CN109032352A - Hand signal treating method and apparatus - Google Patents
Hand signal treating method and apparatus Download PDFInfo
- Publication number
- CN109032352A CN109032352A CN201810782693.2A CN201810782693A CN109032352A CN 109032352 A CN109032352 A CN 109032352A CN 201810782693 A CN201810782693 A CN 201810782693A CN 109032352 A CN109032352 A CN 109032352A
- Authority
- CN
- China
- Prior art keywords
- signal
- hand
- hand signal
- saturation
- obtains
- 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.)
- Granted
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
Landscapes
- Engineering & Computer Science (AREA)
- General Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- User Interface Of Digital Computer (AREA)
Abstract
The present invention relates to a kind of hand signal treating method and apparatus, the hand signal processing method, comprising: the hand signal for obtaining input is modified the time span of the hand signal, obtains amendment hand signal;It obtains the amendment hand signal corresponding waveform diagram and the corresponding saturation signal of the saturation value is determined according to the saturation value corresponding time cycle in the waveform diagram;The corresponding signal measurements of the saturation signal are obtained, successively the signal measurements are reconstructed, remove the background noise in saturation signal, obtain output hand signal.Above-mentioned technical proposal obtains glitch-free hand signal by being modified to the hand signal of input, improves the accuracy of the gesture identification in later period.
Description
Technical field
The present invention relates to human-computer interaction technique fields, more particularly to a kind of hand signal treating method and apparatus.
Background technique
With the development of human-computer interaction technology, gesture identification is as an important branch in field of human-computer interaction, due to tool
There is the features such as lively, vivid and intuitive, becomes the important technical of human-computer interaction.
Currently, carry out gesture identification mainly by sensor acquire hand signal after, according to the hand signal of acquisition come
Identify the gesture actions of hand.But the hand signal acquired by sensor, it is easy the interference by background signal, thus shadow
Ring the precision of subsequent gesture identification.
Summary of the invention
Based on this, it is necessary to aiming at the problem that hand signal is interfered by background signal, provide a kind of hand signal processing
Method.
A kind of hand signal processing method, comprising:
The hand signal for obtaining input, is modified the time span of the hand signal, obtains amendment hand signal;
The corresponding waveform diagram of the amendment hand signal is obtained, according to corresponding week time of saturation value in the waveform diagram
Phase determines the corresponding saturation signal of the saturation value;
The corresponding signal measurements of the saturation signal are obtained, successively the signal measurements are reconstructed, removal is full
With the background noise in signal, output hand signal is obtained.
In one of the embodiments, before the step of obtaining the hand signal of input, further includes: obtain the original of input
Beginning hand signal screens the original hand signal, obtains the hand signal according to the activation threshold of setting.
The time span to the hand signal is modified in one of the embodiments, obtains amendment gesture
The step of signal, comprising: according to the time span of setting, extraction and interpolation are carried out to the hand signal, obtain time span
Unified amendment hand signal.
In one of the embodiments, after the step of obtaining time span unified amendment hand signal, further includes:
According to the filter factor of setting, the amendment hand signal is filtered, the amendment gesture letter after being filtered
Number.
It is described in one of the embodiments, to obtain the corresponding signal measurements of the saturation signal, successively to the letter
The step of number measured value is reconstructed, comprising: obtain the saturation signal starting sample time corresponding first signal measurements,
Second signal measured value, saturation signal end sampling time corresponding third signal measurements, fourth signal measured value, with
And the total number of the corresponding signal measurements of the saturation signal;
According to first signal measurements, the second signal measured value, third signal measurements, fourth signal measurement
Corresponding relationship between value and the total number of the signal measurements, it is corresponding to calculate each signal measurements in the saturation signal
Filter value;The saturation signal is reconstructed according to the filter value.
A kind of gesture identification method, comprising:
The hand signal for obtaining input, is modified the time span of the hand signal, obtains amendment hand signal;
The corresponding waveform diagram of the amendment hand signal is obtained, according to corresponding week time of saturation value in the waveform diagram
Phase determines the corresponding saturation signal of the saturation value;
The corresponding signal measurements of the saturation signal are obtained, successively the signal measurements are reconstructed, removal is full
With the background noise in signal, output hand signal is obtained;
The amplitude for obtaining the output hand signal obtains amplitude according to the sequencing of the amplitude corresponding time
Sequentially, and according to the amplitude sequence gesture identification is carried out to the hand signal of the input.
A kind of hand signal processing unit, comprising:
Hand signal correction module carries out the time span of the hand signal for obtaining the hand signal of input
Amendment obtains amendment hand signal;
Saturation signal obtains module, for obtaining the corresponding waveform diagram of the amendment hand signal, according to the waveform diagram
The middle saturation value corresponding time cycle, determine the corresponding saturation signal of the saturation value;
Hand signal reconstructed module, for obtaining the corresponding signal measurements of the saturation signal, successively to the signal
Measured value is reconstructed, and removes the background noise in saturation signal, obtains output hand signal.
A kind of gesture identifying device, comprising:
Hand signal correction module obtains the hand signal of input, is modified to the time span of the hand signal,
Obtain amendment hand signal;
Saturation signal obtains module, obtains the corresponding waveform diagram of the amendment hand signal, is embezzled according to the waveform diagram
The corresponding time cycle, determines the corresponding saturation signal of the saturation value with value;
Hand signal reconstructed module obtains the corresponding signal measurements of the saturation signal, successively to the signal measurement
Value is reconstructed, and removes the background noise in saturation signal, obtains output hand signal;
Gesture recognition module, for obtaining the amplitude of the output hand signal, according to the amplitude corresponding time
Sequencing obtains amplitude sequence, and carries out gesture identification to the hand signal of the input according to the amplitude sequence.
Above-mentioned hand signal treating method and apparatus, after being modified to the hand signal of input, according to modified
The waveform diagram of hand signal filters out background noise existing for modified hand signal, to obtain glitch-free
Hand signal improves the accuracy of the gesture identification in later period.
Detailed description of the invention
Fig. 1 is the flow diagram of an embodiment hand signal processing method;
Fig. 2 is that an embodiment hand signal acquires waveform diagram;
Fig. 3 is that another embodiment hand signal acquires waveform diagram;
Fig. 4 is the schematic flow diagram of an embodiment gesture identification method;
Fig. 5 is the LED distribution schematic diagram of an embodiment infrared sensor;
Fig. 6 is the structural schematic diagram of the hand signal processing unit of an embodiment;
Fig. 7 is the structural schematic diagram of an embodiment gesture identifying device.
Specific embodiment
It is with reference to the accompanying drawing and preferably real for the effect for further illustrating technological means adopted by the present invention and acquirement
Example is applied, to the technical solution of the embodiment of the present invention, carries out clear and complete description.
As shown in FIG. 1, FIG. 1 is the flow diagrams of an embodiment hand signal processing method, comprising:
Step S21 obtains the hand signal of input, is modified, is corrected to the time span of the hand signal
Hand signal.
In above-mentioned steps, hand signal is the data-signal of the expression hand motion acquired by equipment, hand signal
Length indicates that amendment hand signal is obtained after being adjusted to the time of hand signal with signal duration length.
Step S22 obtains the corresponding waveform diagram of the amendment hand signal, corresponding according to saturation value in the waveform diagram
Time cycle determines the corresponding saturation signal of the saturation value.
In above-mentioned steps, waveform diagram is the figure drawn according to the signal strength of amendment hand signal and the duration of signal
Shape, the horizontal axis of waveform diagram indicate the time cycle of signal, and the longitudinal axis of waveform diagram indicates the intensity of signal.Saturation value goes out for waveform diagram
Existing multiple identical amplitudes, for hand signal before not carrying out background noise and filtering out, background noise is superimposed upon gesture letter
On number, so that the sensor of acquisition hand signal returns to multiple identical amplitudes, these multiple amplitudes are then saturation value.
Step S23 obtains the corresponding signal measurements of the saturation signal, successively carries out weight to the signal measurements
Structure removes the background noise in saturation signal, obtains output hand signal.
In above-mentioned steps, signal measurements indicate the strength levels of signal, indicate signal measurement with waveform in waveform diagram
It is worth the changing rule with the time.It is the signal that a segment signal measured value is amplitude in hand signal comprising signal.Signal is surveyed
It is in order to which according to the changing rule of hand signal, the background interference letter of superposition must be gone up originally by constructing hand signal that magnitude, which is reconstructed,
It number is filtered out, to obtain the hand signal of noiseless interference.
Above-mentioned hand signal processing method is believed after being modified to the hand signal of input according to modified gesture
Number waveform diagram, background noise existing for modified hand signal is filtered out, to obtain glitch-free gesture letter
Number, improve the accuracy of later period gesture identification.
In one of the embodiments, before the step of obtaining the hand signal of input, further includes: obtain the original of input
Beginning hand signal screens the original hand signal, obtains the hand signal according to the activation threshold of setting.It is former
Beginning hand signal is the infrared signal of sensor initial acquisition, is indicated with infrared sampled value.Activation threshold is according to Practical Project
Setting judges the whether effective numerical value of original hand signal, when current time infrared sampled value and last moment infrared sampled value
Difference be denoted as difference value greater than activation threshold, then the infrared hand signal acquired is effective;When current time infrared sampled value and
The difference of last moment infrared sampled value is denoted as difference value less than activation threshold, then the infrared hand signal acquired is invalid.
The time span to the hand signal is modified in one of the embodiments, obtains amendment gesture
The step of signal, comprising: according to the time span of setting, extraction and interpolation are carried out to the hand signal, obtain time span
Unified amendment hand signal.Since the mobile speed speed of gesture is different, so that the hand signal screened is unfavorable for spy
Sign is extracted, it is therefore desirable to is modified to the time span of hand signal.The hand signal too short to time span carries out interpolation,
The signal too long to time span is sampled, so that the hand signal that multichannel is different in size, is unified into designated length
Hand signal.
In one of the embodiments, after the step of obtaining time span unified amendment hand signal, further includes:
According to the filter factor of setting, the amendment hand signal is filtered, the amendment gesture letter after being filtered
Number.In the present embodiment, be filtered amendment hand signal using following formula:
xiThe signal sampling value inputted for i-th,For i-th of filter value, k takes the number between 0 to 1;
Wherein,
S is the time span for correcting hand signal, and l is the resampling length for correcting hand signal.
It is described in one of the embodiments, to obtain the corresponding signal measurements of the saturation signal, successively to the letter
The step of number measured value is reconstructed, comprising: obtain the saturation signal starting sample time corresponding first signal measurements,
Second signal measured value, saturation signal end sampling time corresponding third signal measurements, fourth signal measured value, with
And the total number of the corresponding signal measurements of the saturation signal;
According to first signal measurements, the second signal measured value, third signal measurements, fourth signal measurement
Corresponding relationship between value and the total number of the signal measurements, it is corresponding to calculate each signal measurements in the saturation signal
Filter value;The saturation signal is reconstructed according to the filter value.
When hand signal is there are when background interference, background interference amount can directly be superimposed upon the corresponding signal measurement of hand signal
In value, the Amplitude Ration of measurement is bigger without the amplitude of background interference, so that infrared sensor returns to the same saturation value.If by signal
Measured value is depicted as waveform diagram, then the wave crest of original needle pattern becomes the wave crest of flat-top shape in waveform diagram, as shown in Fig. 2, believing
The change of number shape causes the discrimination of subsequent gesture to decline.Therefore background interference on hand signal is believed using second order interpolation formula
It number is filtered out, to rebuild wave crest shape, obtains waveform diagram as shown in Figure 3.
Obtain in one of the embodiments, the saturation signal starting sample time corresponding first signal measurements,
Second signal measured value, saturation signal end sampling time corresponding third signal measurements, fourth signal measured value, with
And the total number of the corresponding signal measurements of the saturation signal;It is surveyed according to first signal measurements, the second signal
Corresponding relationship between magnitude, third signal measurements, fourth signal measured value and the total number of the signal measurements calculates institute
State the corresponding filter value of each signal measurements in saturation signal;The saturation signal is reconstructed according to the filter value.
The second order interpolation formula filtered out using second order interpolation formula to background noise on hand signal is as follows:
Wherein, xiThe signal sampling value inputted for i-th,For i-th of filter value, x0、x1Before entering saturation for waveform
Endpoint, xn-1、xnEndpoint after going out desaturation for waveform.
As shown in figure 4, Fig. 4 is the schematic flow diagram of an embodiment gesture identification method, comprising:
Step S41 obtains the hand signal of input, is modified, is corrected to the time span of the hand signal
Hand signal.
Step S42 obtains the corresponding waveform diagram of the amendment hand signal, corresponding according to saturation value in the waveform diagram
Time cycle determines the corresponding saturation signal of the saturation value.
Step S43 obtains the corresponding signal measurements of the saturation signal, successively carries out weight to the signal measurements
Structure removes the background noise in saturation signal, obtains output hand signal.
Step S44 obtains the amplitude of the output hand signal, according to the sequencing of the amplitude corresponding time,
Amplitude sequence is obtained, and gesture identification is carried out to the hand signal of the input according to the amplitude sequence.
Above-mentioned gesture identification method, after being modified to the hand signal of input, according to modified hand signal
Waveform diagram filters out background noise existing for modified hand signal, after obtaining glitch-free hand signal, then root
Gesture identification is carried out according to the hand signal, to improve the accuracy of gesture identification.
Below by taking infrared sensor acquisition hand signal carries out gesture identification as an example, to the gesture identification method of the present embodiment
It is illustrated.
As shown in figure 5, Fig. 5 is the LED distribution schematic diagram of an embodiment infrared sensor, comprising: LED1, LED2,
LED3, LED4, four infrared LEDs are square to be distributed in four angles of infrared sensor.
When palm or other four infrared sensors of object proximity, since the angle of incision is different, object and four infrared
The distance of sensor is different, and the respective magnitudes of four infrared sensors are also different.For one section of waveform signal, in same a period of time
Point, there are ordinal relations for four channel amplitudes, such as the four-way amplitude x of synchronization1,x2,x3,x4, it is easy to seek four-way
The descending of amplitude arranges.Wherein, a1,a1,a1,a1Corresponding 1 to 4 one group of arrangement, it is suitable to be denoted as amplitude
Sequence, to obtain 4*3*2*1=24 kind amplitude sequence.
When gesture is rotation gesture, palm or other objects are clockwise around the overhead at the center of four infrared sensors
Rotation, gesture successively rotate to LED2, LED3, LED4 from LED1, return the typical clockwise rotation gesture in the overhead LED1.Then
Hand stop the overhead LED1 when, the corresponding channel of amplitude descending be LED1, LED2, LED3, LED4, be denoted as sequential vector [1,2,3,
4];Then hand stop the overhead LED2 when, the corresponding channel of amplitude descending be LED2, LED3, LED4, LED1, be denoted as sequential vector [2,
3,4,1];Then hand stop the overhead LED3 when, the corresponding channel of amplitude descending be LED3, LED4, LED1, LED2, be denoted as sequence to
It measures [3,4,1,2];When then hand stops the overhead LED4, the corresponding channel of amplitude descending is LED4, LED1, LED2, LED3, is denoted as suitable
Sequence vector [4,1,2,3].The available table 1 of sequence described above is recorded, similarly analyzes and rotates available table 2 counterclockwise.
The amplitude sequence list of the rotation gesture of table 1
Rotate gesture | Clockwise | Clockwise | Clockwise | Clockwise |
Largest passages | 1 | 2 | 3 | 4 |
Secondary big channel | 2 | 3 | 4 | 1 |
Secondary passage aisle | 3 | 4 | 1 | 2 |
Smallest passage | 4 | 1 | 2 | 3 |
The amplitude sequence list of the rotation gesture of table 2
Rotate gesture | Counterclockwise | Counterclockwise | Counterclockwise | Counterclockwise |
Largest passages | 1 | 2 | 3 | 4 |
Secondary big channel | 4 | 1 | 2 | 3 |
Secondary passage aisle | 3 | 4 | 1 | 2 |
Smallest passage | 2 | 3 | 4 | 1 |
When gesture is when moving down gesture of gesture of translation, palm is originated from LED1 or LED2, and reaching LED3, LED4 terminates.
Then hand stop the overhead LED1 when, the corresponding channel of amplitude descending be LED1, LED2, LED4, LED3, be denoted as sequential vector [1,2,
4,3];When then hand stops the overhead LED2, the corresponding channel of amplitude descending is LED2, LED1, LED3, LED4, is denoted as sequential vector
[2,1,3,4];When then hand stops the overhead LED3, the corresponding channel of amplitude descending is LED3, LED4, LED2, LED1, is denoted as sequence
Vector [3,4,2,1];When then hand stops the overhead LED4, the corresponding channel of amplitude descending is LED4, LED3, LED1, LED2, is denoted as
Sequential vector [4,3,1,2].Record the available table 3 of sequence described above, similarly analysis move up, move to left and move to right available table 4,
Table 5, table 6.
Table 3 moves down the amplitude sequence list of gesture
Rotate gesture | Starting | Starting | Terminate | Terminate |
Largest passages | 1 | 2 | 3 | 4 |
Secondary big channel | 2 | 1 | 4 | 3 |
Secondary passage aisle | 4 | 3 | 2 | 1 |
Smallest passage | 3 | 4 | 1 | 2 |
Table 4 moves up the amplitude sequence list of gesture
Rotate gesture | Starting | Starting | Terminate | Terminate |
Largest passages | 4 | 3 | 1 | 2 |
Secondary big channel | 3 | 4 | 2 | 1 |
Secondary passage aisle | 1 | 2 | 4 | 3 |
Smallest passage | 2 | 1 | 3 | 4 |
Table 5 moves to left the amplitude sequence list of gesture
Rotate gesture | Starting | Starting | Terminate | Terminate |
Largest passages | 2 | 3 | 4 | 1 |
Secondary big channel | 3 | 2 | 1 | 4 |
Secondary passage aisle | 1 | 4 | 3 | 2 |
Smallest passage | 4 | 1 | 2 | 3 |
Table 6 moves to right the amplitude sequence list of gesture
Rotate gesture | Starting | Starting | Terminate | Terminate |
Largest passages | 4 | 1 | 2 | 3 |
Secondary big channel | 1 | 4 | 3 | 2 |
Secondary passage aisle | 3 | 2 | 1 | 4 |
Smallest passage | 2 | 3 | 4 | 1 |
When gesture is click gesture, palm is since any overhead of LED1 or LED2, LED3, LED4, straight down, then
Click action is completed straight up and is left.The movement of vertical direction does not change the amplitude sequence of four-way, so clicking starting
Amplitude sequence and click end amplitude it is sequentially identical, and the type of amplitude sequence is more pure.Double-click gesture and substantially singly
Hitter's gesture is identical, double-clicks gesture and is equivalent to the lesser click gesture of time interval twice.
It (moves up, move down, move to left, move to right, click, double-click, rotate clockwise, inverse by analyzing to obtain eight kinds of gestures above
Hour hands rotation) amplitude sequence have biggish difference, therefore using amplitude sequence as the feature of gesture classification.
It is explained below with the process of gesture identification:
Firstly, 240 sections of gesture samples of test set, including the suitable reverse rotation of double-click totally eight kinds of gestures are clicked up and down.By institute
There is hand signal by the pretreatment such as segmentation, resampling, filtering, obtains pretreated hand signal.
Then, amplitude sequential vector feature is extracted to pretreated hand signal.The available multiple groups of one segment signal
Sequential vector, multiple groups sequential vector may be irregular different, then carry out mode filtering to sequential vector.Mode filter effect: if preceding
One group of sequential vector is identical with later group sequential vector, different from current group sequential vector, then currently the output of group sequential vector should
At previous group sequential vector, as soon as group that frequency of occurrence is most in three groups of sequential vectors from back to front is taken, if number is identical not
Change current order vector.
Statistics frequency is carried out to sequential vector.Mode is as follows: merging close on identical sequential vector, obtain it is multiple it is adjacent not
Same sequential vector, the frequency for the sequential vector that adds up reject the too low sequential vector of frequency of occurrence, remaining multiple sequential vectors
As sequential vector group.
Count the type sum of sequential vector group.The sum for the different arrangements that type sum is 1234, identical arrangement is not
It computes repeatedly, be up to 24 kinds.The sequential vector of the initial segment and ending segment is intercepted as feature.
Finally, being identified according to sequential vector to gesture.
Whether the amplitude sequence for judging the amplitude sequence of starting and terminating is identical, then judges the class of sequential vector group if they are the same
Whether type sum, which is less than, is clicked type sum threshold value, and the signal determining is at clicking if being less than threshold value.If continuous determine at clicking,
And click interval it is small at fixed time, then modify judgement at double-click.
Whether identical, the detection ordering Vector Groups if not identical of amplitude sequence for judging the amplitude sequence of starting and terminating,
If the signal determining is at translation, and according to correspondence, different tables is divided into there are the sequence of table 3, table 4, table 5 or table 6 in sequence vector
It moves up, move down, move to left, move to right.
Whether the amplitude sequence for judging the amplitude sequence of starting and terminating is identical, then judges the class of sequential vector group if they are the same
Whether type sum is greater than rotation type sum threshold value, if more than threshold value then detection ordering Vector Groups, if in Vector Groups there are table 1 or
The sequence of table 2, the signal determining are divided into clockwise and anticlockwise at rotation according to different tables.Above-mentioned gesture identification method, greatly
The efficiency of gesture identification is improved greatly.
As shown in fig. 6, Fig. 6 is the structural schematic diagram of the hand signal processing unit of an embodiment, comprising:
Hand signal correction module 610, for obtaining the hand signal of input, to the time span of the hand signal into
Row amendment obtains amendment hand signal.
Wherein, hand signal is the data-signal of the expression hand motion acquired by equipment, and the length of hand signal is used
Signal duration length indicates that amendment hand signal is obtained after being adjusted to the time of hand signal.
Saturation signal obtains module 620, for obtaining the corresponding waveform diagram of the amendment hand signal, according to the waveform
The saturation value corresponding time cycle in figure determines the corresponding saturation signal of the saturation value.
Wherein, waveform diagram is the figure drawn according to the signal strength of amendment hand signal and the duration of signal, wave
The horizontal axis of shape figure indicates the time cycle of signal, and the longitudinal axis of waveform diagram indicates the intensity of signal.Saturation value is what waveform diagram occurred
Multiple identical amplitudes, before not carrying out background noise and filtering out, background noise is superimposed upon on hand signal hand signal,
So that the sensor of acquisition hand signal returns to multiple identical amplitudes, these multiple amplitudes are then saturation value.
Hand signal reconstructed module 630, for obtaining the corresponding signal measurements of the saturation signal, successively to the letter
Number measured value is reconstructed, and removes the background noise in saturation signal, obtains output hand signal.
Wherein, signal measurements indicate signal strength levels, in waveform diagram with waveform indicate signal measurements with
The changing rule of time.It is the signal that a segment signal measured value is amplitude in hand signal comprising signal.Signal measurements into
Row reconstruct is in order to which according to the changing rule of hand signal, the background noise that building hand signal must go up superposition originally is carried out
It filters out, to obtain the hand signal of noiseless interference.
Above-mentioned hand signal processing unit is believed after being modified to the hand signal of input according to modified gesture
Number waveform diagram, background noise existing for modified hand signal is filtered out, to obtain glitch-free gesture letter
Number, improve the accuracy of later period gesture identification.
As shown in fig. 7, Fig. 7 is the structural schematic diagram of an embodiment gesture identifying device, comprising:
Hand signal correction module 710 obtains the hand signal of input, repairs to the time span of the hand signal
Just, amendment hand signal is obtained.
Saturation signal obtains module 720, the corresponding waveform diagram of the amendment hand signal is obtained, according in the waveform diagram
The saturation value corresponding time cycle determines the corresponding saturation signal of the saturation value.
Hand signal reconstructed module 730 obtains the corresponding signal measurements of the saturation signal, successively surveys to the signal
Magnitude is reconstructed, and removes the background noise in saturation signal, obtains output hand signal.
Gesture recognition module 740, for obtaining the amplitude of the output hand signal, according to the amplitude corresponding time
Sequencing, obtain amplitude sequence, and gesture identification is carried out according to hand signal of the amplitude sequence to the input.
Above-mentioned gesture identifying device, after being modified to the hand signal of input, according to modified hand signal
Waveform diagram filters out background noise existing for modified hand signal, after obtaining glitch-free hand signal, then root
Gesture identification is carried out according to the hand signal, to improve the accuracy of gesture identification.
Each technical characteristic of embodiment described above can be combined arbitrarily, for simplicity of description, not to above-mentioned reality
It applies all possible combination of each technical characteristic in example to be all described, as long as however, the combination of these technical characteristics is not deposited
In contradiction, all should be considered as described in this specification.
The embodiments described above only express several embodiments of the present invention, and the description thereof is more specific and detailed, but simultaneously
It cannot therefore be construed as limiting the scope of the patent.It should be pointed out that coming for those of ordinary skill in the art
It says, without departing from the inventive concept of the premise, various modifications and improvements can be made, these belong to protection of the invention
Range.Therefore, the scope of protection of the patent of the invention shall be subject to the appended claims.
Claims (10)
1. a kind of hand signal processing method characterized by comprising
The hand signal for obtaining input, is modified the time span of the hand signal, obtains amendment hand signal;
The corresponding waveform diagram of the amendment hand signal is obtained, according to the saturation value corresponding time cycle in the waveform diagram, really
Determine the corresponding saturation signal of the saturation value;
The corresponding signal measurements of the saturation signal are obtained, successively the signal measurements are reconstructed, removal saturation letter
Background noise in number obtains output hand signal.
2. hand signal processing method according to claim 1, which is characterized in that in the step for the hand signal for obtaining input
Before rapid, further includes:
The original hand signal for obtaining input screens the original hand signal, obtains according to the activation threshold of setting
The hand signal.
3. hand signal processing method according to claim 1, which is characterized in that the time to the hand signal
The step of length is modified, and obtains amendment hand signal, comprising:
According to the time span of setting, extraction and interpolation are carried out to the hand signal, obtain the unified amendment hand of time span
Gesture signal.
4. hand signal processing method according to claim 3, which is characterized in that obtaining the unified amendment of time span
After the step of hand signal, further includes:
According to the filter factor of setting, the amendment hand signal is filtered, the amendment hand after being filtered
Gesture signal.
5. hand signal processing method according to claim 1, which is characterized in that the acquisition saturation signal is corresponding
Signal measurements, the step of successively the signal measurements are reconstructed, comprising:
Obtain the saturation signal starting sample time corresponding first signal measurements, second signal measured value, the saturation
Signal end sampling time corresponding third signal measurements, fourth signal measured value and the corresponding letter of the saturation signal
The total number of number measured value;
According to first signal measurements, the second signal measured value, third signal measurements, fourth signal measured value and
Corresponding relationship between the total number of the signal measurements calculates the corresponding filtering of each signal measurements in the saturation signal
Value;
The saturation signal is reconstructed according to the filter value.
6. hand signal processing method according to any one of claims 1 to 4, which is characterized in that the hand signal is red
Outer hand signal.
7. a kind of gesture identification method characterized by comprising
The hand signal for obtaining input, is modified the time span of the hand signal, obtains amendment hand signal;
The corresponding waveform diagram of the amendment hand signal is obtained, according to the saturation value corresponding time cycle in the waveform diagram, really
Determine the corresponding saturation signal of the saturation value;
The corresponding signal measurements of the saturation signal are obtained, successively the signal measurements are reconstructed, removal saturation letter
Background noise in number obtains output hand signal;
The amplitude for obtaining the output hand signal obtains amplitude sequence according to the sequencing of the amplitude corresponding time,
And gesture identification is carried out to the hand signal of the input according to the amplitude sequence.
8. a kind of hand signal processing unit characterized by comprising
Hand signal correction module, for obtain input hand signal, the time span of the hand signal is modified,
Obtain amendment hand signal;
Saturation signal obtains module and is embezzled for obtaining the corresponding waveform diagram of the amendment hand signal according to the waveform diagram
The corresponding time cycle, determines the corresponding saturation signal of the saturation value with value;
Hand signal reconstructed module, for obtaining the corresponding signal measurements of the saturation signal, successively to the signal measurement
Value is reconstructed, and removes the background noise in saturation signal, obtains output hand signal.
9. a kind of gesture identifying device characterized by comprising
Hand signal correction module obtains the hand signal of input, is modified, obtains to the time span of the hand signal
Correct hand signal;
Saturation signal obtains module, the corresponding waveform diagram of the amendment hand signal is obtained, according to saturation value in the waveform diagram
The corresponding time cycle determines the corresponding saturation signal of the saturation value;
Hand signal reconstructed module obtains the corresponding signal measurements of the saturation signal, successively to the signal measurements into
Row reconstruct, removes the background noise in saturation signal, obtains output hand signal;
Gesture recognition module, it is successive according to the amplitude corresponding time for obtaining the amplitude of the output hand signal
Sequentially, amplitude sequence is obtained, and gesture identification is carried out to the hand signal of the input according to the amplitude sequence.
10. a kind of computer equipment including memory, processor and stores the meter that can be run on a memory and on a processor
Calculation machine program, which is characterized in that the processor realizes claim 1 to 6 any gesture letter when executing described program
The step of number processing method and gesture identification method as claimed in claim 7.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810782693.2A CN109032352B (en) | 2018-07-17 | 2018-07-17 | Gesture signal processing method and device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810782693.2A CN109032352B (en) | 2018-07-17 | 2018-07-17 | Gesture signal processing method and device |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109032352A true CN109032352A (en) | 2018-12-18 |
CN109032352B CN109032352B (en) | 2021-07-06 |
Family
ID=64642897
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810782693.2A Active CN109032352B (en) | 2018-07-17 | 2018-07-17 | Gesture signal processing method and device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109032352B (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109799911A (en) * | 2019-02-01 | 2019-05-24 | 奥佳华智能健康科技集团股份有限公司 | A kind of massage armchair gesture control device and control method |
CN110096151A (en) * | 2019-04-30 | 2019-08-06 | 努比亚技术有限公司 | A kind of gesture identification method, equipment and computer readable storage medium |
CN111638715A (en) * | 2020-05-29 | 2020-09-08 | 北京海益同展信息科技有限公司 | Robot control method, robot, electronic device, and storage medium |
CN112728607A (en) * | 2021-01-20 | 2021-04-30 | 宁波方太厨具有限公司 | Smoke interference prevention control method for non-contact control range hood and range hood |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20130162520A1 (en) * | 2011-12-27 | 2013-06-27 | Maxim Integrated Products, Inc. | Gesture detection and compact representation thereof |
CN103677217A (en) * | 2012-08-30 | 2014-03-26 | 深圳市亿思达显示科技有限公司 | 3D space input system |
CN105807935A (en) * | 2016-04-01 | 2016-07-27 | 中国科学技术大学苏州研究院 | Gesture control man-machine interactive system based on WiFi |
CN107992191A (en) * | 2017-10-30 | 2018-05-04 | 捷开通讯(深圳)有限公司 | A kind of electronic equipment and its control method, camera of electronic equipment module |
-
2018
- 2018-07-17 CN CN201810782693.2A patent/CN109032352B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20130162520A1 (en) * | 2011-12-27 | 2013-06-27 | Maxim Integrated Products, Inc. | Gesture detection and compact representation thereof |
CN103677217A (en) * | 2012-08-30 | 2014-03-26 | 深圳市亿思达显示科技有限公司 | 3D space input system |
CN105807935A (en) * | 2016-04-01 | 2016-07-27 | 中国科学技术大学苏州研究院 | Gesture control man-machine interactive system based on WiFi |
CN107992191A (en) * | 2017-10-30 | 2018-05-04 | 捷开通讯(深圳)有限公司 | A kind of electronic equipment and its control method, camera of electronic equipment module |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109799911A (en) * | 2019-02-01 | 2019-05-24 | 奥佳华智能健康科技集团股份有限公司 | A kind of massage armchair gesture control device and control method |
CN110096151A (en) * | 2019-04-30 | 2019-08-06 | 努比亚技术有限公司 | A kind of gesture identification method, equipment and computer readable storage medium |
CN110096151B (en) * | 2019-04-30 | 2023-05-02 | 东芯泰合(广州)科技有限公司 | Gesture recognition method, device and computer readable storage medium |
CN111638715A (en) * | 2020-05-29 | 2020-09-08 | 北京海益同展信息科技有限公司 | Robot control method, robot, electronic device, and storage medium |
CN111638715B (en) * | 2020-05-29 | 2023-09-01 | 京东科技信息技术有限公司 | Robot control method, robot, electronic device, and storage medium |
CN112728607A (en) * | 2021-01-20 | 2021-04-30 | 宁波方太厨具有限公司 | Smoke interference prevention control method for non-contact control range hood and range hood |
CN112728607B (en) * | 2021-01-20 | 2022-04-19 | 宁波方太厨具有限公司 | Smoke interference prevention control method for non-contact control range hood and range hood |
Also Published As
Publication number | Publication date |
---|---|
CN109032352B (en) | 2021-07-06 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109032352A (en) | Hand signal treating method and apparatus | |
CN110032975B (en) | Seismic facies picking method | |
US9268321B2 (en) | Digital tiggering using finite state machines | |
CN110266552A (en) | Method, model training method and the device of Traffic anomaly detection | |
CN107909042B (en) | continuous gesture segmentation recognition method | |
CN107833206A (en) | The accurate extracting method of power line under a kind of complex background | |
CN106771589B (en) | A kind of implementation method of the real-time Waterfall plot of spectrum analysis | |
CN107067031A (en) | A kind of calligraphy posture automatic identifying method based on Wi Fi signals | |
CN107728210A (en) | The determination method and apparatus in road are lacked in multiple instruments gathered data | |
CN109582176A (en) | A kind of touch screen anti-noise method and device | |
CN107192411A (en) | A kind of many telemetry parameter analysis window division methods of spacecraft | |
CN105136317B (en) | Single-point sampling judgement single-photon detector and its sampling decision method | |
CN105116318A (en) | Method for realizing burr detection in logic analyzer | |
CN105954695A (en) | Synchronization-based homogeneous-sensor mutation parameter recognizing method and device | |
CN107407703A (en) | The spectrum analyzer of real-time bandwidth with raising | |
CN114842280A (en) | Automatic micro-seismic signal identification algorithm based on convolutional neural network | |
CN106502459A (en) | A kind of smooth filtering method of capacitance touching control track noise signal | |
CN103095249B (en) | A kind of Median Filter Circuit and method | |
CN108052927A (en) | Gesture processing method and processing device based on video data, computing device | |
CN106802293A (en) | waveform peak detection method and device | |
CN106782711A (en) | Signal extraction and error filtering method and device | |
CN106502457B (en) | A kind of quality evaluating method of capacitance touching control track | |
CN110008205A (en) | A kind of monitoring system redundant data cleaning method | |
CN109614999A (en) | A kind of data processing method, device, equipment and computer readable storage medium | |
CN110231894A (en) | The determination method and apparatus of the motion profile of object |
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 | ||
GR01 | Patent grant | ||
GR01 | Patent grant | ||
CP03 | Change of name, title or address | ||
CP03 | Change of name, title or address |
Address after: 510000 room 306, 3rd floor, 43 Sicheng Road, software park, Tianhe District, Guangzhou City, Guangdong Province Patentee after: GUANGZHOU ZHIYUAN ELECTRONICS Co.,Ltd. Address before: 510665 room 517, 1023 Gaopu Road, Tianhe Software Park, Tianhe District, Guangzhou City, Guangdong Province Patentee before: GUANGZHOU ZHIYUAN ELECTRONICS Co.,Ltd. |