehabilitation, industrial automation operations and myoelectric control manipulators, it is a technology that uses human muscle electrical signals to control manipulators. This technology collects and analyzes the electrical signals of human muscles and converts them into action instructions for the manipulator, thereby achieving control of the manipulator. Myoelectric control technology has been applied in many fields, including rehabilitation training, industrial robots, bionic prostheses, etc. Although myoelectric control technology has made significant progress in many fields, it still faces some challenges. For example, the execution action of the manipulator deviates from the recognition gesture of the myoelectric signal, resulting in reduced accuracy of the interactive action.

[0003] In the prior art, in order to reduce the deviation between the execution action of the robot arm and the recognized gesture of the electromyographic signal, the difference between the execution action of the robot arm and the recognized gesture of the electromyographic signal is manually observed, and then the control unit of the robot arm is manually debugged. However, the method of the prior art will result in low accuracy in judging the difference due to human factors, resulting in insufficient accuracy of the interactive action after debugging.

[0004] Therefore, how to automatically debug interactive actions and improve the accuracy of debugged interactive actions has become an urgent problem to be solved. Summary of the invention

[0005] Based on the above problems, the present invention is proposed to provide a surface electromyography signal recognition method and system that overcomes the above problems or at least partially solves the above problems.

[0006] According to one aspect of the present invention, a surface electromyography signal recognition method is provided, comprising the following steps:

[0007] Establishing one-to-one correspondence between different single fingers of the corresponding tester and different mechanical fingers of the corresponding mechanical interconnection end, and obtaining mechanical finger sleeves for sleeves on different mechanical fingers, wherein each mechanical finger sleeve corresponds to a different mechanical color;

[0008] Based on the different single fingers of the corresponding test personnel, the morphology is collected respectively, and based on the obtained single morphology, each test finger sleeve is obtained to be respectively sleeved on each single finger, wherein each test finger sleeve corresponds to a different test color;

[0009] Based on each myoelectric control relationship, a one-to-one corresponding color association relationship between each different mechanical color and each different test color is established;

[0010] In response to the first detection end detecting that any single finger has a finger activity corresponding to the single finger activity range, the second detection end is triggered to detect a mechanical finger having a color association relationship with the single finger to obtain a mechanical activity range corresponding to the mechanical finger;

[0011] In response to the mechanical activity range being different from the single-body activity range, an initial myoelectric signal corresponding to driving the mechanical finger to move to the mechanical activity range is acquired, and the initial myoelectric signal is updated based on the single-body activity range to obtain a current myoelectric signal.

[0012] Optionally, in the method according to the present invention, the morphology of each different single finger of the corresponding test person is collected respectively, and each test finger sleeve for each single finger is obtained based on the obtained single morphology, wherein each test finger sleeve corresponds to each different test color, including:

[0013] Creating an image acquisition layer and sending it to a test end corresponding to the tester, wherein the image acquisition layer includes a front view filling slot corresponding to the front view angle at the center position and an upper view filling slot corresponding to the upper view angle located above the front view filling slot;

[0014] In response to the test end interacting with any filling slot, triggering an image acquisition plug-in pre-installed on the test end to perform image acquisition to obtain an acquisition perspective image;

[0015] Comparing the captured viewing angle image with a reference viewing angle image that is pre-filled in the filling slot and corresponds to the same viewing angle as the filling slot, to obtain an image evaluation value corresponding to the captured viewing angle image;

[0016] In response to the image evaluation value being greater than a preset evaluation threshold, replacing the reference perspective image located in the filling slot with the acquisition perspective image;

[0017] In response to all the reference perspective images corresponding to all the filling slots being replaced, morphology acquisition is performed based on the acquisition perspective images corresponding to different perspectives in the image acquisition layer to obtain the monomer forms corresponding to different monomer fingers located on the hand limbs;

[0018] Establishing a model display layer corresponding to the image acquisition layer, and generating finger models corresponding to the individual fingers and finger sleeve models respectively sleeved on the finger models in the model display layer based on the individual shapes;

[0019] Sending the model display layer to the test end, and in response to the test end's interaction with any finger sleeve model, calling a preset color group, wherein the preset color group includes different test colors;

[0020] In response to the test end selecting any test color in the preset color group, rendering the finger sleeve model based on the test color, and removing the test color from the preset color group to obtain an updated preset color group;

[0021] In response to the completion of rendering of all the finger cuff models, test finger cuffs respectively connected with the monomer gloves are produced based on the finger cuff models.

[0022] Optionally, in the method according to the present invention, performing image comparison between the acquired viewing angle image and a reference viewing angle image pre-filled in the filling slot and corresponding to the same viewing angle as the filling slot to obtain an image evaluation value corresponding to the acquired viewing angle image includes:

[0023] Inputting the captured viewing angle image into a pre-trained hand recognition model to determine a captured hand image of a corresponding hand limb located in the captured viewing angle image;

[0024] Retrieving a reference perspective image that is pre-filled in the filling slot and corresponds to the same perspective as the filling slot, wherein the reference perspective image includes a reference hand image of a reference hand;

[0025] Creating a transparent contrast layer, and sequentially stacking the acquired viewing angle image and the reference viewing angle image on the transparent contrast layer to obtain a viewing angle contrast layer;

[0026] Performing pixel processing on the viewing angle comparison layer to obtain each collected hand pixel point constituting the collected hand image and each reference hand pixel point constituting the reference hand image;

[0027] Determine the overlap between each collected hand pixel point and each reference hand pixel point, and obtain the corresponding pixel overlap number based on the determination result;

[0028] Obtaining the number of pixel acquisitions corresponding to each collected hand pixel point, calculating the ratio of the number of pixel acquisitions to the number of pixel overlaps, and obtaining a first evaluation value of the corresponding overlap dimension based on the obtained pixel ratio;

[0029] Acquire a collection center point corresponding to the collected hand image and a reference center point corresponding to the reference hand image based on the perspective comparison layer, and generate a center connection line connecting the collection center point and the reference center point;

[0030] Determine the length of a center line segment corresponding to the center connection line, and obtain a second evaluation value corresponding to the deviation dimension based on the length of the center line segment;

[0031] The first evaluation value and the second evaluation value are summed to obtain an image evaluation value corresponding to the captured viewing angle image.

[0032] Optionally, in the method according to the present invention, morphology acquisition is performed based on the acquired perspective images corresponding to different perspectives in the image acquisition layer to obtain the monomer forms corresponding to different monomer fingers located on the hand limbs, including:

[0033] Determine the captured perspective image corresponding to the frontal perspective in the image capture layer as the frontal perspective captured image, and perform image interception on the captured hand image in the frontal perspective captured image to obtain the frontal perspective captured hand image;

[0034] Inputting the intercepted frontal view captured hand image into a pre-trained palm recognition model to determine the frontal view captured palm image of the corresponding palm part located in the frontal view captured hand image;

[0035] Acquire a palm contour corresponding to the front-view captured palm image, and determine an image portion outside the palm contour in the front-view captured hand image as a front-view captured finger image;

[0036] Acquire finger contours constituting the corresponding frontal captured finger image, and determine the remaining contours in the finger contours except the contours overlapping with the palm contour as morphological contours;

[0037] Performing coordinate processing on the front-view captured finger image, and obtaining the image coordinate points constituting the morphological contour;

[0038] Performing difference calculations based on coordinate values ​​between adjacent image coordinate points to obtain coordinate differences, and determining mutation points with mutation attributes existing in the image coordinate points based on the coordinate differences, wherein the mutation attributes include peak attributes and valley attributes;

[0039] In each mutation point, two mutation points with the smallest corresponding coordinate value and the second smallest corresponding coordinate value are respectively determined as the first marking point and the second marking point, and the mutation points with valley bottom attributes that are arranged continuously are grouped in pairs along the contour extension direction of the morphological contour to obtain valley bottom point groups;

[0040] In response to any valley point group including both the first marked point and the second marked point, a point association relationship is established between the valley point group and the monomer finger corresponding to the thumb among the different monomer fingers, and point association relationships are established between the remaining other valley point groups and the remaining other monomer fingers based on the sequence of the monomers corresponding to the different monomer fingers;

[0041] Based on each valley point group, each monomer form corresponding to each different monomer finger located on the hand limb with a point association relationship is obtained.

[0042] Optionally, in the method according to the present invention, obtaining the monomer forms corresponding to the different monomer fingers having the point association relationship based on the valley point groups includes:

[0043] Vertically mapping two mutation points in the same valley point group to the overlapped contour respectively, obtaining two valley mapping points corresponding to the same valley point group, and obtaining a valley mapping contour composed of the two valley mapping points in the overlapped contour;

[0044] The contour length corresponding to the valley bottom mapping contour is multiplied by the retrieved preset width coefficient to obtain a width shape corresponding to a single finger having a point-association relationship with the valley bottom point group;

[0045] In the morphological contour, a mutation point with a peak attribute between two mutation points in the same valley point group is determined as a fingertip original point, and the fingertip original point is vertically mapped to the coincident contour to obtain a fingertip mapping point;

[0046] Connecting the original fingertip point with the mapped fingertip point to obtain a fingertip connection line segment;

[0047] Acquire the length of the fingertip line segment corresponding to the fingertip connection line segment, and multiply the fingertip line segment length by the retrieved preset length coefficient to obtain the length shape corresponding to the single finger having a point-association relationship with the valley point group;

[0048] Determine the acquisition perspective image corresponding to the upward viewing perspective in the image acquisition layer as the upward viewing perspective acquisition image, and perform image interception on the acquisition hand image located in the upward viewing perspective acquisition image to obtain the upward viewing hand image;

[0049] Determine a perspective conversion relationship between the front-view captured hand image and the upward-view captured hand image, and determine, based on the perspective conversion relationship, conversion points in the upward-view captured hand image corresponding to mutation points having valley bottom attributes;

[0050] Taking each conversion point as a starting point, forming each vertical dividing line extending in a vertical direction, and dividing the upward-viewed hand image based on each vertical dividing line to obtain each divided sub-area corresponding to each valley bottom point group;

[0051] Coordinate processing is performed on the upward-viewed hand image, and the maximum coordinate point and the minimum coordinate point corresponding to the Y-axis maximum value and the Y-axis minimum value located in the same divided sub-area are obtained, and the maximum coordinate point and the minimum coordinate point are coordinate-connected to obtain a coordinate connection line segment;

[0052] Obtaining the length of the coordinate line segment corresponding to the coordinate connecting line segment, and multiplying the length of the coordinate line segment by the retrieved preset thickness coefficient to obtain the thickness shape corresponding to the single finger having a point-association relationship with the valley point group;

[0053] The length form, width form and thickness form corresponding to the same monomer finger are assembled to obtain the monomer forms corresponding to the monomer fingers.

[0054] Optionally, in the method according to the present invention, in response to the first detection end detecting that any single finger has a finger activity corresponding to the single finger activity range, the second detection end is triggered to detect a mechanical finger having a color association relationship with the single finger to obtain a mechanical activity range corresponding to the mechanical finger, including:

[0055] In response to receiving a reference limb determination signal sent by the user end, controlling the first detection end, which includes a first front view detection end and a first upward view detection end, to respectively collect images of the hand limb corresponding to the test person based on the front view angle and the upward view angle, to obtain a front view detection hand image and an upward view detection hand image;

[0056] Control the second detection end, which includes a second front view detection end and a second upward view detection end, to respectively collect images based on the front view angle and the upward view angle for each different mechanical finger, so as to obtain a front view detection mechanical image and an upward view detection mechanical image corresponding to the same mechanical finger;

[0057] In response to receiving a changed limb determination signal sent by the user terminal, controlling the first front view detection terminal and the first upward view detection terminal to respectively collect images of the hand limb based on a front view angle and an upward view angle to obtain a front view changed hand image and an upward view changed hand image;

[0058] The front-view detection hand image and the front-view change hand image corresponding to the front-view viewing angle are determined as the front-view hand comparison group, and the upward-view detection hand image and the upward-view change hand image corresponding to the upward-view viewing angle are determined as the upward-view hand comparison group;

[0059] Performing image comparison on the detected hand image and the changed hand image in the same hand comparison group, and determining whether any single finger located on the hand limb has a finger activity within a corresponding single activity range based on the comparison result;

[0060] In response to finger activity in the corresponding monomer activity range, the second detection end is triggered to collect images of the mechanical finger having a color association relationship with the monomer finger at the same viewing angle as the hand comparison group, to obtain a changed mechanical image corresponding to the mechanical finger;

[0061] The changed mechanical image and the detected mechanical image corresponding to the same viewing angle are compared, and the mechanical activity range corresponding to the mechanical finger is obtained based on the comparison result.

[0062] Optionally, in the method according to the present invention, performing image comparison on the detected hand image and the changed hand image in the same hand comparison group, and determining whether any single finger located on the hand limb has a finger activity within a corresponding single activity range based on the comparison result, includes:

[0063] Performing coordinate processing on the detected hand image and the changed hand image in the same hand comparison group, respectively, to obtain the detected coordinate points constituting the detected hand image and the changed coordinate points constituting the changed hand image;

[0064] Aggregate each detection coordinate point and each change coordinate point corresponding to the same test color into a detection coordinate group and a change coordinate group, respectively, and compare the detection coordinate group and the change coordinate group corresponding to the same test color, and determine each test attribute corresponding to each test color at different viewing angles based on the comparison result, wherein the test attributes include a maintenance attribute and a change attribute;

[0065] In response to the test attribute corresponding to any test color at any viewing angle being a variable attribute, it is determined that a finger monomer corresponding to the test color has a finger activity within a corresponding finger monomer activity range.

[0066] Optionally, in the method according to the present invention, coordinate comparison is performed on the detection coordinate group and the change coordinate group corresponding to the same test color, and based on the comparison result, the test attributes corresponding to each test color at different viewing angles are determined, including:

[0067] Respectively obtaining each detection coordinate point and each change coordinate point at the edge position in the detection coordinate group and the change coordinate group corresponding to any test color, determining a detection coordinate profile and a change coordinate profile based on each detection coordinate point and each change coordinate point at the edge position, and performing a profile comparison between the detection coordinate profile and the change coordinate profile;

[0068] In response to determining that there is a profile difference between the detection coordinate profile and the change coordinate profile based on the comparison result, determining a viewing angle corresponding to the profile difference;

[0069] In response to the viewing angle being an upward viewing angle, determining the test attribute of the test color at the upward viewing angle as a change attribute;

[0070] In response to the viewing angle being a normal viewing angle, the acquired test colors respectively corresponding to different contour differences under the normal viewing angle are determined as linkage groups to be determined, and the test colors respectively corresponding to no contour differences are determined as reference static groups;

[0071] Determine the test colors corresponding to the change coordinate contours that are in a fitting relationship and are continuously arranged in the linkage group to be determined as the same linkage color group;

[0072] In the linkage color group, a test color adjacent to any test color in the reference static group is determined as a linkage color, and a test attribute of the linkage color at the normal viewing angle is determined as a change attribute.

[0073] Optionally, in the method according to the present invention, in response to the mechanical activity range being different from the monomer activity range, acquiring an initial myoelectric signal corresponding to driving the mechanical finger to move to the mechanical activity range, and updating the initial myoelectric signal based on the monomer activity range to obtain a current myoelectric signal, including:

[0074] In response to the mechanical activity range being different from the monomer activity range, acquiring a viewing angle corresponding to the monomer activity range;

[0075] In response to the viewing angle being a front-view viewing angle, determining a rotation drive unit for controlling the rotation of a mechanical finger corresponding to the mechanical activity range, and determining an initial electromyographic signal corresponding to driving the mechanical finger to move to the mechanical activity range based on the rotation drive unit;

[0076] Controlling the rotation drive unit to move the mechanical finger corresponding to the mechanical activity range from the mechanical activity range to an activity range corresponding to the monomer activity range, and obtaining a rotation increment signal;

[0077] The initial electromyographic signal is superimposed on the rotation increment signal to complete the updating of the initial electromyographic signal to obtain a current electromyographic signal;

[0078] and / or

[0079] In response to the viewing angle being an upward viewing angle, determining a bending drive unit for performing bending control on a mechanical finger corresponding to the mechanical activity range, and determining an initial electromyographic signal corresponding to driving the mechanical finger to move to the mechanical activity range based on the bending drive unit;

[0080] Controlling the bending drive unit to move the mechanical finger corresponding to the mechanical activity range from the mechanical activity range to an activity range corresponding to the monomer activity range, and obtaining a bending increment signal;

[0081] The initial electromyographic signal is superimposed on the bending increment signal to complete the updating of the initial electromyographic signal and obtain the current electromyographic signal.

[0082] According to another aspect of the present invention, there is provided a surface electromyography signal recognition system, comprising:

[0083] The myoelectric establishment module is configured to establish myoelectric control relationships between different single fingers of the corresponding tester and different mechanical fingers of the corresponding mechanical interconnection end, and obtain mechanical finger sleeves for sleeves on different mechanical fingers, wherein each mechanical finger sleeve corresponds to a different mechanical color;

[0084] The morphology collection module is configured to collect morphologies of different individual fingers of the corresponding test personnel, and determine the test finger sleeves to be respectively sleeved on the individual fingers based on the obtained individual morphologies, wherein the test finger sleeves correspond to different test colors;

[0085] The association establishment module is configured to establish color association relationships between different mechanical colors and different test colors in a one-to-one correspondence based on each electromyographic control relationship;

[0086] The activity detection module is configured to, in response to the first detection end detecting that any single finger has a finger activity corresponding to the single finger activity range, trigger the second detection end to detect a mechanical finger having a color association relationship with the single finger, and obtain a mechanical activity range corresponding to the mechanical finger;

[0087] The myoelectric updating module is configured to obtain an initial myoelectric signal corresponding to driving the mechanical finger to move to the mechanical activity range in response to the mechanical activity range being different from the single-unit activity range, and update the initial myoelectric signal based on the single-unit activity range to obtain a current myoelectric signal.

[0088] According to the solution of the present invention, the present invention pre-establishes a one-to-one correspondence between the tester's single finger and the mechanical finger, and cleverly uses the different colors of the mechanical finger sleeve and the test finger sleeve for intuitive color association. This method greatly simplifies the complex parameter setting and debugging process in the traditional electromyographic control system. The tester does not need to deeply understand the complex system configuration, but only needs to quickly identify and adjust the control relationship of each finger according to the color correspondence, making the adjustment process intuitive and easy to operate, greatly reducing the difficulty and time cost of operation, and can also achieve the automation of the debugging process while ensuring the accuracy of debugging. BRIEF DESCRIPTION OF THE DRAWINGS

[0089] Figure 1 A flow chart of a surface electromyography signal recognition method according to an embodiment of the present invention is shown;

[0090] Figure 2 A schematic diagram of collecting a hand image from the front view in this embodiment is shown;

[0091] Figure 3 A schematic diagram of upward-looking acquisition of a hand image in this embodiment is shown;

[0092] Figure 4 A structural block diagram of a surface electromyography signal recognition system according to another embodiment of the present invention is shown. DETAILED DESCRIPTION

[0093] The exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0094] In order to solve the above problems in the prior art, the inventor proposes the solution of the present invention. An embodiment of the present invention provides a surface electromyography signal recognition method, which can be executed in a computing device.

[0095] Figure 1 A flow chart of a surface electromyography signal recognition method according to an embodiment of the present invention is shown. The method is suitable for execution in a computing device, wherein the computing device can be understood as a mobile phone or computer with data processing function.

[0096] like Figure 1 As shown, the surface electromyography signal recognition method proposed in this embodiment starts from step S101, wherein step S101 may include the following contents:

[0097] Establish one-to-one corresponding electromyographic control relationships between different single fingers of the corresponding test personnel and different mechanical fingers of the corresponding mechanical interconnection end, and obtain mechanical finger sleeves for sleeves on different mechanical fingers, wherein each mechanical finger sleeve corresponds to a different mechanical color.

[0098] For example, in this embodiment, the mechanical interconnection end can be understood as a mechanical arm, which includes different mechanical fingers. Each mechanical finger has a one-to-one electromyographic control relationship with different individual fingers of the corresponding tester. That is, when conducting the corresponding test, the tester can determine whether the corresponding mechanical finger has the same activity by performing an activity on any test finger, thereby determining the control sensitivity of the mechanical finger, and realizing automatic adjustment of the mechanical finger based on the judgment result, thereby improving the corresponding interaction accuracy.

[0099] It should be noted that in order to be able to distinguish each mechanical finger of the mechanical interconnection end in a visual dimension, a corresponding mechanical finger sleeve can be pre-set for sleeve connection, and each mechanical finger sleeve can have a corresponding different mechanical color. For example, in general, there should be five mechanical fingers, corresponding to the thumb, index finger, middle finger, ring finger and little finger, then the corresponding mechanical colors can be white, black, red, green and blue, or can also be composed of other colors; different mechanical colors are used to distinguish each mechanical finger, so that the target positioning of the corresponding mechanical finger can be quickly completed during the test.

[0100] In step S102, the following contents are included:

[0101] The shapes of the different individual fingers of the corresponding test personnel are collected respectively, and the test finger sleeves respectively fitted on the individual fingers are obtained based on the obtained shapes of the individual fingers, wherein the test finger sleeves respectively correspond to different test colors.

[0102] For example, in this embodiment, after the distinction of each mechanical finger is completed, the different individual fingers of the tester can be distinguished. Similarly, the distinction of the corresponding visual dimension can also be achieved based on color, that is, by collecting the morphology of each different individual finger of the tester, the test finger sleeves that are adapted to each individual finger and can be sleeved are determined based on the obtained individual morphologies, and each test finger sleeve also has a corresponding different test color;

[0103] It should be noted that, here, the mechanical color and the test color can be the same color or different colors. In order to more accurately illustrate the one-to-one electromyographic control relationship between each individual finger and each mechanical finger, the test color and mechanical color corresponding to the individual finger and the mechanical finger with the same electromyographic control relationship can be set to the same color.

[0104] Furthermore, in this embodiment, the above-mentioned “collecting the shapes of different individual fingers of the corresponding test personnel, and obtaining the test finger sleeves respectively sleeved on the individual fingers based on the obtained shapes of the individual fingers, wherein the test finger sleeves respectively correspond to different test colors” may also include the following steps:

[0105] Creating an image acquisition layer and sending it to a test end corresponding to the tester, wherein the image acquisition layer includes a front view filling slot corresponding to the front view angle at the center position and an upper view filling slot corresponding to the upper view angle located above the front view filling slot;

[0106] In response to the test end interacting with any filling slot, triggering an image acquisition plug-in pre-installed on the test end to perform image acquisition to obtain an acquisition perspective image;

[0107] Comparing the captured viewing angle image with a reference viewing angle image that is pre-filled in the filling slot and corresponds to the same viewing angle as the filling slot, to obtain an image evaluation value corresponding to the captured viewing angle image;

[0108] In response to the image evaluation value being greater than a preset evaluation threshold, replacing the reference perspective image located in the filling slot with the acquisition perspective image;

[0109] In response to all the reference perspective images corresponding to all the filling slots being replaced, morphology acquisition is performed based on the acquisition perspective images corresponding to different perspectives in the image acquisition layer to obtain the monomer forms corresponding to different monomer fingers located on the hand limbs;

[0110] Establishing a model display layer corresponding to the image acquisition layer, and generating finger models corresponding to the individual fingers and finger sleeve models respectively sleeved on the finger models in the model display layer based on the individual shapes;

[0111] Sending the model display layer to the test end, and in response to the test end's interaction with any finger sleeve model, calling a preset color group, wherein the preset color group includes different test colors;

[0112] In response to the test end selecting any test color in the preset color group, rendering the finger sleeve model based on the test color, and removing the test color from the preset color group to obtain an updated preset color group;

[0113] In response to the completion of rendering of all the finger cuff models, test finger cuffs respectively connected with the monomer gloves are produced based on the finger cuff models.

[0114] For example, in this embodiment, the test finger sleeve may be made of rubber or other materials, which is customized based on the individual shape of each individual finger. Therefore, in order to obtain the individual shape of each different individual finger of the corresponding tester, it can be obtained through image processing. For example, the detailed process can be described as follows:

[0115] First, an image acquisition layer can be created based on the server and sent to the test end (such as a mobile phone or a computer) of the corresponding tester, wherein the image acquisition layer can be pre-created and stored in the server, and the image acquisition image includes a front view filling slot and an upper view filling slot;

[0116] Next, when the test end receives the image acquisition layer, it means that it can fill the image of the corresponding single finger into the image acquisition layer based on the image acquisition layer. Specifically, when the test end interacts with any filling slot, it triggers the image acquisition plug-in pre-installed on the test end to perform corresponding image acquisition, thereby obtaining the acquisition perspective image;

[0117] Next, after obtaining the corresponding captured perspective image, it is necessary to compare the captured perspective image with the reference perspective image pre-filled in the filling slot, so as to obtain an image evaluation value corresponding to the captured perspective image based on the comparison result. Here, it should be noted that since the filling slot includes a front view filling slot corresponding to the front view perspective and an upward view filling slot corresponding to the upward view perspective, each filling slot includes a reference perspective image of the corresponding perspective. For example, the front view filling slot should be filled with the reference perspective image corresponding to the front view perspective, and the upward view filling slot should be filled with the reference perspective image corresponding to the upward view perspective. Angle image, and the reference view image can be understood as being obtained by shooting based on the reference hand. By comparing the captured view image captured by the test end with the reference view image corresponding to the same view, the capture quality of the captured view image can be judged, so as to further determine whether it is necessary to prompt the test end to re-capture based on the capture quality, thereby improving the corresponding image capture accuracy; in addition, in the image capture layer, the front view filling slot is at the center position, and the upper view filling slot is at the upper side of the front view filling slot, so as to meet the corresponding view conversion relationship and increase the slot distinctiveness based on different view angles for each filling slot;

[0118] Then, after the image evaluation value corresponding to the acquired perspective image is completed, the image evaluation value may be compared with the retrieved preset evaluation threshold, and when the image evaluation value is greater than the preset evaluation threshold, the acquired perspective image may be used to replace the reference perspective image in the filling slot accordingly;

[0119] Then, when all the reference view images corresponding to all the filling slots are replaced, corresponding morphological acquisition can be performed based on each acquisition view image in the image acquisition layer, thereby obtaining each monomer morphology corresponding to each different monomer finger of the test person's hand limb;

[0120] Then, a model display layer corresponding to the image acquisition layer is established, and finger models corresponding to the individual fingers and finger sleeve models respectively sleeved on the finger models are generated in the model display layer based on the corresponding individual shapes;

[0121] Finally, after completing the generation of each finger sleeve model of the corresponding shape, the corresponding model display layer can be sent to the test end, and the test end can render the finger sleeve model by interacting with the finger sleeve model to select the corresponding test color from the retrieved preset color group. After completing the rendering of all finger sleeve models, the test finger sleeves that are respectively connected to each monomer glove can be customized based on the rendered finger sleeve models to improve the corresponding customization requirements.

[0122] Furthermore, in this embodiment, the above-mentioned “comparing the captured viewing angle image with a reference viewing angle image pre-filled in the filling slot and corresponding to the same viewing angle as the filling slot to obtain an image evaluation value corresponding to the captured viewing angle image” may also include the following steps:

[0123] Inputting the captured viewing angle image into a pre-trained hand recognition model to determine a captured hand image of a corresponding hand limb located in the captured viewing angle image;

[0124] Retrieving a reference perspective image that is pre-filled in the filling slot and corresponds to the same perspective as the filling slot, wherein the reference perspective image includes a reference hand image of a reference hand;

[0125] Creating a transparent contrast layer, and sequentially stacking the acquired viewing angle image and the reference viewing angle image on the transparent contrast layer to obtain a viewing angle contrast layer;

[0126] Performing pixel processing on the viewing angle comparison layer to obtain each collected hand pixel point constituting the collected hand image and each reference hand pixel point constituting the reference hand image;

[0127] Determine the overlap between each collected hand pixel point and each reference hand pixel point, and obtain the corresponding pixel overlap number based on the determination result;

[0128] Obtaining the number of pixel acquisitions corresponding to each collected hand pixel point, calculating the ratio of the number of pixel acquisitions to the number of pixel overlaps, and obtaining a first evaluation value of the corresponding overlap dimension based on the obtained pixel ratio;

[0129] Acquire a collection center point corresponding to the collected hand image and a reference center point corresponding to the reference hand image based on the perspective comparison layer, and generate a center connection line connecting the collection center point and the reference center point;

[0130] Determine the length of a center line segment corresponding to the center connection line, and obtain a second evaluation value corresponding to the deviation dimension based on the length of the center line segment;

[0131] The first evaluation value and the second evaluation value are summed to obtain an image evaluation value corresponding to the captured viewing angle image.

[0132] It should be noted that after the corresponding acquisition perspective image is obtained, the acquisition perspective image can be input into the hand recognition model for corresponding recognition, so as to obtain the acquisition hand image of the corresponding hand limb located in the acquisition perspective image. Here, the hand recognition model can be understood as being obtained by training the pre-created image recognition model with the corresponding training set (including multiple images corresponding to hand limbs of different forms). Since this is a prior art, this embodiment will not be described in detail.

[0133] After the acquisition of the collected hand image is completed, a transparent contrast layer can be further created, and the collected hand image and the reference view image located in the filling slot are sequentially superimposed on the transparent contrast layer to obtain the corresponding view contrast layer. It should be noted that the reference view image includes a reference hand image corresponding to the reference hand;

[0134] Performing corresponding pixel processing on the perspective comparison layer, thereby obtaining each collected hand pixel point and each reference hand pixel point that respectively constitute the collected hand image and the reference hand image, and determining the overlap between each hand pixel point located in different hand images, so as to obtain the corresponding pixel overlap number;

[0135] It should be noted that, here, since the reference hand image is obtained by collecting based on the reference hand, the greater the number of pixel overlaps, the higher the image quality of the collected hand image collected by the test end. Therefore, the number of pixel collections corresponding to the collected hand pixels can be further obtained, the ratio of the pixel collection number to the pixel overlap number can be calculated, and the first evaluation value of the corresponding overlap dimension can be obtained based on the obtained pixel ratio;

[0136] For example, the obtained pixel ratio may be pre-normalized to obtain a normalized coincidence value, and the normalized coincidence value and the retrieved coincidence weight may be multiplied to obtain the corresponding first evaluation value;

[0137] After obtaining the first evaluation value of the corresponding overlapping dimension, the second evaluation value of the corresponding deviation dimension can be further obtained, so that the first evaluation value and the second evaluation value can be summed to obtain the image evaluation value corresponding to the acquired viewing angle image; through image evaluation in two different dimensions, it can be ensured that the final acquired viewing angle image has a higher image quality, so as to carry out subsequent electromyographic control adjustment operations.

[0138] It should be noted that the second evaluation value may be obtained by, for example, the following method steps:

[0139] First, the acquisition center point of the corresponding acquired hand image and the reference center point of the corresponding reference hand image can be obtained in the perspective comparison layer, and a central connection line connecting the acquisition center point and the reference center point can be further generated;

[0140] Finally, by determining the length of the center line segment corresponding to the center connecting line to further obtain the second evaluation value of the corresponding deviation dimension, it can be known that the longer the center line segment length is, the greater the degree of deviation between the two, and the corresponding second evaluation value should be lower. Therefore, the obtained center line segment length can be pre-normalized to obtain the length normalization value, and the length normalization value can be inversely calculated to multiply the calculation result with the retrieved length weight to obtain the corresponding second evaluation value.

[0141] Furthermore, in this embodiment, the above-mentioned “performing morphology acquisition based on the acquisition perspective images corresponding to different perspectives located in the image acquisition layer to obtain the monomer forms corresponding to the different monomer fingers located on the hand limbs” may also include the following steps:

[0142] Determine the captured perspective image corresponding to the frontal perspective in the image capture layer as the frontal perspective captured image, and perform image interception on the captured hand image in the frontal perspective captured image to obtain the frontal perspective captured hand image;

[0143] Inputting the intercepted frontal view captured hand image into a pre-trained palm recognition model to determine the frontal view captured palm image of the corresponding palm part located in the frontal view captured hand image;

[0144] Acquire a palm contour corresponding to the front-view captured palm image, and determine an image portion outside the palm contour in the front-view captured hand image as a front-view captured finger image;

[0145] Acquire finger contours constituting the corresponding frontal captured finger image, and determine the remaining contours in the finger contours except the contours overlapping with the palm contour as morphological contours;

[0146] Performing coordinate processing on the front-view captured finger image, and obtaining the image coordinate points constituting the morphological contour;

[0147] Performing difference calculations based on coordinate values ​​between adjacent image coordinate points to obtain coordinate differences, and determining mutation points with mutation attributes existing in the image coordinate points based on the coordinate differences, wherein the mutation attributes include peak attributes and valley attributes;

[0148] In each mutation point, two mutation points with the smallest corresponding coordinate value and the second smallest corresponding coordinate value are respectively determined as the first marking point and the second marking point, and the mutation points with valley bottom attributes that are arranged continuously are grouped in pairs along the contour extension direction of the morphological contour to obtain valley bottom point groups;

[0149] In response to any valley point group including both the first marked point and the second marked point, a point association relationship is established between the valley point group and the monomer finger corresponding to the thumb among the different monomer fingers, and point association relationships are established between the remaining other valley point groups and the remaining other monomer fingers based on the sequence of the monomers corresponding to the different monomer fingers;

[0150] Based on each valley point group, each monomer form corresponding to each different monomer finger located on the hand limb with a point association relationship is obtained.

[0151] For example, in this embodiment, the morphology of each single finger can be collected by the following method steps:

[0152] First, the captured perspective image corresponding to the frontal perspective in the image capture layer can be determined as the frontal perspective captured image, and further image capture is performed on it to obtain the frontal perspective captured hand image;

[0153] Next, after the acquisition of the front-view captured hand image is completed, it can be input into the palm recognition model obtained through pre-training to determine the front-view captured palm image corresponding to the palm part in the front-view captured hand image, wherein, since the hand limbs generally include the palm and fingers, the part corresponding to the fingers does not exist in the front-view captured palm image; and it can be explained that the palm recognition model here is similar to the creation process of the above-mentioned hand recognition model, and this embodiment will not be repeated in detail;

[0154] Next, the palm contour can be determined by the acquired frontal view captured palm image, and since the frontal view captured hand image includes images of the corresponding palm and fingers, the image portion outside the palm contour in the frontal view captured hand image can be determined as the frontal view captured finger image based on the acquired palm contour. It should be noted that since the image portion corresponding to the palm is generally simple in shape, while the image portion corresponding to the finger is relatively complex in shape, the image portion corresponding to the finger can be obtained by first identifying the image portion corresponding to the palm, thereby improving the corresponding data processing efficiency;

[0155] Then, the corresponding finger contour can be obtained based on the obtained front view captured finger image, and since the finger and the palm are connected to each other, the corresponding finger contour and the palm contour also have a corresponding overlapping contour, and in order to obtain the monomer morphology of the corresponding monomer finger, it is necessary to determine the remaining contours in the finger contour except the overlapping contour as the corresponding morphology contour;

[0156] Then, coordinate processing can be performed on the finger image collected in the front view to obtain the image coordinate points constituting the morphological contour, and the difference between adjacent image coordinate points based on the coordinate values ​​can be further calculated to obtain the corresponding coordinate difference values, so as to determine the mutation points with mutation properties existing in the image coordinate points according to the coordinate difference values;

[0157] Next, it should be noted that, since the morphological contour generally corresponds to the contours of all individual fingers, and the fingertips of each individual finger are arc-shaped and have different lengths, in order to obtain the individual morphology of each individual finger, it is necessary to determine the corresponding mutation points based on the morphological contour, and determine the parts corresponding to each individual finger existing in the morphological contour based on the mutation points, that is, the two mutation points with the smallest corresponding coordinate values ​​and the second smallest coordinate values ​​in each mutation point can be respectively determined as the first marking point and the second marking point, and the mutation points with valley attributes that are continuously arranged along the contour extension direction of the morphological contour are grouped in pairs to obtain the corresponding valley point groups;

[0158] Next, when any valley point group includes both the first marked point and the second marked point, it appears that the two are located on both sides of the corresponding thumb, so a point association relationship between the valley point group and the monomer finger corresponding to the thumb in each of the different monomer fingers can be established, and further, based on the monomer sequence of each of the different monomer fingers, point association relationships between the remaining other valley point groups and the remaining other monomer fingers are established;

[0159] Finally, after the establishment of the correlation relationship between each point is completed, the monomer forms corresponding to different monomer fingers located on the hand limbs with point correlation relationship can be obtained based on each valley point group.

[0160] It should be noted that the above-mentioned "monomer order" can be understood as the arrangement order between different monomer fingers, such as: thumb, index finger, middle finger, ring finger and little finger, and the above-mentioned front view acquisition of hand image can be, for example, Figure 2 As shown, the above-mentioned upward-looking acquisition of hand images can be, for example, Figure 3 shown.

[0161] Furthermore, in this embodiment, the above-mentioned “obtaining the monomer forms corresponding to the different monomer fingers having the point association relationship based on the valley point groups” may also include the following steps:

[0162] Vertically mapping two mutation points in the same valley point group to the overlapped contour respectively, obtaining two valley mapping points corresponding to the same valley point group, and obtaining a valley mapping contour composed of the two valley mapping points in the overlapped contour;

[0163] The contour length corresponding to the valley bottom mapping contour is multiplied by the retrieved preset width coefficient to obtain a width shape corresponding to a single finger having a point-association relationship with the valley bottom point group;

[0164] In the morphological contour, a mutation point with a peak attribute between two mutation points in the same valley point group is determined as a fingertip original point, and the fingertip original point is vertically mapped to the coincident contour to obtain a fingertip mapping point;

[0165] Connecting the original fingertip point with the mapped fingertip point to obtain a fingertip connection line segment;

[0166] Acquire the length of the fingertip line segment corresponding to the fingertip connection line segment, and multiply the fingertip line segment length by the retrieved preset length coefficient to obtain the length shape corresponding to the single finger having a point-association relationship with the valley point group;

[0167] Determine the acquisition perspective image corresponding to the upward viewing perspective in the image acquisition layer as the upward viewing perspective acquisition image, and perform image interception on the acquisition hand image located in the upward viewing perspective acquisition image to obtain the upward viewing hand image;

[0168] Determine a perspective conversion relationship between the front-view captured hand image and the upward-view captured hand image, and determine, based on the perspective conversion relationship, conversion points in the upward-view captured hand image corresponding to mutation points having valley bottom attributes;

[0169] Taking each conversion point as a starting point, forming each vertical dividing line extending in a vertical direction, and dividing the upward-viewed hand image based on each vertical dividing line to obtain each divided sub-area corresponding to each valley bottom point group;

[0170] Coordinate processing is performed on the upward-viewed hand image, and the maximum coordinate point and the minimum coordinate point corresponding to the Y-axis maximum value and the Y-axis minimum value located in the same divided sub-area are obtained, and the maximum coordinate point and the minimum coordinate point are coordinate-connected to obtain a coordinate connection line segment;

[0171] Obtaining the length of the coordinate line segment corresponding to the coordinate connecting line segment, and multiplying the length of the coordinate line segment by the retrieved preset thickness coefficient to obtain the thickness shape corresponding to the single finger having a point-association relationship with the valley point group;

[0172] The length form, width form and thickness form corresponding to the same monomer finger are assembled to obtain the monomer forms corresponding to the monomer fingers.

[0173] For example, in this embodiment, in order to obtain the monomer form of each monomer finger, it is necessary to obtain the length, width and thickness of each monomer finger respectively, wherein the corresponding specific process can be described as follows:

[0174] First, since the base of each individual finger is located at the overlapped contour, two valley mapping points corresponding to the same valley point group are obtained by overlapping the vertical mapping values ​​of two mutation points in the same valley point group, and further obtaining a valley mapping contour composed of the two valley mapping points in the overlapped contour, and obtaining the contour length corresponding to the valley mapping contour, multiplying the contour length by the retrieved preset width coefficient, thereby obtaining the width shape corresponding to the individual finger having a point-association relationship with the valley point group;

[0175] Then, in the morphological contour sink, the mutation point with peak attribute between two mutation points in the same valley point group is determined as the original fingertip point, and the original fingertip point is also vertically mapped to the coincident contour to obtain the fingertip mapping point, and the corresponding fingertip connection line segment is obtained by point-to-point connection of the fingertip original point and the fingertip mapping point, and the fingertip line segment length of the corresponding fingertip connection line segment is further multiplied by the retrieved preset length coefficient, so as to obtain the length morphology corresponding to the single finger having a point-to-point association relationship with the valley point group;

[0176] Then, after acquiring the width and length of the corresponding single finger based on the frontal view capture image, the capture image of the corresponding upper view in the image capture layer can be further determined as the upper view capture image, and the capture hand image in the upper view capture image can be intercepted to acquire the thickness of each single finger through the upper view capture hand image.

[0177] Next, since the hand image collected by frontal view and the hand image collected by upward view are collected based on the same hand limb, each conversion point corresponding to each mutation point with valley bottom attribute can be determined in the hand image collected by upward view based on the perspective conversion relationship between frontal view and upward view;

[0178] Next, each conversion point is used as a starting point to form vertical division lines extending in the vertical direction, and the upward-viewed hand image is divided accordingly based on the obtained vertical division lines, thereby obtaining divided sub-areas corresponding to each valley point group;

[0179] Next, coordinate processing is performed on the hand image collected by looking upward, and the maximum coordinate point and the minimum coordinate point corresponding to the maximum value and the minimum value of the Y axis in the same divided sub-area are respectively obtained, so as to further coordinately connect the maximum coordinate point and the minimum coordinate point to obtain a coordinate connection line segment;

[0180] Finally, by obtaining the coordinate segment length of the coordinate connecting segment and multiplying it with the preset thickness coefficient, the thickness shape corresponding to the monomer finger having a point-related relationship with the valley point group is obtained, and the length shape, width shape and thickness shape corresponding to the same monomer finger are assembled to obtain the monomer shapes corresponding to each monomer finger.

[0181] In step S103, the following contents are included:

[0182] Based on each myoelectric control relationship, a one-to-one color association relationship between each different mechanical color and each different test color is established.

[0183] For example, in this embodiment, since each single finger controls a different mechanical finger respectively, a one-to-one color association relationship between each mechanical color and each test color can be established based on the corresponding electromyographic control relationship, so that in the subsequent electromyographic control process, the controlled mechanical finger can be quickly positioned according to the color association relationship.

[0184] In step S104, the following contents are included:

[0185] In response to the first detection end detecting finger activity of any single finger corresponding to the single activity range, the second detection end is triggered to detect the mechanical finger having a color association relationship with the single finger to obtain the mechanical activity range corresponding to the mechanical finger.

[0186] For example, after the corresponding color association relationship is established, the corresponding electromyographic control sensitivity test can be performed. When the pre-configured first detection end detects that any single finger has finger activity, the second detection end needs to be triggered to detect the mechanical finger that has a color association relationship with it and determine the range of its mechanical activity. In the subsequent process, the single activity range corresponding to the single finger is compared with the mechanical activity range to determine whether the mechanical finger has performed the corresponding matching mechanical activity.

[0187] Further, in this embodiment, the above-mentioned “in response to the first detection end detecting that any single finger has a finger activity corresponding to the single finger activity range, triggering the second detection end to detect a mechanical finger having a color association relationship with the single finger, and obtaining a mechanical activity range corresponding to the mechanical finger” may also include the following steps:

[0188] In response to receiving a reference limb determination signal sent by the user end, controlling the first detection end, which includes a first front view detection end and a first upward view detection end, to respectively collect images of the hand limb corresponding to the test person based on the front view angle and the upward view angle, to obtain a front view detection hand image and an upward view detection hand image;

[0189] Control the second detection end, which includes a second front view detection end and a second upward view detection end, to respectively collect images based on the front view angle and the upward view angle for each different mechanical finger, so as to obtain a front view detection mechanical image and an upward view detection mechanical image corresponding to the same mechanical finger;

[0190] In response to receiving a changed limb determination signal sent by the user terminal, controlling the first front view detection terminal and the first upward view detection terminal to respectively collect images of the hand limb based on a front view angle and an upward view angle to obtain a front view changed hand image and an upward view changed hand image;

[0191] The front-view detection hand image and the front-view change hand image corresponding to the front-view viewing angle are determined as the front-view hand comparison group, and the upward-view detection hand image and the upward-view change hand image corresponding to the upward-view viewing angle are determined as the upward-view hand comparison group;

[0192] Performing image comparison on the detected hand image and the changed hand image in the same hand comparison group, and determining whether any single finger located on the hand limb has a finger activity within a corresponding single activity range based on the comparison result;

[0193] In response to finger activity in the corresponding monomer activity range, the second detection end is triggered to collect images of the mechanical finger having a color association relationship with the monomer finger at the same viewing angle as the hand comparison group, to obtain a changed mechanical image corresponding to the mechanical finger;

[0194] The changed mechanical image and the detected mechanical image corresponding to the same viewing angle are compared, and the mechanical activity range corresponding to the mechanical finger is obtained based on the comparison result.

[0195] For example, in this embodiment, in order to enable the first detection end to determine whether each single finger has any finger movement, it is necessary to determine the shape of each single finger in a static state, that is, by receiving the reference limb determination signal sent by the user end (the terminal device used by the tester), the pre-configured first detection end including the first front view detection end and the first upward view detection end are respectively controlled to perform image acquisition at different perspectives on the hand limb corresponding to the tester, thereby obtaining a front view detection hand image and an upward view detection hand image; then, the pre-configured second detection end including the second front view detection end and the second upward view detection end can be controlled to also perform image acquisition at different perspectives on each different mechanical finger, thereby obtaining a front view detection mechanical image and an upward view detection mechanical image corresponding to the same mechanical finger; after completing the image acquisition of the hand limb and each mechanical finger in a static state, the corresponding sensitivity test can be started. After receiving the change limb determination signal sent by the user end, it can be It is determined that the tester has performed corresponding finger activities on at least one single finger. At this time, the first front view detection end and the first upward view detection end can be controlled to collect images of the tester's hand limbs at different perspectives again, so as to obtain corresponding front view changing hand images and upward view changing hand images; at the same time, the front view detection hand image and the front view changing hand image corresponding to the front view perspective are determined as the front view hand comparison group, and the upward view detection hand image and the upward view changing hand image corresponding to the upward view perspective are determined as the upward view hand comparison group; by performing image comparison between the detection hand image and the changing hand image located in the same hand comparison group, it is determined based on the comparison result whether any single finger has finger activities; when any single finger has finger activities corresponding to the single finger activity range, it will trigger the second detection end to collect images with the same perspective as the hand comparison group for the mechanical finger having a color association relationship with the single finger, and then the obtained changing mechanical image can be compared with the detection mechanical image to obtain the corresponding mechanical activity range.

[0196] It should be noted that in this embodiment, when a change is identified in any single finger, the corresponding mechanical finger can be quickly located through the corresponding color association relationship, and the corresponding activity comparison can be performed, thereby greatly reducing the complexity of sensitivity debugging in the prior art and improving the efficiency of the corresponding debugging.

[0197] Furthermore, in this embodiment, the above-mentioned “performing image comparison on the detected hand image and the changed hand image in the same hand comparison group, and determining whether any single finger located on the hand limb has a finger activity within the corresponding single activity range based on the comparison result” may also include the following steps:

[0198] Performing coordinate processing on the detected hand image and the changed hand image in the same hand comparison group, respectively, to obtain the detected coordinate points constituting the detected hand image and the changed coordinate points constituting the changed hand image;

[0199] Aggregate each detection coordinate point and each change coordinate point corresponding to the same test color into a detection coordinate group and a change coordinate group, respectively, and compare the detection coordinate group and the change coordinate group corresponding to the same test color, and determine each test attribute corresponding to each test color at different viewing angles based on the comparison result, wherein the test attributes include a maintenance attribute and a change attribute;

[0200] In response to the test attribute corresponding to any test color at any viewing angle being a variable attribute, it is determined that a finger monomer corresponding to the test color has a finger activity within a corresponding finger monomer activity range.

[0201] For example, in this embodiment, after obtaining the corresponding hand comparison group, the detected hand image and the changed hand image in the same hand comparison group can be coordinate-processed respectively to obtain the detected coordinate points constituting the detected hand image and the changed coordinate points constituting the changed hand image; by respectively aggregating the detected coordinate points and the changed coordinate points corresponding to the same test color into the corresponding detection coordinate group and the changed coordinate group, the detected coordinate group and the changed coordinate group corresponding to the same test color can be coordinate-compared, and the test attributes corresponding to each test color at different viewing angles can be determined based on the comparison results; when the test attribute corresponding to any test color at any viewing angle is a changing attribute, it is determined that the single finger corresponding to the test color has undergone a finger activity within the corresponding single activity range.

[0202] It should be noted that, in this embodiment, based on the frontal view changing hand image corresponding to the frontal view angle, it can be relatively clear to judge whether the corresponding single finger has undergone the corresponding bifurcation activity (that is, any single finger has rotated in the left and right direction), and the upward view changing hand image corresponding to the upward view angle can be relatively clear to judge whether the corresponding single finger has undergone the corresponding bending activity; wherein, the specific judgment method can be as follows:

[0203] First, for bifurcation activities, under the normal viewing angle, the number of coordinates corresponding to the corresponding change coordinate group and the detection coordinate group should be the same, while the corresponding coordinate value intervals are different. Therefore, based on this feature, the coordinate value interval of the corresponding change coordinate group can be compared with the coordinate value interval of the detection coordinate group to determine whether the corresponding change has occurred;

[0204] Secondly, for bending activities, from the upward viewing angle, when any single finger undergoes bending activity, the number of coordinates corresponding to the single finger should increase or decrease. Therefore, based on this feature, the number of coordinates of the corresponding change coordinate group can be compared with the number of coordinates of the detection coordinate group to determine whether the corresponding change has occurred.

[0205] Furthermore, in this embodiment, the above-mentioned “comparing the detection coordinate group and the change coordinate group corresponding to the same test color, and determining the test attributes corresponding to each test color at different viewing angles based on the comparison result” may also include the following steps:

[0206] Respectively obtaining each detection coordinate point and each change coordinate point at the edge position in the detection coordinate group and the change coordinate group corresponding to any test color, determining a detection coordinate profile and a change coordinate profile based on each detection coordinate point and each change coordinate point at the edge position, and performing a profile comparison between the detection coordinate profile and the change coordinate profile;

[0207] In response to determining that there is a profile difference between the detection coordinate profile and the change coordinate profile based on the comparison result, determining a viewing angle corresponding to the profile difference;

[0208] In response to the viewing angle being an upward viewing angle, determining the test attribute of the test color at the upward viewing angle as a change attribute;

[0209] In response to the viewing angle being a normal viewing angle, the acquired test colors respectively corresponding to different contour differences under the normal viewing angle are determined as linkage groups to be determined, and the test colors respectively corresponding to no contour differences are determined as reference static groups;

[0210] Determine the test colors corresponding to the change coordinate contours that are in a fitting relationship and are continuously arranged in the linkage group to be determined as the same linkage color group;

[0211] In the linkage color group, a test color adjacent to any test color in the reference static group is determined as a linkage color, and a test attribute of the linkage color at the normal viewing angle is determined as a change attribute.

[0212] For example, in this embodiment, in normal finger activities, if any single finger has a finger activity corresponding to a bifurcated activity, it may generate linkage activities of other single fingers that are adjacent to it due to physiological structure reasons. Therefore, when determining the test attribute of any single finger, it is necessary to make a corresponding determination in combination with whether linkage activities are generated, and the specific method steps can be described as follows:

[0213] First, each detection coordinate point and each change coordinate point at the edge position can be obtained in the detection coordinate group and the change coordinate group corresponding to any test color, and the detection coordinate profile and the change coordinate profile can be further determined based on the obtained detection coordinate points and the change coordinate points, so as to compare the detection group coordinate profile with the change coordinate profile;

[0214] Finally, when it is determined that there is a contour difference between the detection coordinate contour and the change coordinate contour, the viewing angle corresponding to the contour difference can be determined; wherein, when the viewing angle is an upward viewing angle, it indicates that the single finger corresponding to the test color has undergone bending activity, and at this time, the test attribute of the test color under the corresponding upward viewing angle can be directly determined as a change attribute; and if the corresponding viewing angle is a frontal viewing angle, it is necessary to determine each test color with different contour differences under the frontal viewing angle as a linkage group to be determined, and determine each test color without contour differences as a reference static group, so as to determine whether there is linkage activity based on the linkage group to be determined; since each single finger with linkage activity generally has a fitting relationship, it is possible to determine each test color corresponding to each change coordinate contour that has a fitting relationship and is continuously arranged in the linkage group to be determined as the same linkage color group, and further determine the test color that is adjacent to any test color in the reference static group in the linkage color group as a linkage color, and determine the test attribute of the linkage color under the frontal viewing angle as a change attribute.

[0215] In step S105, the following contents are included:

[0216] In response to the mechanical activity range being different from the single-body activity range, an initial myoelectric signal corresponding to driving the mechanical finger to move to the mechanical activity range is acquired, and the initial myoelectric signal is updated based on the single-body activity range to obtain a current myoelectric signal.

[0217] For example, in this embodiment, after obtaining the corresponding mechanical activity range, the mechanical activity range can be compared with the single-body activity range. When the mechanical activity range is different from the single-body activity range, it indicates that the sensitivity of the myoelectric control of the mechanical finger is poor, and corresponding automatic debugging is required. The corresponding debugging method is to obtain the initial myoelectric signal that drives the mechanical finger to move to the mechanical activity range, and update the initial myoelectric signal based on the single-body activity range to obtain the current myoelectric signal.

[0218] It should be noted that, in the present embodiment, each robotic finger can be driven by an independent driving unit to perform corresponding mechanical activities. The driving unit is, for example, a motor or a cylinder. When an error occurs between the mechanical activity range of any robotic finger and the corresponding single-body activity range, it indicates that there is a certain difference in the driving amount generated by the initial electromyographic signal, and an automatic update is required to improve the corresponding control sensitivity.

[0219] Further, in this embodiment, the above-mentioned “in response to the mechanical activity range being different from the monomer activity range, obtaining an initial myoelectric signal corresponding to driving the mechanical finger to move to the mechanical activity range, and updating the initial myoelectric signal based on the monomer activity range to obtain a current myoelectric signal” may also include the following steps:

[0220] In response to the mechanical activity range being different from the monomer activity range, acquiring a viewing angle corresponding to the monomer activity range;

[0221] In response to the viewing angle being a front-view viewing angle, determining a rotation drive unit for controlling the rotation of a mechanical finger corresponding to the mechanical activity range, and determining an initial electromyographic signal corresponding to driving the mechanical finger to move to the mechanical activity range based on the rotation drive unit;

[0222] Controlling the rotation drive unit to move the mechanical finger corresponding to the mechanical activity range from the mechanical activity range to an activity range corresponding to the monomer activity range, and obtaining a rotation increment signal;

[0223] The initial electromyographic signal is superimposed on the rotation increment signal to complete the updating of the initial electromyographic signal to obtain a current electromyographic signal;

[0224] and / or

[0225] In response to the viewing angle being an upward viewing angle, determining a bending drive unit for performing bending control on a mechanical finger corresponding to the mechanical activity range, and determining an initial electromyographic signal corresponding to driving the mechanical finger to move to the mechanical activity range based on the bending drive unit;

[0226] Controlling the bending drive unit to move the mechanical finger corresponding to the mechanical activity range from the mechanical activity range to an activity range corresponding to the monomer activity range, and obtaining a bending increment signal;

[0227] The initial electromyographic signal is superimposed on the bending increment signal to complete the updating of the initial electromyographic signal and obtain the current electromyographic signal.

[0228] For example, in this embodiment, it can be known from the above content that the finger movement of a single finger may include a rotation movement corresponding to the front view angle and a bending movement corresponding to the upward view angle. Therefore, the update of the initial electromyographic signal may also be performed differently based on the two aspects:

[0229] First, for the perspective being a normal view, a rotation driving unit for controlling the rotation of a mechanical finger corresponding to the mechanical activity range can be determined, and an initial electromyographic signal corresponding to driving the mechanical finger to move to the mechanical activity range can be further determined based on the rotation driving unit; after completing the acquisition of the initial electromyographic signal, the rotation driving unit is controlled to move the mechanical finger corresponding to the mechanical activity range from the mechanical activity range to the activity range corresponding to the monomer activity range, and the corresponding rotation increment signal is added to the initial electromyographic signal, thereby completing the update of the initial electromyographic signal to obtain the current electromyographic signal;

[0230] Second, for the upward viewing angle, the bending drive unit that controls the bending control of the robotic finger can be determined based on the above method, and the initial electromyographic signal corresponding to the mechanical activity range of the robotic finger can be determined; the robotic finger is moved from the mechanical activity range to the activity range corresponding to the single-body activity range by controlling the bending drive unit, so that the acquired bending increment signal is superimposed with the initial electromyographic signal, thereby completing the update of the initial electromyographic signal and obtaining the corresponding current electromyographic signal.

[0231] In summary, this embodiment pre-establishes a one-to-one correspondence between the tester's single finger and the mechanical finger, and cleverly uses the different colors of the mechanical finger sleeve and the test finger sleeve for intuitive color association. This method greatly simplifies the complex parameter setting and debugging process in the traditional myoelectric control system. The tester does not need to deeply understand the complex system configuration, but only needs to quickly identify and adjust the control relationship of each finger according to the color correspondence, making the adjustment process intuitive and easy to operate, greatly reducing the difficulty and time cost of operation, and can also achieve the automation of the debugging process while ensuring the accuracy of debugging.

[0232] In addition, this method also performs well in cost control. The color association design not only reduces the need for additional hardware, such as complex sensors or additional recognition devices, but also avoids tedious software programming and algorithm optimization, thereby effectively reducing the overall cost of the system. At the same time, since color is an intuitive and universally recognizable symbol, no additional professional training or technical support is required, further saving human resources and training costs.

[0233] In addition, the surface electromyography signal recognition method not only simplifies the adjustment process and improves operational efficiency through innovative color association design, but also significantly reduces the research and development and implementation costs of the system, providing a cost-effective technical solution for the widespread application of myoelectric controlled manipulators in medical rehabilitation, industrial automation and other fields.

[0234] Another embodiment of the present invention provides a surface electromyography signal recognition system. Figure 4 As the corresponding system block diagram, the system includes:

[0235] The myoelectric establishment module is configured to establish myoelectric control relationships between different single fingers of the corresponding tester and different mechanical fingers of the corresponding mechanical interconnection end, and obtain mechanical finger sleeves for sleeves on different mechanical fingers, wherein each mechanical finger sleeve corresponds to a different mechanical color;

[0236] The morphology collection module is configured to collect morphologies of different individual fingers of the corresponding test personnel, and determine the test finger sleeves to be respectively sleeved on the individual fingers based on the obtained individual morphologies, wherein the test finger sleeves correspond to different test colors;

[0237] The association establishment module is configured to establish color association relationships between different mechanical colors and different test colors in a one-to-one correspondence based on each electromyographic control relationship;

[0238] The activity detection module is configured to, in response to the first detection end detecting that any single finger has a finger activity corresponding to the single finger activity range, trigger the second detection end to detect a mechanical finger having a color association relationship with the single finger, and obtain a mechanical activity range corresponding to the mechanical finger;

[0239] The myoelectric updating module is configured to obtain an initial myoelectric signal corresponding to driving the mechanical finger to move to the mechanical activity range in response to the mechanical activity range being different from the single-unit activity range, and update the initial myoelectric signal based on the single-unit activity range to obtain a current myoelectric signal.

[0240] In the description provided herein, algorithms and displays are not inherently related to any particular computer, virtual system or other device. Various general purpose systems can also be used together with the examples of the present invention. According to the above description, it is obvious that the structure required for constructing such systems. In addition, the present invention is not directed to any specific programming language either. It should be understood that various programming languages ​​can be utilized to implement the content of the present invention described herein, and the above description of specific languages ​​is for the purpose of disclosing the preferred embodiment of the present invention.

[0241] In the description provided herein, a large number of specific details are described. However, it is understood that embodiments of the present invention can be practiced without these specific details. In some instances, well-known methods, structures and techniques are not shown in detail so as not to obscure the understanding of this description.

[0242] Similarly, it should be understood that in order to streamline the present disclosure and aid in understanding one or more of the various inventive aspects, in the above description of exemplary embodiments of the present invention, various features of the present invention are sometimes grouped together into a single embodiment, figure, or description thereof.

[0243] Those skilled in the art will appreciate that the modules or units or components of the devices in the examples disclosed herein may be arranged in the devices described in the embodiment, or alternatively may be located in one or more devices different from the devices in the examples. The modules in the foregoing examples may be combined into one module or may be divided into multiple submodules.

[0244] Those skilled in the art will appreciate that the modules in the devices in the embodiments may be adaptively changed and arranged in one or more devices different from the embodiments. The modules or units or components in the embodiments may be combined into one module or unit or component, and furthermore may be divided into a plurality of submodules or subunits or subcomponents.

[0245] Furthermore, those skilled in the art will appreciate that although some embodiments described herein include certain features included in other embodiments but not other features, the combination of features from different embodiments is meant to be within the scope of the present invention and to form different embodiments.

[0246] In addition, some of the embodiments are described herein as methods or combinations of method elements that can be implemented by a processor of a computer system or by other devices that perform the functions. Therefore, a processor with necessary instructions for implementing the method or method elements forms a device for implementing the method or method elements. In addition, the elements described herein of the device embodiments are examples of devices for implementing the functions performed by the elements for the purpose of implementing the invention.

[0247] As used herein, unless otherwise specified, the use of ordinal numbers "first," "second," "third," etc. to describe common objects merely indicates that different instances of similar objects are involved, and is not intended to imply that the objects so described must have a given order in time, space, order, or in any other manner.

[0248] Although the present invention has been described according to a limited number of embodiments, it will be apparent to those skilled in the art, with the benefit of the above description, that other embodiments may be envisioned within the scope of the invention thus described. In addition, it should be noted that the language used in this specification is primarily selected for readability and instructional purposes, rather than for explaining or limiting the subject matter of the present invention.

Claims

1. A surface electromyography signal recognition method, characterized in that: The following steps are involved: Establishing one-to-one correspondence between different single fingers of the corresponding tester and different mechanical fingers of the corresponding mechanical interconnection end, and obtaining mechanical finger sleeves for sleeves on different mechanical fingers, wherein each mechanical finger sleeve corresponds to a different mechanical color; Based on the different single fingers of the corresponding test personnel, the morphology is collected respectively, and based on the obtained single morphology, each test finger sleeve is obtained to be respectively sleeved on each single finger, wherein each test finger sleeve corresponds to a different test color; Based on each myoelectric control relationship, a one-to-one corresponding color association relationship between each different mechanical color and each different test color is established; In response to the first detection end detecting that any single finger has a finger activity corresponding to the single finger activity range, the second detection end is triggered to detect a mechanical finger having a color association relationship with the single finger to obtain a mechanical activity range corresponding to the mechanical finger; In response to the mechanical activity range being different from the monomer activity range, acquiring an initial myoelectric signal corresponding to driving the mechanical finger to move to the mechanical activity range, and updating the initial myoelectric signal based on the monomer activity range to obtain a current myoelectric signal; The morphology of each different single finger of the corresponding tester is collected respectively, and the test finger sleeves for sleeves on each single finger are obtained based on the obtained single morphology, wherein each test finger sleeve corresponds to a different test color, including: Creating an image acquisition layer and sending it to a test end corresponding to the tester, wherein the image acquisition layer includes a front view filling slot corresponding to the front view angle at the center position and an upper view filling slot corresponding to the upper view angle located above the front view filling slot; In response to the test end interacting with any filling slot, triggering an image acquisition plug-in pre-installed on the test end to perform image acquisition to obtain an acquisition perspective image; Comparing the captured viewing angle image with a reference viewing angle image that is pre-filled in the filling slot and corresponds to the same viewing angle as the filling slot, to obtain an image evaluation value corresponding to the captured viewing angle image; In response to the image evaluation value being greater than a preset evaluation threshold, replacing the reference perspective image located in the filling slot with the acquisition perspective image; In response to all the reference perspective images corresponding to all the filling slots being replaced, morphology acquisition is performed based on the acquisition perspective images corresponding to different perspectives in the image acquisition layer to obtain the monomer forms corresponding to different monomer fingers located on the hand limbs; Establishing a model display layer corresponding to the image acquisition layer, and generating finger models corresponding to the individual fingers and finger sleeve models respectively sleeved on the finger models in the model display layer based on the individual shapes; Sending the model display layer to the test end, and in response to the test end's interaction with any finger sleeve model, calling a preset color group, wherein the preset color group includes different test colors; In response to the test end selecting any test color in the preset color group, rendering the finger sleeve model based on the test color, and removing the test color from the preset color group to obtain an updated preset color group; In response to the completion of rendering of all the finger sleeve models, producing the test finger sleeves respectively connected with the monomer gloves based on the finger sleeve models; Based on the captured perspective images corresponding to different perspectives in the image capture layer, the morphology is captured to obtain the monomer morphologies corresponding to different monomer fingers located on the hand limbs, including: Determine the captured perspective image corresponding to the frontal perspective in the image capture layer as the frontal perspective captured image, and perform image interception on the captured hand image in the frontal perspective captured image to obtain the frontal perspective captured hand image; Inputting the intercepted frontal view captured hand image into a pre-trained palm recognition model to determine the frontal view captured palm image of the corresponding palm part located in the frontal view captured hand image; Acquire a palm contour corresponding to the front-view captured palm image, and determine an image portion outside the palm contour in the front-view captured hand image as a front-view captured finger image; Acquire finger contours constituting the corresponding frontal captured finger image, and determine the remaining contours in the finger contours except the contours overlapping with the palm contour as morphological contours; Performing coordinate processing on the front-view captured finger image, and obtaining the image coordinate points constituting the morphological contour; Performing difference calculations based on coordinate values ​​between adjacent image coordinate points to obtain coordinate differences, and determining mutation points with mutation attributes existing in the image coordinate points based on the coordinate differences, wherein the mutation attributes include peak attributes and valley attributes; In each mutation point, two mutation points with the smallest corresponding coordinate value and the second smallest corresponding coordinate value are respectively determined as the first marking point and the second marking point, and the mutation points with valley bottom attributes that are arranged continuously are grouped in pairs along the contour extension direction of the morphological contour to obtain valley bottom point groups; In response to any valley point group including both the first marked point and the second marked point, a point association relationship is established between the valley point group and the monomer finger corresponding to the thumb among the different monomer fingers, and point association relationships are established between the remaining other valley point groups and the remaining other monomer fingers based on the sequence of the monomers corresponding to the different monomer fingers; Based on each valley point group, the monomer forms corresponding to different monomer fingers located on the hand limbs with point association relationship are obtained; In response to the first detection end detecting that any single finger has a finger activity corresponding to the single finger activity range, the second detection end is triggered to detect a mechanical finger having a color association relationship with the single finger, and obtain a mechanical activity range corresponding to the mechanical finger, including: In response to receiving a reference limb determination signal sent by the user end, controlling the first detection end, which includes a first front view detection end and a first upward view detection end, to respectively collect images of the hand limb corresponding to the test person based on the front view angle and the upward view angle, to obtain a front view detection hand image and an upward view detection hand image; Control the second detection end, which includes a second front view detection end and a second upward view detection end, to respectively collect images based on the front view angle and the upward view angle for each different mechanical finger, so as to obtain a front view detection mechanical image and an upward view detection mechanical image corresponding to the same mechanical finger; In response to receiving a changed limb determination signal sent by the user terminal, controlling the first front view detection terminal and the first upward view detection terminal to respectively collect images of the hand limb based on a front view angle and an upward view angle to obtain a front view changed hand image and an upward view changed hand image; The front-view detection hand image and the front-view change hand image corresponding to the front-view viewing angle are determined as the front-view hand comparison group, and the upward-view detection hand image and the upward-view change hand image corresponding to the upward-view viewing angle are determined as the upward-view hand comparison group; Performing image comparison on the detected hand image and the changed hand image in the same hand comparison group, and determining whether any single finger located on the hand limb has a finger activity within a corresponding single activity range based on the comparison result; In response to finger activity in the corresponding monomer activity range, the second detection end is triggered to collect images of the mechanical finger having a color association relationship with the monomer finger at the same viewing angle as the hand comparison group, to obtain a changed mechanical image corresponding to the mechanical finger; Perform image comparison on the changed mechanical image and the detected mechanical image corresponding to the same viewing angle, and obtain the mechanical activity range corresponding to the mechanical finger based on the comparison result; Performing image comparison on the detected hand image and the changed hand image in the same hand comparison group, and determining whether any single finger located on the hand limb has a finger activity within a corresponding single activity range based on the comparison result, including: Performing coordinate processing on the detected hand image and the changed hand image in the same hand comparison group, respectively, to obtain the detected coordinate points constituting the detected hand image and the changed coordinate points constituting the changed hand image; Aggregate each detection coordinate point and each change coordinate point corresponding to the same test color into a detection coordinate group and a change coordinate group, respectively, and compare the detection coordinate group and the change coordinate group corresponding to the same test color, and determine each test attribute corresponding to each test color at different viewing angles based on the comparison result, wherein the test attributes include a maintenance attribute and a change attribute; In response to the test attribute corresponding to any test color at any viewing angle being a variable attribute, it is determined that a single finger corresponding to the test color has a finger activity within a corresponding single finger activity range.

2. The surface electromyography signal recognition method according to claim 1, characterized in that: Comparing the captured viewing angle image with a reference viewing angle image that is pre-filled in the filling slot and corresponds to the same viewing angle as the filling slot, to obtain an image evaluation value corresponding to the captured viewing angle image, including: Inputting the captured viewing angle image into a pre-trained hand recognition model to determine a captured hand image of a corresponding hand limb located in the captured viewing angle image; Retrieving a reference perspective image that is pre-filled in the filling slot and corresponds to the same perspective as the filling slot, wherein the reference perspective image includes a reference hand image of a reference hand; Creating a transparent contrast layer, and sequentially stacking the acquired viewing angle image and the reference viewing angle image on the transparent contrast layer to obtain a viewing angle contrast layer; Performing pixel processing on the viewing angle comparison layer to obtain each collected hand pixel point constituting the collected hand image and each reference hand pixel point constituting the reference hand image; Determine the overlap between each collected hand pixel point and each reference hand pixel point, and obtain the corresponding pixel overlap number based on the determination result; Obtaining the number of pixel acquisitions corresponding to each collected hand pixel point, calculating the ratio of the number of pixel acquisitions to the number of pixel overlaps, and obtaining a first evaluation value of the corresponding overlap dimension based on the obtained pixel ratio; Acquire a collection center point corresponding to the collected hand image and a reference center point corresponding to the reference hand image based on the perspective comparison layer, and generate a center connection line connecting the collection center point and the reference center point; Determine the length of a center line segment corresponding to the center connection line, and obtain a second evaluation value corresponding to the deviation dimension based on the length of the center line segment; The first evaluation value and the second evaluation value are summed to obtain an image evaluation value corresponding to the captured viewing angle image.

3. The surface electromyography signal recognition method according to claim 1, characterized in that: Based on each valley point group, the monomer forms corresponding to different monomer fingers with point association relationships are obtained, including: Vertically mapping two mutation points in the same valley point group to the overlapped contour respectively, obtaining two valley mapping points corresponding to the same valley point group, and obtaining a valley mapping contour composed of the two valley mapping points in the overlapped contour; The contour length corresponding to the valley bottom mapping contour is multiplied by the retrieved preset width coefficient to obtain a width shape corresponding to a single finger having a point-association relationship with the valley bottom point group; In the morphological contour, a mutation point with a peak attribute between two mutation points in the same valley point group is determined as a fingertip original point, and the fingertip original point is vertically mapped to the coincident contour to obtain a fingertip mapping point; Connecting the original fingertip point with the mapped fingertip point to obtain a fingertip connection line segment; Acquire the length of the fingertip line segment corresponding to the fingertip connection line segment, and multiply the fingertip line segment length by the retrieved preset length coefficient to obtain the length shape corresponding to the single finger having a point-association relationship with the valley point group; Determine the acquisition perspective image corresponding to the upward viewing perspective in the image acquisition layer as the upward viewing perspective acquisition image, and perform image interception on the acquisition hand image located in the upward viewing perspective acquisition image to obtain the upward viewing hand image; Determine a perspective conversion relationship between the front-view captured hand image and the upward-view captured hand image, and determine, based on the perspective conversion relationship, conversion points in the upward-view captured hand image corresponding to mutation points having valley bottom attributes; Taking each conversion point as a starting point, forming each vertical dividing line extending in a vertical direction, and dividing the upward-viewed hand image based on each vertical dividing line to obtain each divided sub-area corresponding to each valley bottom point group; Coordinate processing is performed on the upward-viewed hand image, and the maximum coordinate point and the minimum coordinate point corresponding to the Y-axis maximum value and the Y-axis minimum value located in the same divided sub-area are obtained, and the maximum coordinate point and the minimum coordinate point are coordinate-connected to obtain a coordinate connection line segment; Obtaining the length of the coordinate line segment corresponding to the coordinate connecting line segment, and multiplying the length of the coordinate line segment by the retrieved preset thickness coefficient to obtain the thickness shape corresponding to the single finger having a point-association relationship with the valley point group; The length form, width form and thickness form corresponding to the same monomer finger are assembled to obtain the monomer forms corresponding to the monomer fingers.

4. The surface electromyography signal recognition method according to claim 1, characterized in that: The detection coordinate group and the change coordinate group corresponding to the same test color are compared, and the test attributes corresponding to each test color at different viewing angles are determined based on the comparison results, including: Respectively obtaining each detection coordinate point and each change coordinate point at the edge position in the detection coordinate group and the change coordinate group corresponding to any test color, determining a detection coordinate profile and a change coordinate profile based on each detection coordinate point and each change coordinate point at the edge position, and performing a profile comparison between the detection coordinate profile and the change coordinate profile; In response to determining that there is a profile difference between the detection coordinate profile and the change coordinate profile based on the comparison result, determining a viewing angle corresponding to the profile difference; In response to the viewing angle being an upward viewing angle, determining the test attribute of the test color at the upward viewing angle as a change attribute; In response to the viewing angle being a normal viewing angle, the acquired test colors respectively corresponding to different contour differences under the normal viewing angle are determined as linkage groups to be determined, and the test colors respectively corresponding to no contour differences are determined as reference static groups; Determine the test colors corresponding to the change coordinate contours that are in a fitting relationship and are continuously arranged in the linkage group to be determined as the same linkage color group; In the linkage color group, a test color adjacent to any test color in the reference static group is determined as a linkage color, and a test attribute of the linkage color at the normal viewing angle is determined as a change attribute.

5. The surface electromyography signal recognition method according to claim 4, characterized in that: In response to the mechanical activity range being different from the monomer activity range, an initial myoelectric signal corresponding to driving the mechanical finger to move to the mechanical activity range is acquired, and the initial myoelectric signal is updated based on the monomer activity range to obtain a current myoelectric signal, including: In response to the mechanical activity range being different from the monomer activity range, acquiring a viewing angle corresponding to the monomer activity range; In response to the viewing angle being a front-view viewing angle, determining a rotation drive unit for controlling the rotation of a mechanical finger corresponding to the mechanical activity range, and determining an initial electromyographic signal corresponding to driving the mechanical finger to move to the mechanical activity range based on the rotation drive unit; Controlling the rotation drive unit to move the mechanical finger corresponding to the mechanical activity range from the mechanical activity range to an activity range corresponding to the monomer activity range, and obtaining a rotation increment signal; Superimposing the initial electromyographic signal with the rotation increment signal to update the initial electromyographic signal and obtain a current electromyographic signal; and / or In response to the viewing angle being an upward viewing angle, determining a bending drive unit for performing bending control on a mechanical finger corresponding to the mechanical activity range, and determining an initial electromyographic signal corresponding to driving the mechanical finger to move to the mechanical activity range based on the bending drive unit; Controlling the bending drive unit to move the mechanical finger corresponding to the mechanical activity range from the mechanical activity range to an activity range corresponding to the monomer activity range, and obtaining a bending increment signal; The initial electromyographic signal is superimposed on the bending increment signal to complete the updating of the initial electromyographic signal and obtain the current electromyographic signal.

6. A surface electromyography signal recognition system, characterized in that: include: The myoelectric establishment module is configured to establish myoelectric control relationships between different single fingers of the corresponding tester and different mechanical fingers of the corresponding mechanical interconnection end, and obtain mechanical finger sleeves for sleeves on different mechanical fingers, wherein each mechanical finger sleeve corresponds to a different mechanical color; The morphology collection module is configured to collect morphologies of different individual fingers of the corresponding test personnel, and determine the test finger sleeves to be respectively sleeved on the individual fingers based on the obtained individual morphologies, wherein the test finger sleeves correspond to different test colors; The association establishment module is configured to establish color association relationships between different mechanical colors and different test colors in a one-to-one correspondence based on each electromyographic control relationship; The activity detection module is configured to, in response to the first detection end detecting that any single finger has a finger activity corresponding to the single finger activity range, trigger the second detection end to detect a mechanical finger having a color association relationship with the single finger, and obtain a mechanical activity range corresponding to the mechanical finger; an electromyographic update module, configured to obtain an initial electromyographic signal corresponding to driving the mechanical finger to move to the mechanical activity range in response to the mechanical activity range being different from the monomer activity range, and to update the initial electromyographic signal based on the monomer activity range to obtain a current electromyographic signal; The morphology of each different single finger of the corresponding tester is collected respectively, and the test finger sleeves for sleeves on each single finger are obtained based on the obtained single morphology, wherein each test finger sleeve corresponds to a different test color, including: Creating an image acquisition layer and sending it to a test end corresponding to the tester, wherein the image acquisition layer includes a front view filling slot corresponding to the front view angle at the center position and an upper view filling slot corresponding to the upper view angle located above the front view filling slot; In response to the test end interacting with any filling slot, triggering an image acquisition plug-in pre-installed on the test end to perform image acquisition to obtain an acquisition perspective image; Comparing the captured viewing angle image with a reference viewing angle image that is pre-filled in the filling slot and corresponds to the same viewing angle as the filling slot, to obtain an image evaluation value corresponding to the captured viewing angle image; In response to the image evaluation value being greater than a preset evaluation threshold, replacing the reference perspective image located in the filling slot with the acquisition perspective image; In response to all the reference perspective images corresponding to all the filling slots being replaced, morphology acquisition is performed based on the acquisition perspective images corresponding to different perspectives in the image acquisition layer to obtain the monomer forms corresponding to different monomer fingers located on the hand limbs; Establishing a model display layer corresponding to the image acquisition layer, and generating finger models corresponding to the individual fingers and finger sleeve models respectively sleeved on the finger models in the model display layer based on the individual shapes; Sending the model display layer to the test end, and in response to the test end's interaction with any finger sleeve model, calling a preset color group, wherein the preset color group includes different test colors; In response to the test end selecting any test color in the preset color group, rendering the finger sleeve model based on the test color, and removing the test color from the preset color group to obtain an updated preset color group; In response to the completion of rendering of all the finger sleeve models, producing the test finger sleeves respectively connected with the monomer gloves based on the finger sleeve models; Based on the captured perspective images corresponding to different perspectives in the image capture layer, the morphology is captured to obtain the monomer morphologies corresponding to different monomer fingers located on the hand limbs, including: Determine the captured perspective image corresponding to the frontal perspective in the image capture layer as the frontal perspective captured image, and perform image interception on the captured hand image in the frontal perspective captured image to obtain the frontal perspective captured hand image; Inputting the intercepted frontal view captured hand image into a pre-trained palm recognition model to determine the frontal view captured palm image of the corresponding palm part located in the frontal view captured hand image; Acquire a palm contour corresponding to the front-view captured palm image, and determine an image portion outside the palm contour in the front-view captured hand image as a front-view captured finger image; Acquire finger contours constituting the corresponding frontal captured finger image, and determine the remaining contours in the finger contours except the contours overlapping with the palm contour as morphological contours; Performing coordinate processing on the front-view captured finger image, and obtaining the image coordinate points constituting the morphological contour; Performing difference calculations based on coordinate values ​​between adjacent image coordinate points to obtain coordinate differences, and determining mutation points with mutation attributes existing in the image coordinate points based on the coordinate differences, wherein the mutation attributes include peak attributes and valley attributes; In each mutation point, two mutation points with the smallest corresponding coordinate value and the second smallest corresponding coordinate value are respectively determined as the first marking point and the second marking point, and the mutation points with valley bottom attributes that are arranged continuously are grouped in pairs along the contour extension direction of the morphological contour to obtain valley bottom point groups; In response to any valley point group including both the first marked point and the second marked point, a point association relationship is established between the valley point group and the monomer finger corresponding to the thumb among the different monomer fingers, and point association relationships are established between the remaining other valley point groups and the remaining other monomer fingers based on the sequence of the monomers corresponding to the different monomer fingers; Based on each valley point group, the monomer forms corresponding to different monomer fingers located on the hand limbs with point association relationship are obtained; In response to the first detection end detecting that any single finger has a finger activity corresponding to the single finger activity range, the second detection end is triggered to detect a mechanical finger having a color association relationship with the single finger, and obtain a mechanical activity range corresponding to the mechanical finger, including: In response to receiving a reference limb determination signal sent by the user end, controlling the first detection end, which includes a first front view detection end and a first upward view detection end, to respectively collect images of the hand limb corresponding to the test person based on the front view angle and the upward view angle, to obtain a front view detection hand image and an upward view detection hand image; Control the second detection end, which includes a second front view detection end and a second upward view detection end, to respectively collect images based on the front view angle and the upward view angle for each different mechanical finger, so as to obtain a front view detection mechanical image and an upward view detection mechanical image corresponding to the same mechanical finger; In response to receiving a changed limb determination signal sent by the user terminal, controlling the first front view detection terminal and the first upward view detection terminal to respectively collect images of the hand limb based on a front view angle and an upward view angle to obtain a front view changed hand image and an upward view changed hand image; The front-view detection hand image and the front-view change hand image corresponding to the front-view viewing angle are determined as the front-view hand comparison group, and the upward-view detection hand image and the upward-view change hand image corresponding to the upward-view viewing angle are determined as the upward-view hand comparison group; Performing image comparison on the detected hand image and the changed hand image in the same hand comparison group, and determining whether any single finger located on the hand limb has a finger activity within a corresponding single activity range based on the comparison result; In response to finger activity in the corresponding monomer activity range, the second detection end is triggered to collect images of the mechanical finger having a color association relationship with the monomer finger at the same viewing angle as the hand comparison group, to obtain a changed mechanical image corresponding to the mechanical finger; Perform image comparison on the changed mechanical image and the detected mechanical image corresponding to the same viewing angle, and obtain the mechanical activity range corresponding to the mechanical finger based on the comparison result; Performing image comparison on the detected hand image and the changed hand image in the same hand comparison group, and determining whether any single finger located on the hand limb has a finger activity within a corresponding single activity range based on the comparison result, including: Performing coordinate processing on the detected hand image and the changed hand image in the same hand comparison group, respectively, to obtain the detected coordinate points constituting the detected hand image and the changed coordinate points constituting the changed hand image; Aggregate each detection coordinate point and each change coordinate point corresponding to the same test color into a detection coordinate group and a change coordinate group, respectively, and compare the detection coordinate group and the change coordinate group corresponding to the same test color, and determine each test attribute corresponding to each test color at different viewing angles based on the comparison result, wherein the test attributes include a maintenance attribute and a change attribute; In response to the test attribute corresponding to any test color at any viewing angle being a variable attribute, it is determined that a single finger corresponding to the test color has a finger activity within a corresponding single finger activity range.

Images