CN112826516A - Electromyographic signal processing method, device, equipment, readable storage medium and product - Google Patents
Electromyographic signal processing method, device, equipment, readable storage medium and product Download PDFInfo
- Publication number
- CN112826516A CN112826516A CN202110019793.1A CN202110019793A CN112826516A CN 112826516 A CN112826516 A CN 112826516A CN 202110019793 A CN202110019793 A CN 202110019793A CN 112826516 A CN112826516 A CN 112826516A
- Authority
- CN
- China
- Prior art keywords
- electromyographic signal
- corrected
- electromyographic
- recognition model
- trained
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000003672 processing method Methods 0.000 title claims abstract description 33
- 238000003860 storage Methods 0.000 title claims abstract description 13
- 238000012549 training Methods 0.000 claims abstract description 81
- 238000012937 correction Methods 0.000 claims abstract description 80
- 238000012545 processing Methods 0.000 claims abstract description 33
- 238000000034 method Methods 0.000 claims abstract description 30
- 230000009471 action Effects 0.000 claims description 26
- 230000003183 myoelectrical effect Effects 0.000 claims description 16
- 238000004590 computer program Methods 0.000 claims description 8
- 238000001514 detection method Methods 0.000 claims description 8
- 238000002372 labelling Methods 0.000 claims description 7
- 230000006870 function Effects 0.000 claims description 6
- 230000000694 effects Effects 0.000 abstract description 15
- 238000004891 communication Methods 0.000 description 10
- 238000010586 diagram Methods 0.000 description 8
- 230000008569 process Effects 0.000 description 6
- 238000005516 engineering process Methods 0.000 description 5
- 230000003287 optical effect Effects 0.000 description 4
- 238000005457 optimization Methods 0.000 description 4
- 230000005236 sound signal Effects 0.000 description 4
- 210000002569 neuron Anatomy 0.000 description 3
- 230000001133 acceleration Effects 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 239000003814 drug Substances 0.000 description 2
- 230000003993 interaction Effects 0.000 description 2
- 238000007726 management method Methods 0.000 description 2
- 210000003205 muscle Anatomy 0.000 description 2
- 210000005036 nerve Anatomy 0.000 description 2
- 230000002232 neuromuscular Effects 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 230000036982 action potential Effects 0.000 description 1
- 230000004913 activation Effects 0.000 description 1
- 230000006978 adaptation Effects 0.000 description 1
- 238000003491 array Methods 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000002567 electromyography Methods 0.000 description 1
- 230000005057 finger movement Effects 0.000 description 1
- 210000000245 forearm Anatomy 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 210000001087 myotubule Anatomy 0.000 description 1
- 238000003909 pattern recognition Methods 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
Images
Landscapes
- User Interface Of Digital Computer (AREA)
Abstract
The embodiment of the disclosure provides an electromyographic signal processing method, an electromyographic signal processing device, an electromyographic signal processing apparatus, a readable storage medium and a product, wherein the method comprises the following steps: acquiring an electromyographic signal to be identified; inputting the electromyographic signals to be recognized into the corrected electromyographic signal recognition model; recognizing a gesture corresponding to the electromyographic signal to be recognized through the corrected electromyographic signal recognition model, and outputting a gesture type; the corrected electromyographic signal recognition model is obtained by correcting a correction data set and training parameter information corresponding to the electromyographic signal recognition model to be corrected, and the correction data set comprises a plurality of groups of surface electromyographic signals different from the to-be-trained sample acquisition positions in the to-be-trained data set and corresponding label information. Therefore, a good training effect can be achieved only by a small amount of data, the data amount required by correction of the electromyographic signal recognition model is reduced on the basis of improving the accuracy of the electromyographic signal recognition model, and the training efficiency of the electromyographic signal recognition model is improved.
Description
Technical Field
The embodiment of the disclosure relates to the field of artificial intelligence, in particular to a myoelectric signal processing method, a myoelectric signal processing device, myoelectric signal processing equipment, a readable storage medium and a product.
Background
The surface electromyographic signal is the comprehensive effect of the superficial muscle electromyographic signal and the electrification activity of the nerve trunk on the skin surface, and can reflect the activity of the neuromuscular to a certain extent. Therefore, the surface electromyographic signals have important practical values in the aspects of clinical medicine, human-computer efficiency, rehabilitation medicine, sports science and the like. The electromyographic control pattern recognition has great potential in the aspect of multi-degree-of-freedom myoelectric intention decoding. But the lack of robustness of the complex factors prevents the large-scale use thereof in the field of prosthesis production. Whereas for forearm electromyography recognition systems, the distance moved is often proportional to the classification error, e.g., the classification error increases from about 5% to about 20% per longitudinal movement of one centimeter. Therefore, how to improve the recognition accuracy of the electromyographic recognition model becomes a problem to be solved urgently.
In order to solve the above technical problems, in the prior art, a model is generally trained by identifying and deploying a surface electromyographic signal characteristic having robustness to an external factor. For example, a statistical measurement variance map of the spatial correlation among channels is utilized to replace a method for extracting features from a single myoelectric channel, so that the method has higher robustness to electrode drift
In the course of implementing the present disclosure, the inventors found that at least the following problems exist in the prior art: according to the method, the spatial information is often acquired through high-definition electromyogram recording, the cost is high, the number of acquisition channels is small, the corresponding acquired data volume is limited, and the accuracy of the electromyogram recognition model cannot be effectively improved.
Disclosure of Invention
The embodiment of the disclosure provides an electromyographic signal processing method, an electromyographic signal processing device, an electromyographic signal processing apparatus, a readable storage medium and a product, which are used for solving the technical problems that the existing electromyographic recognition model processing method is high in correction cost and the data size cannot effectively improve the accuracy of an electromyographic recognition model.
In a first aspect, an embodiment of the present disclosure provides an electromyographic signal processing method, including:
acquiring an electromyographic signal to be identified;
inputting the electromyographic signals to be recognized into a corrected electromyographic signal recognition model;
recognizing a gesture corresponding to an electromyographic signal to be recognized through the corrected electromyographic signal recognition model, and outputting a gesture type;
the corrected electromyographic signal recognition model is obtained by correcting the electromyographic signal recognition model through a correction data set and training parameter information corresponding to the electromyographic signal recognition model to be corrected, the electromyographic signal recognition model to be corrected is obtained by training a data set to be trained corresponding to a plurality of gestures, and the correction data set comprises a plurality of groups of surface electromyographic signals with different acquisition positions from the samples to be trained in the data set to be trained and corresponding label information
In a second aspect, an embodiment of the present disclosure provides an electromyographic signal processing apparatus, including:
the acquisition module is used for acquiring the electromyographic signals to be identified;
the input module is used for inputting the electromyographic signals to be recognized into the corrected electromyographic signal recognition model;
the processing module is used for identifying a gesture corresponding to an electromyographic signal to be identified through the corrected electromyographic signal identification model and outputting a gesture type;
the corrected electromyographic signal recognition model is obtained by correcting the electromyographic signal recognition model through a correction data set and training parameter information corresponding to the electromyographic signal recognition model to be corrected, the electromyographic signal recognition model to be corrected is obtained after training through a data set to be trained corresponding to a plurality of gestures, and the correction data set comprises a plurality of groups of surface electromyographic signals different from the acquisition positions of samples to be trained in the data set to be trained and label information corresponding to the surface electromyographic signals.
In a third aspect, an embodiment of the present disclosure provides an electronic device, including: a memory, a processor;
a memory; a memory for storing the processor-executable instructions;
the processor is used for calling the program instructions in the memory to execute the electromyographic signal processing method according to the first aspect.
In a fourth aspect, the present disclosure provides a computer-readable storage medium, in which computer-executable instructions are stored, and when the computer-executable instructions are executed by a processor, the computer-executable instructions are used to implement the electromyographic signal processing method according to the first aspect.
In a fifth aspect, the present disclosure provides a computer program product comprising a computer program which, when executed by a processor, implements the electromyographic signal processing method according to the first aspect.
According to the electromyographic signal processing method, the electromyographic signal processing device, the electromyographic signal processing equipment, the readable storage medium and the electromyographic signal recognition model, the electromyographic signal to be recognized is recognized by the electromyographic signal recognition model which is corrected by the correction data set and comprises a plurality of groups of surface electromyographic signals different from the to-be-trained sample acquisition positions in the to-be-trained data set and the label information corresponding to the surface electromyographic signals, and therefore the recognition accuracy of the electromyographic signal recognition model can be improved. In addition, the electromyographic signal recognition model to be corrected, which is obtained after the data set to be trained corresponding to the gestures is adopted for training, is corrected, so that a good training effect can be realized only by a small amount of data, the data amount required by the electromyographic signal recognition model is reduced on the basis of improving the accuracy of the electromyographic signal recognition model, and the training efficiency of the electromyographic signal recognition model is improved.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present disclosure and together with the description, serve to explain the principles of the disclosure.
FIG. 1 is a schematic diagram of a network architecture upon which the present disclosure is based;
fig. 2 is a schematic flowchart of an electromyographic signal processing method according to a first embodiment of the disclosure;
fig. 3 is a schematic flowchart of an electromyographic signal processing method according to a second embodiment of the disclosure;
FIG. 4 is a schematic structural diagram of a model to be trained according to an embodiment of the present disclosure;
fig. 5 is a schematic flowchart of an electromyographic signal processing method according to a third embodiment of the disclosure;
fig. 6 is a schematic flowchart of an electromyographic signal processing method according to a fourth embodiment of the disclosure;
fig. 7 is a schematic structural diagram of an electromyographic signal processing apparatus according to a fifth embodiment of the present disclosure;
fig. 8 is a schematic structural diagram of an electronic device according to a sixth embodiment of the present disclosure.
With the foregoing drawings in mind, certain embodiments of the disclosure have been shown and described in more detail below. These drawings and written description are not intended to limit the scope of the disclosed concepts in any way, but rather to illustrate the concepts of the disclosure to those skilled in the art by reference to specific embodiments.
Detailed Description
Reference will now be made in detail to the exemplary embodiments, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, like numbers in different drawings represent the same or similar elements unless otherwise indicated. The implementations described in the exemplary embodiments below are not intended to represent all implementations consistent with the present disclosure. Rather, they are merely examples of apparatus and methods consistent with certain aspects of the present disclosure, as detailed in the appended claims.
First, terms related to embodiments of the present disclosure are explained:
electromyographic signals: the myoelectric signals are superposition of action potentials of movement units in a plurality of myofibers on time and space;
surface electromyographic signals: the method is a comprehensive effect of a superficial muscle electromyographic signal and the electrical activity of a nerve trunk on the surface of the skin, and can reflect the activity of neuromuscular to a certain extent.
In order to solve the technical problems that the conventional electromyographic recognition model processing method is high in correction cost and the data volume cannot effectively improve the accuracy of the electromyographic recognition model, the disclosure provides an electromyographic signal processing method, an electromyographic signal processing device, electromyographic signal processing equipment, a readable storage medium and a product.
It should be noted that the electromyographic signal processing method, device, apparatus, readable storage medium and product provided by the present disclosure may be applied to optimization scenarios of various electromyographic signal recognition models.
In order to reduce errors in the use process of the electromyographic signal recognition model, in the prior art, the model is generally trained through the surface electromyographic signal with robustness to external factors by recognizing and deploying the surface electromyographic signal characteristics with robustness. However, the robust surface electromyogram signal acquired by the method is often high in acquisition cost and limited in data volume, and the optimization operation of the electromyogram signal recognition model cannot be effectively realized.
In the process of solving the technical problem, the inventor finds out through research that in order to reduce the data volume required by the correction of the electromyographic signal recognition model on the basis of improving the accuracy of the electromyographic signal recognition model, the optimization operation of the model can be realized by selecting a supervision self-adaption mode.
The inventor further researches and discovers that a preset model to be trained can be trained through training of a data set to be trained corresponding to a plurality of gestures to obtain an electromyographic signal recognition model to be corrected. And correcting the electromyographic signal recognition model to be corrected by adopting a small amount of surface electromyographic signals different from the to-be-trained sample acquisition positions in the to-be-trained data set and label information corresponding to the surface electromyographic signals to obtain the corrected electromyographic signal recognition model.
The electromyographic signal processing method provided by the embodiment of the disclosure aims to solve the above technical problems in the prior art.
The following describes the technical solutions of the present disclosure and how to solve the above technical problems in specific embodiments. The following several specific embodiments may be combined with each other, and details of the same or similar concepts or processes may not be repeated in some embodiments. Embodiments of the present disclosure will be described below with reference to the accompanying drawings.
Fig. 1 is a schematic diagram of a network architecture based on the present disclosure, and as shown in fig. 1, the network architecture based on the present disclosure at least includes: a terminal device 1 and a server 2, wherein the server 2 can be provided with a myoelectric signal processing device. The electromyographic signal processing device can be written by C/C + +, Java, Shell or Python languages and the like; the terminal device 1 may be a desktop computer, a tablet computer, or the like.
Fig. 2 is a schematic flow chart of an electromyographic signal processing method according to a first embodiment of the disclosure, and as shown in fig. 2, the method includes:
Step 202, inputting the electromyographic signals to be recognized into a corrected electromyographic signal recognition model.
the corrected electromyographic signal recognition model is obtained by correcting the electromyographic signal recognition model through a correction data set and training parameter information corresponding to the electromyographic signal recognition model to be corrected, the electromyographic signal recognition model to be corrected is obtained after training through a data set to be trained corresponding to a plurality of gestures, and the correction data set comprises a plurality of groups of surface electromyographic signals different from the acquisition positions of samples to be trained in the data set to be trained and label information corresponding to the surface electromyographic signals.
The execution main body of the embodiment is an electromyographic signal processing device, and the electromyographic signal processing device can be coupled in a server.
In the embodiment, on the basis of improving the accuracy of the electromyographic signal identification model, the data size required by the electromyographic signal identification model is reduced, and the optimization operation of the model can be realized by selecting a supervision self-adaptive mode. Specifically, a preset model to be trained may be trained by using a data set to be trained corresponding to a plurality of gestures to obtain an electromyographic signal recognition model to be corrected. After the model to be trained is trained on the basis of the data set to be trained corresponding to the gestures, the model to be corrected already has the capability of myoelectric signal gesture recognition.
In order to improve the recognition accuracy of the model, the myoelectric signal recognition model to be corrected can be corrected by adopting a correction data set to obtain a corrected myoelectric signal recognition model. Therefore, the electromyographic signal identification operation can be subsequently carried out according to the corrected electromyographic signal identification model. The correction data set comprises a plurality of groups of surface electromyographic signals which are different from the acquisition positions of the samples to be trained in the data set to be trained and corresponding label information.
Further, in order to realize the recognition operation of the electromyographic signals, firstly, the electromyographic signals to be recognized may be acquired. The electromyographic signal to be identified may be sent by a terminal device, which includes but is not limited to a desktop computer, a tablet computer, and the like.
After acquiring the electromyographic signal to be recognized, inputting the electromyographic signal to be recognized into the corrected electromyographic signal recognition model. The corrected electromyographic signal recognition model can perform gesture recognition operation on the electromyographic signal to be recognized to obtain an output gesture type.
In the electromyographic signal processing method provided by this embodiment, the electromyographic signal to be recognized is recognized by using the electromyographic signal recognition model after the correction data set correction, which includes a plurality of groups of surface electromyographic signals different from the to-be-trained sample collection positions in the to-be-trained data set and the label information corresponding to the surface electromyographic signals, so that the recognition accuracy of the electromyographic signal recognition model can be improved. In addition, the electromyographic signal recognition model to be corrected, which is obtained after the data set to be trained corresponding to the gestures is adopted for training, is corrected, so that a good training effect can be realized only by a small amount of data, the data amount required by the electromyographic signal recognition model is reduced on the basis of improving the accuracy of the electromyographic signal recognition model, and the training efficiency of the electromyographic signal recognition model is improved. The gesture recognition operation is carried out through the corrected electromyographic signal recognition model, and the recognition precision can be improved.
Fig. 3 is a schematic flowchart of an electromyographic signal processing method according to a second embodiment of the disclosure, and on the basis of the first embodiment, as shown in fig. 3, before step 202, the method further includes:
301, acquiring an electromyographic signal recognition model to be corrected, which is trained in advance through a data set to be trained corresponding to a plurality of gestures, and training parameter information corresponding to the electromyographic signal recognition model to be corrected.
In this embodiment, a preset model to be trained is trained by using a data set to be trained corresponding to a plurality of gestures to obtain an electromyographic signal recognition model to be corrected. Acquiring an electromyographic signal recognition model to be corrected after training through a data set to be trained corresponding to a plurality of gestures in advance, and training parameter information corresponding to the electromyographic signal recognition model to be corrected. The training parameter information can be used as an initial training parameter corresponding to the electromyographic signal identification model to be corrected, so that the training efficiency of the model can be further improved.
Specifically, the electromyographic signal recognition model to be corrected, which is trained in advance through the data set to be trained corresponding to the plurality of gestures, and the training parameter information corresponding to the electromyographic signal recognition model to be corrected may be obtained. And acquiring a correction data set, wherein the correction data set comprises a plurality of groups of correction data, and the correction data comprises a surface electromyogram signal and label information corresponding to the surface electromyogram signal. It should be noted that the surface electromyogram signal in the correction data set is different from the surface electromyogram signal acquisition position in the sample to be trained. The collecting channel can be used for collecting surface electromyographic signals after moving a preset distance in a preset direction. The data volume of the correction data set is smaller than the data volume of the data set to be trained.
After the correction data set is obtained, the electromyographic signal recognition model to be corrected is retrained again through the correction data set and training parameter information corresponding to the electromyographic signal recognition model to be corrected, so that the correction operation of the electromyographic signal recognition model to be corrected is realized, and the corrected electromyographic signal recognition model is obtained. Accordingly, after the corrected data set is input to the electromyographic signal recognition model to be corrected, the corrected electromyographic signal recognition model can output the gesture type corresponding to each correction data in the correction data set.
Further, on the basis of the first embodiment, step 301 specifically includes:
the method comprises the steps of obtaining a data set to be trained corresponding to a plurality of gestures, wherein the data set to be trained comprises a plurality of groups of samples to be trained, and the samples to be trained comprise surface electromyographic signals and label information corresponding to the surface electromyographic signals.
And pre-training a preset model to be trained by adopting the data set to be trained to obtain an electromyographic signal recognition model to be corrected and training parameter information corresponding to the electromyographic signal recognition model to be corrected.
In this embodiment, before the electromyographic signal to be corrected is corrected, the electromyographic signal to be corrected is obtained by training.
Specifically, a data set to be trained corresponding to a plurality of gestures may be obtained, the data set to be trained includes a plurality of groups of samples to be trained, and the samples to be trained include surface electromyogram signals and label information corresponding to the surface electromyogram signals. And carrying out iterative training operation on a preset model to be trained through the sample to be trained to obtain a loss value corresponding to the model to be trained. And adjusting the training parameters of the model to be trained according to the loss value until the loss value of the adjusted model to be trained tends to be stable, and obtaining the electromyographic signal identification model to be corrected and training parameter information corresponding to the electromyographic signal identification model to be corrected.
Further, on the basis of the first embodiment, the acquiring a to-be-trained data set corresponding to a plurality of gestures includes:
acquiring myoelectric signals to be processed corresponding to different gestures acquired by a plurality of preset acquisition channels.
And performing active segment detection on the electromyographic signals to be processed to obtain the electromyographic signals corresponding to effective gesture actions.
And labeling the electromyographic signals corresponding to each effective gesture action according to the gesture actions to obtain the data sets to be trained corresponding to the gestures.
In this embodiment, in order to improve the recognition accuracy of the trained myoelectric signal recognition model to be corrected, first, a data set to be trained for training the model may be optimized so that the data set only includes the myoelectric signal and the label information corresponding to the valid gesture.
Specifically, firstly, to-be-processed electromyographic signals corresponding to different gestures collected by a plurality of preset collection channels can be obtained. In the data acquisition process, the electromyographic signals to be processed may have signals of actions irrelevant to gestures, so that the electromyographic signals to be processed can be subjected to activity segment detection to obtain the electromyographic signals corresponding to effective gestures.
And labeling the electromyographic signals corresponding to each effective gesture action according to the gesture actions to obtain the data sets to be trained corresponding to the gestures. Therefore, after the model to be trained is trained through the data set to be trained subsequently, the trained electromyographic signal recognition model to be corrected can have the recognition capability of the electromyographic signal.
Further, on the basis of the first embodiment, step 301 specifically includes:
and pre-training a preset model to be trained by adopting a minimized square and error function and the data set to be trained to obtain an electromyographic signal recognition model to be corrected and training parameter information corresponding to the electromyographic signal recognition model to be corrected.
In this embodiment, a pre-training operation may be performed on a preset to-be-trained model by using a minimized square and error function and a to-be-trained data set, so as to obtain an to-be-corrected electromyographic signal recognition model and training parameter information corresponding to the to-be-corrected electromyographic signal recognition model. Wherein the error function is shown in equation 1:
wherein x isnN is 1, N is data in the input data set to be trained, y (x)nW) is the output prediction parameter of MLP, w is the weight array of neurons, tnAre actual parameters.
In order to improve the training efficiency of the model, in this embodiment, a Levenberg-Marquardt method pair may be specifically adopted to fit the network weight values of the model to be trained.
Fig. 4 is a schematic structural diagram of a model 40 to be trained according to an embodiment of the present disclosure, and as shown in fig. 4, the model 40 to be trained includes an input layer 41, a hidden layer 42, and an output layer 43. Wherein, the hidden layer 42 includes three neurons. The output layer 43 may act as a decoder for finger movements. The sample to be trained is input into the input layer 41, 3 neurons in the hidden layer 42 all use hyperbolic tangent activation functions, and the output layer 43 is decoded output and only contains one output parameter.
According to the electromyographic signal processing method provided by the embodiment, a preset model to be trained is trained by adopting a data set to be trained corresponding to a plurality of gestures, so that an electromyographic signal recognition model to be corrected is obtained. And training the electromyographic signal recognition model to be corrected again through the correction data set and training parameter information corresponding to the electromyographic signal recognition model to be corrected, so as to realize the correction operation of the electromyographic signal recognition model to be corrected and obtain the corrected electromyographic signal recognition model. Therefore, a good training effect can be achieved only by a small amount of data, the data amount required by correction of the electromyographic signal recognition model is reduced on the basis of improving the accuracy of the electromyographic signal recognition model, and the training efficiency of the electromyographic signal recognition model is improved. The gesture recognition operation is carried out through the corrected electromyographic signal recognition model, and the recognition precision can be improved.
Fig. 5 is a schematic flow chart of an electromyographic signal processing method provided in a third embodiment of the present disclosure, where on the basis of any one of the embodiments, step 302 specifically includes:
and 501, controlling the plurality of preset acquisition channels to move a preset number of channel positions in a preset direction to obtain the plurality of moved acquisition channels.
And 502, acquiring electromyographic signals to be processed corresponding to different gestures acquired by the plurality of moved acquisition channels.
In this embodiment, the data in the correction data set may be the surface electromyographic signals acquired after the acquisition channel for acquiring the surface electromyographic signals in the data set to be trained moves a preset distance in a preset direction. Specifically, the preset plurality of acquisition channels may be controlled to move a preset number of channel positions in a preset direction, so as to obtain a plurality of moved acquisition channels. For practical applications, for example, the acquisition positions of 5 channels may be uniformly shifted to the left or right by half a channel position. And performing electromyographic signal acquisition operation through the plurality of moved acquisition channels to obtain electromyographic signals to be processed corresponding to different gestures. And acquiring the correction data set according to the electromyographic signals to be processed corresponding to different gestures acquired by the plurality of moved acquisition channels.
Further, on the basis of any of the above embodiments, step 503 specifically includes:
and performing active segment detection on the electromyographic signals to be processed corresponding to different gestures collected by the plurality of moved collection channels to obtain target electromyographic signals corresponding to effective gesture actions.
And for a target electromyographic signal corresponding to each effective gesture action, labeling the target electromyographic signal according to the gesture action to obtain the correction data set.
In this embodiment, first, the correction data set for the correction model may be optimized to include only the electromyographic signals corresponding to valid gesture actions and the tag information.
Specifically, firstly, to-be-processed electromyographic signals corresponding to different gestures collected by a plurality of preset collection channels can be obtained. In the data acquisition process, signals of actions irrelevant to gestures may exist in the electromyographic signals to be processed corresponding to different gestures acquired by the plurality of moved acquisition channels, so that the electromyographic signals to be processed corresponding to different gestures acquired by the plurality of moved acquisition channels can be subjected to activity segment detection to obtain target electromyographic signals corresponding to effective gestures.
And marking the target electromyographic signals corresponding to each effective gesture action according to the gesture actions to obtain a plurality of correction data sets corresponding to the gestures.
According to the electromyographic signal processing method provided by the embodiment, the acquisition channel for acquiring the surface electromyographic signals in the to-be-trained data set is moved to the preset direction for the preset distance, and then the acquired surface electromyographic signals are acquired, so that the acquisition of the correction data set can be realized, the correction operation on the to-be-corrected electromyographic signal recognition model can be realized, and the recognition accuracy of the electromyographic signal recognition model is improved.
Fig. 6 is a schematic flow chart of an electromyographic signal processing method according to a fourth embodiment of the present disclosure, where on the basis of any of the foregoing embodiments, step 303 specifically includes:
step 601, taking training parameter information corresponding to the electromyographic signal identification model to be corrected as an initial parameter of the electromyographic signal identification model to be corrected, and obtaining the model to be corrected.
In this embodiment, since the electromyographic signal identification model to be corrected already has a certain electromyographic signal identification capability, a fine correction operation can be performed on the electromyographic signal identification model to be corrected on the basis of the electromyographic signal identification model to be corrected. Specifically, training parameter information corresponding to the electromyographic signal identification model to be corrected may be used as an initial parameter of the electromyographic signal identification model to be corrected, so as to obtain the model to be corrected. And training the model to be corrected through the correction data until the model to be corrected is converged, and obtaining the corrected electromyographic signal recognition model.
According to the electromyographic signal processing method provided by the embodiment, training parameter information corresponding to the electromyographic signal identification model to be corrected is used as an initial parameter of the electromyographic signal identification model to be corrected, so that a good training effect can be achieved only by a small amount of data in the following process, and the training efficiency of the electromyographic signal identification model is improved.
Fig. 7 is a schematic structural diagram of an electromyographic signal processing apparatus according to a fifth embodiment of the present disclosure, and as shown in fig. 7, the electromyographic signal processing apparatus 70 includes: the system comprises an acquisition module 71, an input module 72 and a processing module 73, wherein the acquisition module 71 is used for acquiring the electromyographic signals to be identified. An input module 72, configured to input the electromyographic signal to be recognized into the corrected electromyographic signal recognition model. And the processing module 73 is configured to recognize a gesture corresponding to the to-be-recognized electromyographic signal through the corrected electromyographic signal recognition model, and output a gesture type. The corrected electromyographic signal recognition model is obtained by correcting the electromyographic signal recognition model through a correction data set and training parameter information corresponding to the electromyographic signal recognition model to be corrected, the electromyographic signal recognition model to be corrected is obtained after training through a data set to be trained corresponding to a plurality of gestures, and the correction data set comprises a plurality of groups of surface electromyographic signals different from the acquisition positions of samples to be trained in the data set to be trained and label information corresponding to the surface electromyographic signals.
According to the electromyographic signal processing device provided by the embodiment, the electromyographic signal to be recognized is recognized by the electromyographic signal recognition model which is corrected by the correction data set and comprises a plurality of groups of surface electromyographic signals different from the to-be-trained sample acquisition positions in the to-be-trained data set and the corresponding label information, so that the recognition accuracy of the electromyographic signal recognition model can be improved. In addition, the electromyographic signal recognition model to be corrected, which is obtained after the data set to be trained corresponding to the gestures is adopted for training, is corrected, so that a good training effect can be realized only by a small amount of data, the data amount required by the electromyographic signal recognition model is reduced on the basis of improving the accuracy of the electromyographic signal recognition model, and the training efficiency of the electromyographic signal recognition model is improved.
Further, on the basis of the fifth embodiment, the apparatus further includes: and a training module. The acquisition module is further used for acquiring an electromyographic signal recognition model to be corrected, which is trained in advance through a data set to be trained corresponding to a plurality of gestures, and training parameter information corresponding to the electromyographic signal recognition model to be corrected; acquiring a correction data set, wherein the correction data set comprises a plurality of groups of surface electromyographic signals different from the acquisition positions of the to-be-trained samples in the to-be-trained data set and corresponding label information, and the data volume of the correction data set is smaller than that of the to-be-trained data set. The training module is used for carrying out correction operation on the electromyographic signal identification model to be corrected through the correction data set and training parameter information corresponding to the electromyographic signal identification model to be corrected to obtain a corrected electromyographic signal identification model; and outputting the gesture type corresponding to each correction data in the correction data set through the corrected electromyographic signal recognition model.
Further, on the basis of the fifth embodiment, the obtaining module is further configured to obtain a data set to be trained corresponding to the plurality of gestures, the data set to be trained includes a plurality of groups of samples to be trained, and the samples to be trained includes surface electromyographic signals and label information corresponding to the surface electromyographic signals. And pre-training a preset model to be trained by adopting the data set to be trained to obtain an electromyographic signal recognition model to be corrected and training parameter information corresponding to the electromyographic signal recognition model to be corrected.
Further, on the basis of the fifth embodiment, the acquiring module is further configured to acquire to-be-processed electromyographic signals corresponding to different gestures acquired by a plurality of preset acquisition channels. And performing active segment detection on the electromyographic signals to be processed to obtain the electromyographic signals corresponding to effective gesture actions. And labeling the electromyographic signals corresponding to each effective gesture action according to the gesture actions to obtain the data sets to be trained corresponding to the gestures.
Further, on the basis of the fifth embodiment, the training module is configured to perform pre-training operation on a preset model to be trained by using a minimized square and error function and the data set to be trained, so as to obtain an electromyographic signal identification model to be corrected and training parameter information corresponding to the electromyographic signal identification model to be corrected.
Further, on the basis of any of the above embodiments, the obtaining module is further configured to control the preset multiple collecting channels to move to a preset direction by a preset number of channel positions, so as to obtain the moved multiple collecting channels. And acquiring myoelectric signals to be processed corresponding to different gestures acquired by the plurality of moved acquisition channels. And acquiring the correction data set according to the electromyographic signals to be processed, which are acquired by the plurality of moved acquisition channels and correspond to different gestures.
Further, on the basis of any of the above embodiments, the obtaining module is further configured to perform activity segment detection on the electromyographic signals to be processed, which are acquired by the plurality of moved acquisition channels and correspond to different gestures, so as to obtain target electromyographic signals corresponding to effective gesture actions;
and for a target electromyographic signal corresponding to each effective gesture action, labeling the target electromyographic signal according to the gesture action to obtain the correction data set.
Further, on the basis of any of the above embodiments, the training module is configured to: and taking training parameter information corresponding to the electromyographic signal identification model to be corrected as an initial parameter of the electromyographic signal identification model to be corrected to obtain the model to be corrected. And training the model to be corrected through the correction data until the model to be corrected is converged, and obtaining the corrected electromyographic signal recognition model.
Fig. 8 is a schematic structural diagram of an electronic device according to a sixth embodiment of the present disclosure, as shown in fig. 8, the electronic device may be a computer, a tablet device, a medical device, an exercise device, a personal digital assistant, or the like.
The apparatus 800 may include one or more of the following components: a processing component 802, a memory 804, a power component 806, a multimedia component 808, an audio component 810, an input/output (I/O) interface 812, a sensor component 814, and a communication component 816.
The processing component 802 generally controls overall operation of the device 800, such as operations associated with display, telephone calls, data communications, camera operations, and recording operations. The processing components 802 may include one or more processors 820 to execute instructions to perform all or a portion of the steps of the methods described above. Further, the processing component 802 can include one or more modules that facilitate interaction between the processing component 802 and other components. For example, the processing component 802 can include a multimedia module to facilitate interaction between the multimedia component 808 and the processing component 802.
The memory 804 is configured to store various types of data to support operations at the apparatus 800. Examples of such data include instructions for any application or method operating on device 800, contact data, phonebook data, messages, pictures, videos, and so forth. The memory 804 may be implemented by any type or combination of volatile or non-volatile memory devices such as Static Random Access Memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic or optical disks.
The multimedia component 808 includes a screen that provides an output interface between the device 800 and a user. In some embodiments, the screen may include a Liquid Crystal Display (LCD) and a Touch Panel (TP). If the screen includes a touch panel, the screen may be implemented as a touch screen to receive an input signal from a user. The touch panel includes one or more touch sensors to sense touch, slide, and gestures on the touch panel. The touch sensor may not only sense the boundary of a touch or slide action, but also detect the duration and pressure associated with the touch or slide operation. In some embodiments, the multimedia component 808 includes a front facing camera and/or a rear facing camera. The front camera and/or the rear camera may receive external multimedia data when the device 800 is in an operating mode, such as a shooting mode or a video mode. Each front camera and rear camera may be a fixed optical lens system or have a focal length and optical zoom capability.
The audio component 810 is configured to output and/or input audio signals. For example, the audio component 810 includes a Microphone (MIC) configured to receive external audio signals when the apparatus 800 is in an operational mode, such as a call mode, a recording mode, and a voice recognition mode. The received audio signals may further be stored in the memory 804 or transmitted via the communication component 816. In some embodiments, audio component 810 also includes a speaker for outputting audio signals.
The I/O interface 812 provides an interface between the processing component 802 and peripheral interface modules, which may be keyboards, click wheels, buttons, etc. These buttons may include, but are not limited to: a home button, a volume button, a start button, and a lock button.
The sensor assembly 814 includes one or more sensors for providing various aspects of state assessment for the device 800. For example, the sensor assembly 814 may detect the open/closed status of the device 800, the relative positioning of components, such as a display and keypad of the device 800, the sensor assembly 814 may also detect a change in the position of the device 800 or a component of the device 800, the presence or absence of user contact with the device 800, the orientation or acceleration/deceleration of the device 800, and a change in the temperature of the device 800. Sensor assembly 814 may include a proximity sensor configured to detect the presence of a nearby object without any physical contact. The sensor assembly 814 may also include a light sensor, such as a CMOS or CCD image sensor, for use in imaging applications. In some embodiments, the sensor assembly 814 may also include an acceleration sensor, a gyroscope sensor, a magnetic sensor, a pressure sensor, or a temperature sensor.
The communication component 816 is configured to facilitate communications between the apparatus 800 and other devices in a wired or wireless manner. The device 800 may access a wireless network based on a communication standard, such as WiFi, 2G or 3G, or a combination thereof. In an exemplary embodiment, the communication component 816 receives a broadcast signal or broadcast related information from an external broadcast management system via a broadcast channel. In an exemplary embodiment, the communication component 816 further includes a Near Field Communication (NFC) module to facilitate short-range communications. For example, the NFC module may be implemented based on Radio Frequency Identification (RFID) technology, infrared data association (IrDA) technology, Ultra Wideband (UWB) technology, Bluetooth (BT) technology, and other technologies.
In an exemplary embodiment, the apparatus 800 may be implemented by one or more Application Specific Integrated Circuits (ASICs), Digital Signal Processors (DSPs), Digital Signal Processing Devices (DSPDs), Programmable Logic Devices (PLDs), Field Programmable Gate Arrays (FPGAs), controllers, micro-controllers, microprocessors or other electronic components for performing the above-described methods.
In an exemplary embodiment, a non-transitory computer-readable storage medium comprising instructions, such as the memory 804 comprising instructions, executable by the processor 820 of the device 800 to perform the above-described method is also provided. For example, the non-transitory computer readable storage medium may be a ROM, a Random Access Memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, and the like.
A non-transitory computer-readable storage medium, in which instructions, when executed by a processor of a terminal device, enable the terminal device to perform the electromyogram signal processing method of the terminal device.
Another embodiment of the present disclosure further provides a computer program product, which includes a computer program, and when the computer program is executed by a processor, the computer program can implement the electromyographic signal processing method according to any one of the above embodiments.
Other embodiments of the disclosure will be apparent to those skilled in the art from consideration of the specification and practice of the disclosure disclosed herein. This disclosure is intended to cover any variations, uses, or adaptations of the disclosure following, in general, the principles of the disclosure and including such departures from the present disclosure as come within known or customary practice within the art to which the disclosure pertains. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the disclosure being indicated by the following claims.
It will be understood that the present disclosure is not limited to the precise arrangements described above and shown in the drawings and that various modifications and changes may be made without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.
Claims (12)
1. An electromyographic signal processing method, comprising:
acquiring an electromyographic signal to be identified;
inputting the electromyographic signals to be recognized into a corrected electromyographic signal recognition model;
recognizing a gesture corresponding to an electromyographic signal to be recognized through the corrected electromyographic signal recognition model, and outputting a gesture type;
the corrected electromyographic signal recognition model is obtained by correcting the electromyographic signal recognition model through a correction data set and training parameter information corresponding to the electromyographic signal recognition model to be corrected, the electromyographic signal recognition model to be corrected is obtained after training through a data set to be trained corresponding to a plurality of gestures, and the correction data set comprises a plurality of groups of surface electromyographic signals different from the acquisition positions of samples to be trained in the data set to be trained and label information corresponding to the surface electromyographic signals.
2. The method according to claim 1, characterized in that before inputting the electromyographic signal to be recognized into the corrected electromyographic signal recognition model, the method further comprises:
acquiring an electromyographic signal recognition model to be corrected, which is trained in advance through a data set to be trained corresponding to a plurality of gestures, and training parameter information corresponding to the electromyographic signal recognition model to be corrected;
acquiring a correction data set, wherein the correction data set comprises a plurality of groups of surface electromyographic signals different from the acquisition positions of to-be-trained samples in the to-be-trained data set and corresponding label information, and the data volume of the correction data set is smaller than that of the to-be-trained data set;
correcting the electromyographic signal recognition model to be corrected through the correction data set and training parameter information corresponding to the electromyographic signal recognition model to be corrected to obtain a corrected electromyographic signal recognition model; and outputting the gesture type corresponding to each correction data in the correction data set through the corrected electromyographic signal recognition model.
3. The method according to claim 2, wherein the acquiring of the electromyographic signal recognition model to be corrected trained after being trained by the data set to be trained corresponding to a plurality of gestures in advance and the training parameter information corresponding to the electromyographic signal recognition model to be corrected comprises:
acquiring a data set to be trained corresponding to a plurality of gestures, wherein the data set to be trained comprises a plurality of groups of samples to be trained, and the samples to be trained comprise surface electromyographic signals and label information corresponding to the surface electromyographic signals;
and pre-training a preset model to be trained by adopting the data set to be trained to obtain an electromyographic signal recognition model to be corrected and training parameter information corresponding to the electromyographic signal recognition model to be corrected.
4. The method according to claim 3, wherein the acquiring the data set to be trained corresponding to the plurality of gestures comprises:
acquiring myoelectric signals to be processed corresponding to different gestures acquired by a plurality of preset acquisition channels;
performing active segment detection on the electromyographic signals to be processed to obtain electromyographic signals corresponding to effective gesture actions;
and labeling the electromyographic signals corresponding to each effective gesture action according to the gesture actions to obtain the data sets to be trained corresponding to the gestures.
5. The method according to claim 3, wherein the pre-training operation is performed on a preset model to be trained by using the data set to be trained to obtain an electromyographic signal recognition model to be corrected and training parameter information corresponding to the electromyographic signal recognition model to be corrected, and the method includes:
and pre-training a preset model to be trained by adopting a minimized square and error function and the data set to be trained to obtain an electromyographic signal recognition model to be corrected and training parameter information corresponding to the electromyographic signal recognition model to be corrected.
6. The method of claim 4, wherein said obtaining a correction data set comprises:
controlling the plurality of preset acquisition channels to move a preset number of channel positions in a preset direction to obtain a plurality of moved acquisition channels;
acquiring myoelectric signals to be processed corresponding to different gestures acquired by the plurality of moved acquisition channels;
and acquiring the correction data set according to the electromyographic signals to be processed, which are acquired by the plurality of moved acquisition channels and correspond to different gestures.
7. The method according to claim 6, wherein obtaining the correction data set according to the electromyographic signals to be processed corresponding to different gestures collected by the plurality of collected channels after the movement comprises:
performing active segment detection on electromyographic signals to be processed corresponding to different gestures collected by the plurality of moved collection channels to obtain target electromyographic signals corresponding to effective gesture actions;
and for a target electromyographic signal corresponding to each effective gesture action, labeling the target electromyographic signal according to the gesture action to obtain the correction data set.
8. The method according to any one of claims 2 to 7, wherein the obtaining of the corrected electromyographic signal recognition model by performing a correction operation on the electromyographic signal recognition model to be corrected through the correction data set and training parameter information corresponding to the electromyographic signal recognition model to be corrected comprises:
training parameter information corresponding to the electromyographic signal identification model to be corrected is used as an initial parameter of the electromyographic signal identification model to be corrected, and a model to be corrected is obtained;
and training the model to be corrected through the correction data until the model to be corrected is converged, and obtaining the corrected electromyographic signal recognition model.
9. An electromyographic signal processing apparatus, comprising:
the acquisition module is used for acquiring the electromyographic signals to be identified;
the input module is used for inputting the electromyographic signals to be recognized into the corrected electromyographic signal recognition model;
the processing module is used for identifying a gesture corresponding to an electromyographic signal to be identified through the corrected electromyographic signal identification model and outputting a gesture type;
the corrected electromyographic signal recognition model is obtained by correcting the electromyographic signal recognition model through a correction data set and training parameter information corresponding to the electromyographic signal recognition model to be corrected, the electromyographic signal recognition model to be corrected is obtained after training through a data set to be trained corresponding to a plurality of gestures, and the correction data set comprises a plurality of groups of surface electromyographic signals different from the acquisition positions of samples to be trained in the data set to be trained and label information corresponding to the surface electromyographic signals.
10. An electronic device, comprising: a memory, a processor;
a memory; a memory for storing the processor-executable instructions;
wherein the processor is used for calling the program instructions in the memory to execute the electromyographic signal processing method according to any one of claims 1-8.
11. A computer-readable storage medium, in which computer-executable instructions are stored, which, when executed by a processor, are used to implement the electromyographic signal processing method according to any one of claims 1 to 8.
12. A computer program product, characterized in that it comprises a computer program which, when being executed by a processor, implements an electromyographic signal processing method according to any one of claims 1-8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110019793.1A CN112826516A (en) | 2021-01-07 | 2021-01-07 | Electromyographic signal processing method, device, equipment, readable storage medium and product |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110019793.1A CN112826516A (en) | 2021-01-07 | 2021-01-07 | Electromyographic signal processing method, device, equipment, readable storage medium and product |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112826516A true CN112826516A (en) | 2021-05-25 |
Family
ID=75928322
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110019793.1A Pending CN112826516A (en) | 2021-01-07 | 2021-01-07 | Electromyographic signal processing method, device, equipment, readable storage medium and product |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112826516A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113625882A (en) * | 2021-10-12 | 2021-11-09 | 四川大学 | Myoelectric gesture recognition method based on sparse multichannel correlation characteristics |
| CN114129169A (en) * | 2021-11-22 | 2022-03-04 | 中节能风力发电股份有限公司 | Bioelectric signal data identification method, system, medium, and device |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090327171A1 (en) * | 2008-06-26 | 2009-12-31 | Microsoft Corporation | Recognizing gestures from forearm emg signals |
| CN109085918A (en) * | 2018-06-28 | 2018-12-25 | 天津大学 | Acupuncture needling manipulation training method based on myoelectricity |
| CN110109540A (en) * | 2019-04-23 | 2019-08-09 | 刘简 | Write correcting system and its working method |
| CN111176441A (en) * | 2019-11-27 | 2020-05-19 | 广州雪利昂生物科技有限公司 | Surface myoelectricity-based man-machine interaction training method and device and storage medium |
| CN112132192A (en) * | 2020-09-07 | 2020-12-25 | 北京海益同展信息科技有限公司 | Model training method and device, electronic equipment and storage medium |
-
2021
- 2021-01-07 CN CN202110019793.1A patent/CN112826516A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090327171A1 (en) * | 2008-06-26 | 2009-12-31 | Microsoft Corporation | Recognizing gestures from forearm emg signals |
| CN109085918A (en) * | 2018-06-28 | 2018-12-25 | 天津大学 | Acupuncture needling manipulation training method based on myoelectricity |
| CN110109540A (en) * | 2019-04-23 | 2019-08-09 | 刘简 | Write correcting system and its working method |
| CN111176441A (en) * | 2019-11-27 | 2020-05-19 | 广州雪利昂生物科技有限公司 | Surface myoelectricity-based man-machine interaction training method and device and storage medium |
| CN112132192A (en) * | 2020-09-07 | 2020-12-25 | 北京海益同展信息科技有限公司 | Model training method and device, electronic equipment and storage medium |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113625882A (en) * | 2021-10-12 | 2021-11-09 | 四川大学 | Myoelectric gesture recognition method based on sparse multichannel correlation characteristics |
| CN113625882B (en) * | 2021-10-12 | 2022-06-14 | 四川大学 | Myoelectric gesture recognition method based on sparse multichannel correlation characteristics |
| CN114129169A (en) * | 2021-11-22 | 2022-03-04 | 中节能风力发电股份有限公司 | Bioelectric signal data identification method, system, medium, and device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110084775B (en) | Image processing method and device, electronic equipment and storage medium | |
| CN109614876B (en) | Key point detection method and device, electronic equipment and storage medium | |
| CN108121952B (en) | Face key point positioning method, device, equipment and storage medium | |
| CN107492115B (en) | Target object detection method and device | |
| CN106845398B (en) | Face key point positioning method and device | |
| CN107463903B (en) | Face key point positioning method and device | |
| US11138422B2 (en) | Posture detection method, apparatus and device, and storage medium | |
| EP3933552A1 (en) | Method and device for determining gaze position of user, storage medium, and electronic apparatus | |
| CN108633249B (en) | Physiological signal quality judgment method and device | |
| CN106778531A (en) | Face detection method and device | |
| CN107967459B (en) | Convolution processing method, convolution processing device and storage medium | |
| CN109325908B (en) | Image processing method and device, electronic equipment and storage medium | |
| CN112826516A (en) | Electromyographic signal processing method, device, equipment, readable storage medium and product | |
| WO2022095674A1 (en) | Method and apparatus for operating mobile device | |
| CN105975961A (en) | Human face recognition method, device and terminal | |
| CN110619325A (en) | Text recognition method and device | |
| CN109409382B (en) | Image processing method and device, electronic equipment and storage medium | |
| CN107239758B (en) | Method and device for positioning key points of human face | |
| US20210158031A1 (en) | Gesture Recognition Method, and Electronic Device and Storage Medium | |
| CN113642551A (en) | Nail key point detection method and device, electronic equipment and storage medium | |
| CN114821799A (en) | Motion recognition method, device and equipment based on space-time graph convolutional network | |
| CN112861592B (en) | Training method of image generation model, image processing method and device | |
| CN115686187A (en) | Gesture recognition method and device, electronic equipment and storage medium | |
| CN112784858B (en) | Image data processing method and device and electronic equipment | |
| CN107665340B (en) | Fingerprint identification method and device and electronic equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| CB02 | Change of applicant information |
Address after: 601, 6 / F, building 2, No. 18, Kechuang 11th Street, Daxing District, Beijing, 100176 Applicant after: Jingdong Technology Information Technology Co.,Ltd. Address before: 601, 6 / F, building 2, No. 18, Kechuang 11th Street, Beijing Economic and Technological Development Zone, Beijing 100176 Applicant before: Jingdong Shuke Haiyi Information Technology Co.,Ltd. |
|
| CB02 | Change of applicant information | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210525 |
|
| RJ01 | Rejection of invention patent application after publication |