CN107885312A - A kind of gestural control method and device - Google Patents
A kind of gestural control method and device Download PDFInfo
- Publication number
- CN107885312A CN107885312A CN201610862392.1A CN201610862392A CN107885312A CN 107885312 A CN107885312 A CN 107885312A CN 201610862392 A CN201610862392 A CN 201610862392A CN 107885312 A CN107885312 A CN 107885312A
- Authority
- CN
- China
- Prior art keywords
- signal
- pressure signal
- gesture
- pressure
- group
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/01—Input arrangements or combined input and output arrangements for interaction between user and computer
- G06F3/017—Gesture based interaction, e.g. based on a set of recognized hand gestures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/01—Input arrangements or combined input and output arrangements for interaction between user and computer
- G06F3/048—Interaction techniques based on graphical user interfaces [GUI]
- G06F3/0487—Interaction techniques based on graphical user interfaces [GUI] using specific features provided by the input device, e.g. functions controlled by the rotation of a mouse with dual sensing arrangements, or of the nature of the input device, e.g. tap gestures based on pressure sensed by a digitiser
- G06F3/0488—Interaction techniques based on graphical user interfaces [GUI] using specific features provided by the input device, e.g. functions controlled by the rotation of a mouse with dual sensing arrangements, or of the nature of the input device, e.g. tap gestures based on pressure sensed by a digitiser using a touch-screen or digitiser, e.g. input of commands through traced gestures
- G06F3/04883—Interaction techniques based on graphical user interfaces [GUI] using specific features provided by the input device, e.g. functions controlled by the rotation of a mouse with dual sensing arrangements, or of the nature of the input device, e.g. tap gestures based on pressure sensed by a digitiser using a touch-screen or digitiser, e.g. input of commands through traced gestures for inputting data by handwriting, e.g. gesture or text
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2203/00—Indexing scheme relating to G06F3/00 - G06F3/048
- G06F2203/048—Indexing scheme relating to G06F3/048
- G06F2203/04808—Several contacts: gestures triggering a specific function, e.g. scrolling, zooming, right-click, when the user establishes several contacts with the surface simultaneously; e.g. using several fingers or a combination of fingers and pen
Abstract
The invention discloses a kind of gestural control method and device, method includes:Encoder model and decoder model are established, two models are trained:Multigroup pressure signal is obtained, is pre-processed;One group of pressure signal is selected in circulation;Pressure signal is encoded by encoder, obtains code signal;The code signal is decoded by decoder, obtains reconstruction signal;Judge whether reconstruction signal and the difference of this group of pressure signal are more than given threshold;If so, parameters revision then is carried out to encoder and decoder;Gesture control process:One group of pressure signal corresponding to gesture is gathered, and is pre-processed;This group of pressure signal is encoded by encoder model, obtains code signal;Obtain control instruction corresponding to code signal.The present invention solves the problems, such as to be identified based on images of gestures in the prior art existing big to outside condition depended;The accuracy of model training is improved, improves gesture identification accuracy.
Description
Technical field
The invention belongs to technical field of hand gesture recognition, is to be related to a kind of gestural control method and device specifically.
Background technology
In daily life, gesture is a kind of conventional intuitively exchange way naturally, can be in many specific scene tables
Up to special meaning.Compared to the keyboard in traditional man-machine interaction, mouse, remote control etc., gesture can be with more freely, certainly
Right form interacts with machine;Manipulated compared to touching, although touch interaction has provided the user brand-new operating experience,
User has also inevitably been limited in before display, and the transmission of gesture information can be with not limited.With technology
Development, gesture is possibly realized applied to man-machine interaction, gesture identification also turn into industrial quarters research with apply focus.
Traditional Gesture Recognition is normally based on vision realization, and gesture video or image are obtained by camera,
Then the gesture of Land use models identification technology identification user.
But this method based on image recognition have two it is larger the defects of:1. the dependence of pair external environment condition is bigger,
The acquisition of images of gestures needs condition, the users such as good light, background not to be had by strict being limited between camera
There is the scope blocked so that the scene that it is applied is relatively limited;2. needing extra increase camera, it is related to privacy of user exposure
The problem of under picture pick-up device.
The content of the invention
The invention provides a kind of gestural control method, solves above-mentioned technical problem present in prior art.
In order to solve the above technical problems, the present invention is achieved using following technical proposals:
A kind of gestural control method, methods described include:
Encoder model and decoder model are established, and two models are trained, training process is:
(1)Multigroup pressure signal is obtained, and is pre-processed;
(2)One group of pressure signal is selected in circulation;One group of selected pressure signal is encoded by encoder, obtains message in cipher
Number;The code signal is decoded by decoder, obtains reconstruction signal;Judge the difference of reconstruction signal and this group of pressure signal
Whether value is more than given threshold;
If so, parameters revision then is carried out to encoder and decoder;
Gesture control process:
(3)One group of pressure signal corresponding to gesture is gathered, and is pre-processed;
(4)This group of pressure signal is encoded by the encoder model trained, obtains code signal;
(5)Control instruction corresponding to code signal is obtained, and is exported to controlled device.
Further, the step(1)Specifically include:
Multiple pressure sensors are being laid on hand, and multiple pressure sensors are laid in diverse location, gather the pressure of diverse location
Signal, obtain one group of pressure signal;
For same gesture, multi collect pressure signal, multigroup pressure signal is obtained.
Further, in step(1)And step(3)In, pretreatment specifically includes:Windows detecting, filtering, normalization.
A kind of gesture control device, described device include:
Multiple pressure sensors, the pressure signal of the diverse location for gathering hand;
Pretreatment unit, for carrying out pretreatment operation to pressure signal;
Encoder model, for being encoded to pressure signal, obtain code signal;
Decoder model, for being decoded to code signal, obtain reconstruction signal;
Computing unit, for calculating the difference of reconstruction signal and pressure signal;
Judging unit, for judging whether difference is more than given threshold;
Amending unit, for carrying out parameters revision to encoder model, decoder model;
Memory cell, for storing the corresponding relation of code signal and control instruction;
Output unit, for obtaining control instruction corresponding to code signal, and export to controlled device.
Further, pressure sensor is provided with eight, is laid in the diverse location of hand.
Further, the pretreatment unit includes windows detecting unit, filter unit, normalization unit.
Compared with prior art, the advantages and positive effects of the present invention are:The gestural control method and device of the present invention, is adopted
One group of pressure signal corresponding to collecting gesture motion, the encoder model trained is input to after being pre-processed, obtains message in cipher
Number, and control instruction corresponding to code signal is obtained, controlled device is controlled, solves and is based on gesture figure in the prior art
As the problem of existing big to outside condition depended is identified, it is not necessary to set camera, avoid exposing privacy of user;And
The learning training process of encoder and decoder model is a kind of automatic learning process, and human intervention is less, is reduced artificial dry
Incompleteness caused by pre-, improves the accuracy of model training, and then improves the accuracy of gesture identification, realizes to quilt
Control the accurate control of equipment.
After the embodiment of the present invention is read in conjunction with the figure, the other features and advantages of the invention will become more clear
Chu.
Brief description of the drawings
Fig. 1 is the flow chart of one embodiment of gestural control method proposed by the present invention;
Fig. 2 is the flow chart of part steps in Fig. 1;
Fig. 3 is the flow chart of part steps in Fig. 1;
Fig. 4 is the structural representation of one embodiment of gesture control device proposed by the present invention;
Fig. 5 is the structural representation of pretreatment unit in Fig. 4.
Embodiment
In order to make the purpose , technical scheme and advantage of the present invention be clearer, below with reference to drawings and examples,
The present invention is described in further detail.
The gestural control method and device of the present embodiment, pressure signal is gathered by pressure sensor, utilizes encoder mould
Type, specific gesture is identified, and map it onto corresponding control instruction, so as to reach the purpose interacted with equipment.Under
Face, the gestural control method and device are described in detail.
The gestural control method of the present embodiment, following step is specifically included, it is shown in Figure 1.
Step S1:Encoder model and decoder model are established, and model is trained.
The learning training process of two models mainly comprises the steps, shown in Figure 2.
Step S11:Obtain multigroup pressure signal.
Multiple pressure sensors are being laid on hand, and multiple pressure sensors are laid in the diverse location of hand, gather different positions
The pressure signal put, obtain one group of pressure signal.
In the present embodiment, 8 pressure sensors are being laid with hand, i.e. every group of pressure signal, which includes 8 pressure, to be believed
Number.
For different gesture motions, pressure sensor can collect different pressure signals.
For same gesture motion, due to differences such as the angles of gesture motion, the pressure signal collected also has difference
Not, it is therefore desirable to carry out multi collect.Therefore, to each gesture motion, it is required for gathering multigroup pressure signal.
Step S12:Pretreatment.
The pretreatment operations such as windows detecting, filtering, normalization are carried out to pressure signal.Preprocessing process can be found in existing skill
Art, here is omitted.
Step S13:One group of pressure signal is selected in circulation.
Step S14:One group of selected pressure signal is encoded by encoder, obtains code signal.
Step S15:The code signal is decoded by decoder, obtains reconstruction signal.
Step S16:Calculate the difference of reconstruction signal and this group of pressure signal.
Step S17:Judge whether difference is more than given threshold.
If so, then perform step S18.
Step S18:Parameters revision is carried out to encoder and decoder.
So far, encoder model and the training of decoder model learning are completed.
One group of pressure signal is input to the encoder trained, encoder output password signal is defeated by the code signal
Enter to the decoder trained, decoder output reconstruction signal, the difference of the reconstruction signal and pressure signal is less than given threshold.
The code signal that the encoder model trained exports is mapped one by one with corresponding control instruction, and preserved.
Therefore, each gesture motion maps a control instruction.It is the operation of controllable controlled device by gesture motion.
Step S2:Gesture control process.
A gesture motion is made, controlled device is controlled.The main of the process of gesture control comprise the steps,
It is shown in Figure 3.
Step S21:Gather one group of pressure signal corresponding to gesture.
A gesture motion is made, the pressure signal of multiple pressure sensor collection diverse locations, obtains one group of pressure letter
Number.
Step S22:Pretreatment.
The pretreatment operations such as windows detecting, filtering, normalization are carried out to pressure signal.
Step S23:This group of pressure signal is encoded by the encoder model trained, obtains code signal.
Step S24:Control instruction corresponding to obtaining code signal, export to controlled device, control controlled device operation.
Control instruction corresponding to obtaining code signal, that is, identify gesture motion, will be controlled by wired or wireless mode
Instruction processed is transmitted to controlled device, so as to realize the control and interaction to controlled device.
If code signal corresponding to the gesture motion does not get control instruction, illustrate gesture motion mistake, it is necessary to
Again correct gesture is made.
The gestural control method of the present embodiment, one group of pressure signal corresponding to gesture motion is gathered, it is defeated after being pre-processed
Enter to the encoder model trained, obtain code signal, and obtain control instruction corresponding to code signal, controlled device is entered
Row control, solve the problems, such as in the prior art based on images of gestures be identified existing for it is big to outside condition depended, be not required to
Camera is set, avoids exposing privacy of user;And the learning training process of encoder and decoder model is a kind of automatic
Learning process, human intervention is less, incompleteness caused by reducing human intervention, improves the accuracy of model training, enters
And the accuracy of gesture identification is improved, realize the accurate control to controlled device;User interacts with controlled device
When, it is simple direct, without excessive cumbersome action, improve Consumer's Experience.
Based on the design of above-mentioned gestural control method, the present embodiment also proposed a kind of gesture control device, the gesture
Control device mainly includes multiple pressure sensors, encoder model, decoder model, computing unit, judging unit, amendment list
Member, memory cell, output unit, it is shown in Figure 4.
Multiple pressure sensors, the pressure signal of the diverse location for gathering hand.In the present embodiment, pressure sensor
Eight are provided with, is laid in the diverse location of hand.
Pretreatment unit, for carrying out pretreatment operation to pressure signal.Pretreatment unit mainly includes windows detecting list
Member, filter unit, normalization unit etc., it is shown in Figure 5.
Encoder model, for being encoded to pressure signal, obtain code signal.
Decoder model, for being decoded to code signal, obtain reconstruction signal.
Computing unit, for calculating the difference of reconstruction signal and pressure signal.
Judging unit, for judging whether difference is more than given threshold.
Amending unit, for carrying out parameters revision to encoder model, decoder model.
Memory cell, for storing the corresponding relation of code signal and control instruction.
Output unit, for obtaining control instruction corresponding to code signal, and export to controlled device.
The course of work of specific gesture control device, is described in detail in above-mentioned gestural control method, not superfluous herein
State.
The gesture control device of the present embodiment, solve in the prior art based on images of gestures be identified it is existing externally
The problem of portion's condition depended is big, it is not necessary to camera is set, avoids exposing privacy of user;Encoder model and decoder model essence
Degree is high, improves the accuracy of gesture identification, realizes the accurate control to controlled device.
The above embodiments are merely illustrative of the technical solutions of the present invention, rather than is limited;Although with reference to foregoing reality
Example is applied the present invention is described in detail, for the person of ordinary skill of the art, still can be to foregoing implementation
Technical scheme described in example is modified, or carries out equivalent substitution to which part technical characteristic;And these are changed or replaced
Change, the essence of appropriate technical solution is departed from the spirit and scope of claimed technical solution of the invention.
Claims (6)
- A kind of 1. gestural control method, it is characterised in that:Methods described includes:Encoder model and decoder model are established, and two models are trained, training process is:(1)Multigroup pressure signal is obtained, and is pre-processed;(2)One group of pressure signal is selected in circulation;One group of selected pressure signal is encoded by encoder, obtains message in cipher Number;The code signal is decoded by decoder, obtains reconstruction signal;Judge the difference of reconstruction signal and this group of pressure signal Whether value is more than given threshold;If so, parameters revision then is carried out to encoder and decoder;Gesture control process:(3)One group of pressure signal corresponding to gesture is gathered, and is pre-processed;(4)This group of pressure signal is encoded by the encoder model trained, obtains code signal;(5)Control instruction corresponding to code signal is obtained, and is exported to controlled device.
- 2. gestural control method according to claim 1, it is characterised in that:The step(1)Specifically include:Multiple pressure sensors are being laid on hand, and multiple pressure sensors are laid in diverse location, gather the pressure of diverse location Signal, obtain one group of pressure signal;For same gesture, multi collect pressure signal, multigroup pressure signal is obtained.
- 3. gestural control method according to claim 1, it is characterised in that:In step(1)And step(3)In, pretreatment Specifically include:Windows detecting, filtering, normalization.
- A kind of 4. gesture control device, it is characterised in that:Described device includes:Multiple pressure sensors, the pressure signal of the diverse location for gathering hand;Pretreatment unit, for carrying out pretreatment operation to pressure signal;Encoder model, for being encoded to pressure signal, obtain code signal;Decoder model, for being decoded to code signal, obtain reconstruction signal;Computing unit, for calculating the difference of reconstruction signal and pressure signal;Judging unit, for judging whether difference is more than given threshold;Amending unit, for carrying out parameters revision to encoder model, decoder model;Memory cell, for storing the corresponding relation of code signal and control instruction;Output unit, for obtaining control instruction corresponding to code signal, and export to controlled device.
- 5. gesture identifying device according to claim 4, it is characterised in that:Pressure sensor is provided with eight, is laid in The diverse location of hand.
- 6. gesture identifying device according to claim 4, it is characterised in that:The pretreatment unit includes windows detecting list Member, filter unit, normalization unit.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610862392.1A CN107885312A (en) | 2016-09-29 | 2016-09-29 | A kind of gestural control method and device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610862392.1A CN107885312A (en) | 2016-09-29 | 2016-09-29 | A kind of gestural control method and device |
Publications (1)
Publication Number | Publication Date |
---|---|
CN107885312A true CN107885312A (en) | 2018-04-06 |
Family
ID=61769075
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610862392.1A Pending CN107885312A (en) | 2016-09-29 | 2016-09-29 | A kind of gestural control method and device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107885312A (en) |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103941860A (en) * | 2014-03-31 | 2014-07-23 | 天津三星通信技术研究有限公司 | Gesture recognition system of wrist strap type portable terminal and method of gesture recognition system |
CN104679229A (en) * | 2013-11-27 | 2015-06-03 | 中国移动通信集团公司 | Gesture recognition method and apparatus |
CN105335713A (en) * | 2015-10-28 | 2016-02-17 | 小米科技有限责任公司 | Fingerprint identification method and device |
CN105334958A (en) * | 2015-09-11 | 2016-02-17 | 南京西西弗信息科技有限公司 | Gesture recognition system and realization method |
-
2016
- 2016-09-29 CN CN201610862392.1A patent/CN107885312A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104679229A (en) * | 2013-11-27 | 2015-06-03 | 中国移动通信集团公司 | Gesture recognition method and apparatus |
CN103941860A (en) * | 2014-03-31 | 2014-07-23 | 天津三星通信技术研究有限公司 | Gesture recognition system of wrist strap type portable terminal and method of gesture recognition system |
CN105334958A (en) * | 2015-09-11 | 2016-02-17 | 南京西西弗信息科技有限公司 | Gesture recognition system and realization method |
CN105335713A (en) * | 2015-10-28 | 2016-02-17 | 小米科技有限责任公司 | Fingerprint identification method and device |
Non-Patent Citations (1)
Title |
---|
韩伟 等: "深度学习理论及其应用专题讲座(二) 第3讲 深度学习中的经典网络模型及训练方法", 《军事通信技术》 * |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108983979B (en) | Gesture tracking recognition method and device and intelligent equipment | |
CN102193633B (en) | dynamic sign language recognition method for data glove | |
CN110765854B (en) | Video motion recognition method | |
CN103345644B (en) | The object detection method of on-line training and device | |
CN106547356B (en) | Intelligent interaction method and device | |
CN110796018B (en) | Hand motion recognition method based on depth image and color image | |
CN108921942B (en) | Method and device for 2D (two-dimensional) conversion of image into 3D (three-dimensional) | |
CN104616002A (en) | Facial recognition equipment used for judging age groups | |
CN105912142A (en) | Step recording and behavior identification method based on acceleration sensor | |
CN206379992U (en) | A kind of skin quality testing and analysis system | |
CN107944420A (en) | The photo-irradiation treatment method and apparatus of facial image | |
CN103472915A (en) | Reading control method and reading control device on basis of pupil tracking and display equipment | |
CN104777910A (en) | Method and system for applying expression recognition to display device | |
CN205644515U (en) | Intelligence electron frame | |
CN102982015A (en) | Method of producing electronic courseware by utilizing electronic whiteboard and corresponding display method | |
CN116524195A (en) | Semantic segmentation method, semantic segmentation device, electronic equipment and storage medium | |
CN113724132A (en) | Image style migration processing method and device, electronic equipment and storage medium | |
CN110827355B (en) | Moving target rapid positioning method and system based on video image coordinates | |
CN107885312A (en) | A kind of gestural control method and device | |
CN117008491A (en) | Intelligent gesture control system and method | |
CN110223258A (en) | A kind of multi-mode fast video image defogging method and device | |
CN113916899B (en) | Method, system and device for detecting large transfusion soft bag product based on visual identification | |
CN206212381U (en) | Interior illumination control system based on virtual reality device use state | |
CN115690592A (en) | Image processing method and model training method | |
CN112822393B (en) | Image processing 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 | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180406 |