CN105204611A - Finger movement mode based Morse code character inputting system and method - Google Patents
Finger movement mode based Morse code character inputting system and method Download PDFInfo
- Publication number
- CN105204611A CN105204611A CN201410289073.7A CN201410289073A CN105204611A CN 105204611 A CN105204611 A CN 105204611A CN 201410289073 A CN201410289073 A CN 201410289073A CN 105204611 A CN105204611 A CN 105204611A
- Authority
- CN
- China
- Prior art keywords
- signal
- acceleration
- morse code
- contact
- finger
- 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
- 238000000034 method Methods 0.000 title claims abstract description 50
- 230000005057 finger movement Effects 0.000 title abstract description 3
- 230000001133 acceleration Effects 0.000 claims abstract description 113
- 210000003811 finger Anatomy 0.000 claims abstract description 97
- 210000003813 thumb Anatomy 0.000 claims abstract description 80
- 210000000707 wrist Anatomy 0.000 claims abstract description 25
- 238000000605 extraction Methods 0.000 claims abstract description 20
- 230000009471 action Effects 0.000 claims abstract description 19
- 230000004927 fusion Effects 0.000 claims abstract description 12
- 238000012545 processing Methods 0.000 claims abstract description 11
- 230000033001 locomotion Effects 0.000 claims description 53
- 230000008569 process Effects 0.000 claims description 25
- 239000000203 mixture Substances 0.000 claims description 22
- 230000008859 change Effects 0.000 claims description 21
- 230000011218 segmentation Effects 0.000 claims description 16
- 238000001914 filtration Methods 0.000 claims description 11
- 238000006243 chemical reaction Methods 0.000 claims description 9
- 210000003205 muscle Anatomy 0.000 claims description 9
- 230000005540 biological transmission Effects 0.000 claims description 8
- 230000009467 reduction Effects 0.000 claims description 7
- 230000009466 transformation Effects 0.000 claims description 2
- 239000000523 sample Substances 0.000 description 7
- 238000006073 displacement reaction Methods 0.000 description 6
- 238000004891 communication Methods 0.000 description 5
- 238000001514 detection method Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 230000002159 abnormal effect Effects 0.000 description 2
- 210000001145 finger joint Anatomy 0.000 description 2
- 238000012905 input function Methods 0.000 description 2
- 230000002093 peripheral effect Effects 0.000 description 2
- 238000005070 sampling Methods 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- YKGYQYOQRGPFTO-UHFFFAOYSA-N bis(8-methylnonyl) hexanedioate Chemical compound CC(C)CCCCCCCOC(=O)CCCCC(=O)OCCCCCCCC(C)C YKGYQYOQRGPFTO-UHFFFAOYSA-N 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 239000004615 ingredient Substances 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 230000004118 muscle contraction Effects 0.000 description 1
Landscapes
- User Interface Of Digital Computer (AREA)
Abstract
The invention provides a finger movement mode based Morse code character inputting system and method. Gesture signals are respectively obtained through a bioelectrical sensor arranged on the wrist and an acceleration sensor arranged on the thumb, then contact actions and contact time of fingers are obtained through signal processing, feature extraction and feature fusion, three actions of long-contact, short-contact and finger loosening are respectively mapped into tick, tock and blank space in a basically-formed unit of Morse codes, then corresponding characters are obtained according to character codes of the Morse codes, and accordingly Morse code input is achieved. The problem that a Morse code inputting device in the prior art needs a special device and is inconvenient to carry is solved. Due to the fact that the simple wearable sensors are arranged on the wrist and the thumb in the scheme, the device is very convenient to carry, does not need other complicated structures and is convenient to carry and use.
Description
Technical field
The present invention relates to a kind of virtual input device, specifically a kind of Morse code character input system based on, finger motion pattern and method.
Background technology
The language of telegraph communication is made up of code sign.Telegraph communication is invented by the Morse of the U.S. the earliest, so code sign is also called Morse code (Morsecode).Code sign is made up of two kinds of baseband signals and different interval times: very brief some signal ". ", read " " (Di); Keep the long letter number "-" of certain hour, read " answering-" (Da).Be widely used in telegraph communication in wartime in early days, reporter does not stop by individual facing to sending box, be contact according to certain rule " striker " and " needle plate " (this is both vivid word) by sending box in fact, thus produce long or short electric signal.The sending box of reciever this electric signal can be translated into into business signal such as above DIDA.And operator converts cipher graphic to according to sound, contrast cipher table is translating into word.And the person of transmitting messages carries out is exactly inverse operations.Morse code defines and comprises: English alphabet A-Z (distinguishing without capital and small letter) tens digit 0-9, and "? " "/" " () " "-" ". ", still has a lot of local in use so far.
Morse code is made up of two kinds of baseband signals and different interval times: very brief some signal " ", read " " (Di); Keep the long letter "-" of certain hour, read " answering " (Da).Interval time: drip, 1t; Answer, 3t; Ticking, 1t; Between letter, 3t; Between word, 5t.Generation history Morse code is the earliest some numeral points and draws.The corresponding word of numeral, needs to search a code table and just can know the number that each word is corresponding.Point can be knocked out with a telegraph key, draw and the pause of centre.Morse code is very important on radio in early days, and each radio communication person institute must must know.
Morse code needs special input equipment when inputting, as the equipment needed on the table or in user's hand, placement one is special inputs, as telegraph, PC, mobile phone etc.As disclosed a kind of radio Morse code transmitting-receiving training aids in Chinese patent literature CN201946145U, there is display, reservoir, mainboard, keyboard, synaeresis device and loudspeaker, display is connected with mainboard, reservoir is connected on mainboard, and keyboard is connected on mainboard, and mainboard is connected with synaeresis device, the program can carry out reception and the dispatch training of Morse code, this equipment is inputted Morse code by keyboard, needs the keyboard equipment by outside, not portable.
Summary of the invention
For this reason, technical matters to be solved by this invention is the problem that Morse code input equipment of the prior art is not easily carried with, and provides a kind of and can carry with, the portable Morse code character input system based on finger motion pattern and method.
For solving the problems of the technologies described above, of the present inventionly provide a kind of Morse code character input system based on finger motion pattern and method.
Based on a Morse code character input system for finger motion pattern, comprising:
Signal gathering unit, comprise the biopotential sensor being arranged on wrist and the acceleration transducer be arranged on finger, described biopotential sensor gathers bio-electrical information during user's gesture motion, and described acceleration transducer gathers movement locus and the acceleration of thumb;
Signal Pretreatment unit, the bioelectrical signals obtain described signal gathering unit and acceleration signal carry out noise reduction filtering process, and carry out analog to digital conversion;
Signal segmentation unit, carries out dividing processing by the bioelectrical signals after Signal Pretreatment cell processing and acceleration signal, obtains gesture active segment;
Feature extraction unit, carries out feature extraction for the bioelectrical signals in gesture active segment and acceleration signal respectively;
Fusion Features analytic unit, is undertaken merging and analyzing by the eigenwert of described bioelectrical signals and acceleration signal, identify thumb and other contact actions pointed and duration of contact length, be converted into the basic composition unit of corresponding Morse code;
Symbol maps unit, according to the corresponding relation of Morse code character and Morse code basic composition unit, obtains gesture motion and does corresponding character.
Preferably, also comprise transmission unit, by wireless transmission method, the character after identifying is sent to external unit.
Preferably, described biopotential sensor is placed on the muscle group surface of user's wrist, obtains the bioelectrical signals of user gesture motion, and described bioelectrical signals comprises the impedance transformation signal of electromyographic signal under skin and skin surface.
Based on a Morse code characters input method for finger motion pattern, comprise the steps:
Be arranged on bio-electrical information during biopotential sensor collection user's gesture motion of wrist, the acceleration transducer be arranged on thumb gathers movement locus and the acceleration of thumb;
Noise reduction filtering process is carried out to described bioelectrical signals and acceleration signal, and carries out analog to digital conversion;
Bioelectrical signals after process and acceleration signal are carried out dividing processing, obtains gesture active segment; Feature extraction is carried out respectively for the bioelectrical signals in gesture active segment and acceleration signal;
The eigenwert of described bioelectrical signals and acceleration signal is carried out merging and analyzing, identify contact action between finger and duration of contact length, be converted into the basic composition unit of corresponding Morse code;
Symbol maps unit, according to the corresponding relation of Morse code character and Morse code basic composition unit, obtains gesture motion and does corresponding character.
Preferably, the process of described " bioelectrical signals after process and acceleration signal are carried out dividing processing, obtains gesture active segment ", comprising:
By bioelectrical signals, first obtained the baseline of signal by average filter, then use original signal to deduct baseline, be eliminated the stationary signal of low frequency wonder; Then, service time, window obtained the window self-energy of bioelectrical signals; Afterwards this energy magnitude is normalized, absolute figure by signal becomes the relative value between 0-1, a threshold value is set in 0-1, if signal energy has several sampled points continuously higher than this threshold value, then think that signal is the starting point of gesture active segment herein, after this be that active segment is inner, until several sampled points lower than threshold value, then think that signal is the terminal of gesture active segment herein continuously.
Preferably, the process of described " carrying out feature extraction for the bioelectrical signals in gesture active segment and acceleration signal respectively ", comprising:
By the bioelectrical signals in an active segment, by the energy magnitude of each sensor composition multi-C vector, as bioelectricity eigenwert;
By the acceleration signal in an active segment, calculate the change of the acceleration on three-dimensional respectively, as acceleration signature value.
Preferably, described " eigenwert of described bioelectrical signals and acceleration signal is carried out merging and analyzing, identify contact action between finger and duration of contact length, be converted into the basic composition unit of corresponding Morse code " process, comprising:
Judge it is contact or separate between finger according to bioelectricity eigenwert;
The length of finger Contact time is judged according to acceleration signature value.
Preferably, judge that the process of length pointing the Contact time comprises according to acceleration signature value: what acceleration change was fast is short contact, slow the contacting for long of acceleration change.
Preferably, described in when being converted into the basic composition unit of corresponding Morse code, comprise
Short contact, long contact and finger unclamp the code in short-term of corresponding Morse code respectively, long time-code and space.
Preferably, the process by wireless transmission method, the character after identification being sent to external unit is also comprised.
Technical scheme of the present invention has the following advantages compared to existing technology,
(1) the invention provides a kind of Morse code character input system based on finger motion pattern, comprise signal gathering unit, Signal Pretreatment unit, signal segmentation unit, feature extraction unit, Fusion Features analytic unit, symbol maps unit, by being arranged on the biopotential sensor of wrist and being arranged on the acceleration transducer of thumb, obtain hand signal respectively, then signal transacting is passed through, feature extraction, Fusion Features obtains the contact action between finger, and the length of duration of contact, length is contacted, short contact, unclamp between finger three actions be mapped as respectively Morse code substantially make in unit drip, answer, space, corresponding character is obtained again according to the character code of Morse code, thus achieve the input of Morse code.By two sensors at wrist and thumb place in the program, to finger between contact action and duration of contact length identify, thus achieve the input of Morse code, solve Morse code input equipment in prior art and need special device, carry inconvenient word, except being provided with simply wearable sensor at wrist and thumb in the solution of the present invention, carrying of this equipment is made to become abnormal convenient, without the need to other labyrinth, easy to carry and use, and input Morse code by the mode of finger contact, make input mode simpler answer and easy to operate and grasp, the input of Morse code can be used whenever and wherever possible widely.
(2) the Morse code character input system based on finger motion pattern of the present invention, biopotential sensor is responsible for the posture identifying hand, and namely thumb contacts with middle finger or separates.Because these two kinds of gestures, the form of hand is different, so the tensity of involved muscle group is different, thus the combination of different bioelectricity energy is produced at biopotential sensor, the N dimensional vector that the energy magnitude of each sensor is formed, as the eigenwert of classification, judge that thumb contacts with middle finger or separates.
(3) the Morse code character input system based on finger motion pattern of the present invention, is positioned at the acceleration transducer of thumb position, is responsible for the length of the duration of contact identifying thumb and middle finger.Because sensor can identify thumb velocity variations in three axial directions for this reason, and for short contact, its velocity variations is faster, and the velocity variations of long contact is slower.So, by the acceleration change of three axis, as the eigenwert of classification, can judge that thumb and middle finger are long contacts or short contact.
(4) the present invention also provides the input method of the above-mentioned Morse code character input system based on finger motion pattern, comprise and obtain bioelectrical signals and acceleration signal, gesture active segment is identified after pre-service, and feature extraction and fusion is carried out in this active segment, obtain corresponding finger movement, thus be converted into the elementary cell of corresponding Morse code, complete the input function of Morse code character, be suitable in the environment of movement, complete the input to external unit and control, achieve man-machine interaction.
(5) input method of the present invention, when obtaining gesture active segment, service time, window obtained window self-energy, then by arranging threshold value to obtain the new section satisfied condition, which is simple, but can filter out noise etc., the signal segment and other signal segments with gesture motion are effectively distinguished, for subsequent treatment obtains rational data.
(6) input method of the present invention, the eigenwert of the acceleration transducer on the bioelectricity eigenwert of wrist and thumb is merged, just can judge three kinds of gestures, and by these three kinds of gestures, be mapped to the basic composition unit of Morse code, that is: the long contact of thumb middle finger (" clatter "), thumb middle finger short contact (" ticking "), thumb middle finger unclamp in (space).Alternately repeat this three kinds of gestures, then can form different Morse code characters.
Accompanying drawing explanation
In order to make content of the present invention be more likely to be clearly understood, below according to a particular embodiment of the invention and by reference to the accompanying drawings, the present invention is further detailed explanation, wherein
Fig. 1 is the sensor scheme of installation of the Morse code character input system based on finger motion pattern of the present invention;
Fig. 2 is the system architecture schematic diagram of the Morse code character input system based on finger motion pattern of the present invention;
Fig. 3 is the process flow diagram of the input method of the Morse code character input system based on finger motion pattern of the present invention;
Fig. 4 is Morse's code table.
Embodiment
embodiment 1:
Provide the embodiment that of the Morse code character input system based on finger motion pattern of the present invention is concrete below, based on the Morse code character input system of finger motion pattern, signal gathering unit 101, Signal Pretreatment unit 102, signal segmentation unit 103, feature extraction unit 104, Fusion Features unit 105, symbol maps unit 108 should be comprised.The program is by being arranged on the biopotential sensor 1 of wrist and being arranged on the acceleration transducer 2 of thumb, obtain hand signal respectively, then the contact action between pointing and the length of duration of contact is obtained by signal transacting, feature extraction, Fusion Features, substantially the making dripping in unit, answer of Morse code, space is mapped as respectively by unclamping three actions between long contact, short contact, finger, obtain corresponding character according to the character code of Morse code again, thus achieve the input of Morse code.
Fig. 1 gives the parts of this Morse code character input system position palm part based on finger motion pattern, comprises the biopotential sensor 1 of wrist wrist strap inside and is arranged on the acceleration transducer 2 of thumb finger ring inside.Should form and treatment scheme based on the system of the Morse code character input system of finger motion pattern, as shown in Figure 2.Wherein, information acquisition unit 101, Signal Pretreatment unit 102, signal segmentation unit 103, characteristic extracting module 104, all comprise two and be not only parallel to each other, but also have mutual signal transacting stream, i.e. bioelectrical signals stream and acceleration signal stream.Two bars streams, at Fusion Features analysis module 105, merge, and judge to obtain corresponding Morse code elementary cell to the action of finger, then by gesture and character being mapped in symbol maps unit 108, obtain the character of input.Each concrete ingredient is as follows:
(1) signal gathering unit 101, comprise the biopotential sensor 1 being positioned at wrist wrist strap inside and the acceleration transducer 2 being arranged on thumb finger ring inside, as shown in Figure 1, described biopotential sensor 1 gathers bio-electrical information during user's gesture motion, and described acceleration transducer 2 gathers movement locus and the acceleration of thumb.Wherein, the biopotential sensor of wrist is responsible for the posture identifying hand, and namely thumb contacts with middle finger or separates.Because these two kinds of gestures, the form of hand is different, so the tensity of involved muscle group is different, thus the combination of different bioelectricity energy is produced at biopotential sensor, the N dimensional vector that the energy magnitude of each sensor is formed, as the eigenwert of classification, judge that thumb contacts with middle finger or separates.Be positioned at the acceleration transducer of thumb position, acceleration transducer 2 is placed on user's thumb surface, obtain the movement locus of user's thumb on three dimensions, and thumb X, Y, Z tri-straight-line displacement acceleration axially and and angular displacement acceleration, the length of the duration of contact identifying thumb and middle finger (also can be other three fingers except thumb) is responsible for by acceleration transducer.Because sensor can identify thumb velocity variations in three axial directions for this reason, and for short contact, its velocity variations is faster, and the velocity variations of long contact is slower.So, by the acceleration change of three axis, as the eigenwert of classification, can judge that thumb and middle finger are long contacts or short contact.
(2) Signal Pretreatment unit 102, the bioelectrical signals obtain described signal gathering unit and acceleration signal carry out noise reduction filtering process, and carry out analog to digital conversion.
In this Signal Pretreatment unit, needing the original bioelectrical signals to collecting and acceleration signal, carrying out filtering and noise reduction and analog/digital conversion.First, respectively bandpass filtering is carried out to bioelectrical signals, low-pass filtering is carried out to acceleration signal, to eliminate the impact of the factors such as neighbourhood noise.Then with the sampling rate of 500-1000Hz, analog/digital conversion is carried out to two kinds of signals, obtain discrete digital signal samples point.
(3) signal segmentation unit 103, the bioelectrical signals after being processed by Signal Pretreatment unit 102 and acceleration signal carry out dividing processing, obtain gesture active segment.
In this unit, respectively bioelectrical signals and acceleration signal are processed, thus extract independently hand signal.
For bioelectrical signals, first use 5 moving average filters, obtain the baseline of signal; Then deduct baseline by original signal, be eliminated the stationary signal of low frequency wonder.Then, use the time window of fixed width, get the window self-energy of bioelectrical signals.Afterwards, be normalized by this energy magnitude, the absolute figure by signal becomes the relative value between 0-1, then sets a threshold value between 0-1, if signal energy has several sampled points continuously higher than this threshold value, then think that signal is the starting point of gesture motion active segment herein; After this active segment inside is; If have again several sampled points of continuous print lower than threshold value, then think that this number is the terminal of gesture motion active segment herein.Between starting point and terminal, it is a complete gesture of user.
For acceleration signal, also use same based on the detection starting point of threshold value and the logic of terminal.Then, the starting point obtain two kinds of modes and terminal compare and on average, thus complete the segmentation of user's gesture active segment, obtain the signal data in gesture active segment.
(4) feature extraction unit 104, carries out feature extraction for the bioelectrical signals in gesture active segment and acceleration signal respectively, so that identify the action of finger subsequently through Fusion Features.
For bioelectrical signals, when feature extraction, by the bioelectrical signals in an active segment, by the energy magnitude of each sensor composition multi-C vector, as bioelectricity eigenwert.Each sensor probe (electromyographic signal sensor, impedance signal sensor) in biopotential sensor 1 detects dissimilar sensor signal, sensor probes as different in total N kind obtains N kind sensor signal, thus the combination of different bioelectricity energy is produced at biopotential sensor, the N dimensional vector that the energy magnitude of each sensor is formed, as bioelectrical signals eigenwert, judge that thumb contacts with middle finger or separates.
For acceleration signal, by the acceleration signal in an active segment, calculate the change of the acceleration on three-dimensional respectively, as acceleration signature value.Because thumb is when touching different fingers and different finger-joint, different space motion paths will be formed; The acceleration-deceleration process that thumb is different in three axial directions simultaneously, also can form different acceleration in three axial directions.And the length of the duration of contact identifying thumb and middle finger is responsible for by acceleration transducer.Because sensor can identify thumb velocity variations in three axial directions for this reason, and for short contact, its velocity variations is faster, and the velocity variations of long contact is slower.So, by the acceleration change of three axis, as the eigenwert of classification, can judge that thumb and middle finger are long contacts or short contact.
(5) Fusion Features unit 105, is undertaken merging and analyzing by the eigenwert of described bioelectrical signals and acceleration signal, identify contact action between finger and duration of contact length, be converted into the basic composition unit of corresponding Morse code.
In this unit, the eigenwert of the two class signals obtained in feature extraction unit 104 is merged, judge it is contact or separate between finger according to bioelectricity eigenwert; Judge the length of finger Contact time according to acceleration signature value, what acceleration change was fast is short contact, and what acceleration change was slow is long contact, and short contact, long contact and finger unclamp the code in short-term of corresponding Morse code respectively, long time-code and space.
(6) symbol maps unit 108, according to the corresponding relation of Morse code character and Morse code basic composition unit, obtains gesture motion and does corresponding character.Then, by wireless transmission method, the character after identifying is sent to external unit, as the character after identification sent to peripheral hardware by the mode such as bluetooth, WIFI.
By two sensors at wrist and thumb place in the program, to finger between contact action and duration of contact length identify, thus achieve the input of Morse code, solve Morse code input equipment in prior art and need special device, carry inconvenient word, except being provided with simply wearable sensor at wrist and thumb in the solution of the present invention, carrying of this equipment is made to become abnormal convenient, without the need to other labyrinth, easy to carry and use, and input Morse code by the mode of finger contact, make input mode simpler answer and easy to operate and grasp, the input of Morse code can be used whenever and wherever possible widely.
embodiment 2:
There is provided the input method of the Morse code character input system based on finger motion pattern in a kind of above-described embodiment in the present embodiment, process flow diagram as shown in Figure 3, comprises the steps:
S1, the bio-electrical information when biopotential sensor 1 being arranged on wrist gathers user's finger motion, the acceleration transducer 2 be arranged on thumb gathers movement locus and the acceleration of thumb.
Biopotential sensor 1 is placed on the muscle group surface of user's wrist, comprise multiple sensor probe, as the impedance detection probe etc. of electromyographic signal detection probe, skin surface, for obtaining the bio-electrical information of user's finger motion, as obtained the impedance change signal etc. of electromyographic signal below skin and skin surface.Acceleration transducer 2 is placed on user's thumb surface, obtains the movement locus of user's thumb on three dimensions, and thumb is at X, Y, Z tri-straight-line displacement acceleration axially and and angular displacement acceleration.
Wherein, the biopotential sensor of wrist is responsible for the posture identifying hand, and namely thumb contacts with middle finger or separates.Because these two kinds of gestures, the form of hand is different, so the tensity of involved muscle group is different, thus the combination of different bioelectricity energy is produced at biopotential sensor, the N dimensional vector that the energy magnitude of each sensor is formed, as the eigenwert of classification, judge that thumb contacts with middle finger or separates.Be positioned at the acceleration transducer of thumb position, acceleration transducer 2 is placed on user's thumb surface, obtain the movement locus of user's thumb on three dimensions, and thumb X, Y, Z tri-straight-line displacement acceleration axially and and angular displacement acceleration, the length of the duration of contact identifying thumb and middle finger (also can be other three fingers except thumb) is responsible for by acceleration transducer.Because sensor can identify thumb velocity variations in three axial directions for this reason, and for short contact, its velocity variations is faster, and the velocity variations of long contact is slower.So, by the acceleration change of three axis, as the eigenwert of classification, can judge that thumb and middle finger are long contacts or short contact.。
S2, noise reduction filtering process is carried out to described bioelectrical signals and acceleration signal, and carry out analog to digital conversion.
First, respectively bandpass filtering is carried out to bioelectrical signals, low-pass filtering is carried out to acceleration signal, to eliminate the impact of the factors such as neighbourhood noise.Then with the sampling rate of 500-1000Hz, analog/digital conversion is carried out to two kinds of signals, obtain discrete digital signal samples point.
S3, by process after bioelectrical signals and acceleration signal carry out dividing processing, obtain gesture active segment.
By bioelectrical signals, first obtained the baseline of signal by average filter, then use original signal to deduct baseline, be eliminated the stationary signal of low frequency wonder; Then, service time, window obtained the window self-energy of bioelectrical signals; Afterwards, this energy magnitude is normalized, absolute figure by signal becomes the relative value between 0-1, reset a threshold value between 0-1, if signal energy has several sampled points continuously higher than this threshold value, then thinking that signal is the starting point of gesture active segment herein, is after this that active segment is inner, until several sampled points lower than threshold value, then think that signal is the terminal of gesture active segment herein continuously.
For acceleration signal, also use same based on the detection starting point of threshold value and the logic of terminal.Then, the starting point obtain two kinds of modes and terminal compare and on average, thus complete the segmentation of user's gesture active segment, obtain the signal data in gesture active segment.
In this step, when obtaining gesture active segment, service time, window obtained window self-energy, then by arranging threshold value to obtain the new section satisfied condition, which is simple, but can filter out noise etc., the signal segment and other signal segments with gesture motion are effectively distinguished, for subsequent treatment obtains rational data.
S4, carry out feature extraction for the bioelectrical signals in gesture active segment and acceleration signal respectively, so that identify the action of finger subsequently through Fusion Features.
For bioelectrical signals, when feature extraction, by the bioelectrical signals in an active segment, by the energy magnitude of each sensor composition multi-C vector, as bioelectricity eigenwert.Each sensor probe (electromyographic signal sensor, impedance signal sensor) in biopotential sensor 1 detects dissimilar sensor signal, sensor probes as different in total N kind obtains N kind sensor signal, thus the combination of different bioelectricity energy is produced at biopotential sensor, the N dimensional vector that the energy magnitude of each sensor is formed, as bioelectrical signals eigenwert, judge that thumb contacts with middle finger or separates.
For acceleration signal, by the acceleration signal in an active segment, calculate the change of the acceleration on three-dimensional respectively, as acceleration signature value.Because thumb is when touching different fingers and different finger-joint, different space motion paths will be formed; The acceleration-deceleration process that thumb is different in three axial directions simultaneously, also can form different acceleration in three axial directions.And the length of the duration of contact identifying thumb and middle finger is responsible for by acceleration transducer.Because sensor can identify thumb velocity variations in three axial directions for this reason, and for short contact, its velocity variations is faster, and the velocity variations of long contact is slower.So, by the acceleration change of three axis, as the eigenwert of classification, can judge that thumb and middle finger are long contacts or short contact.。
S5, the eigenwert of described bioelectrical signals and acceleration signal is carried out merging and analyzing, identify contact action between finger and duration of contact length, be converted into the basic composition unit of corresponding Morse code.Judge it is contact or separate between finger according to bioelectricity eigenwert; Judge the length of finger Contact time according to acceleration signature value, what acceleration change was fast is short contact, and what acceleration change was slow is long contact, and short contact, long contact and finger unclamp the code in short-term of corresponding Morse code respectively, long time-code and space.
When thumb and other finger contact, the muscle of wrist near palm of the hand side can shrink, and drives two fingers close, causes the signal amplitude of the biopotential electrode of corresponding site to raise; When thumb and other finger separate, the muscle of wrist near the back of the hand side can shrink, and drives two fingers separately, causes the signal amplitude of the biopotential electrode of corresponding site to raise.Can judge that finger is contact or separates thus.When finger contact, the length of wrist contraction of muscle duration and the speed of acceleration change, can judge long contact or short contact.
S6, corresponding relation according to Morse code character and Morse code basic composition unit, obtain gesture motion and do corresponding character, as shown in Figure 4, gives the corresponding relation of letter and numeral and Morse code.Then, by wireless transmission method, the character after identifying is sent to external unit, as the character after identification sent to peripheral hardware by the mode such as bluetooth, WIFI.
Due to each character of Morse code, all be made up of some length (" clatter ") short (" ticking ") signals, then, in this invention, only need identification 3 kinds of gestures, that is: the long contact of thumb middle finger (" clatter "), thumb middle finger short contact (" ticking "), thumb middle finger unclamp in (space).Two kinds of sensors in the present invention, perform different functions respectively.Wherein, biopotential sensor is responsible for the posture identifying hand, and namely thumb contacts with middle finger or separates.Because these two kinds of gestures, the form of hand is different, so the tensity of involved muscle group is different, thus the combination of different bioelectricity energy is produced at biopotential sensor, the N dimensional vector that the energy magnitude of each sensor is formed, as the eigenwert of classification, judge that thumb contacts with middle finger or separates.And acceleration transducer, be responsible for the length of the duration of contact identifying thumb and middle finger.Because sensor can identify thumb velocity variations in three axial directions for this reason, and for short contact, its velocity variations is faster, and the velocity variations of long contact is slower.So, by the acceleration change of three axis, as the eigenwert of classification, can judge that thumb and middle finger are long contacts or short contact.The eigenwert of the acceleration transducer on the bioelectricity eigenwert of wrist and thumb is merged, just can judge three kinds of gestures, and by these three kinds of gestures, be mapped to the basic composition unit of Morse code, that is: the long contact of thumb middle finger (" clatter "), thumb middle finger short contact (" ticking "), thumb middle finger unclamp in (space).Alternately repeat this three kinds of gestures, then can form different Morse code characters.Finally, by this character by Bluetooth wireless communication module, be sent on external unit, complete input function.
Obviously, above-described embodiment is only for clearly example being described, and the restriction not to embodiment.For those of ordinary skill in the field, can also make other changes in different forms on the basis of the above description.Here exhaustive without the need to also giving all embodiments.And thus the apparent change of extending out or variation be still among the protection domain of the invention.
Claims (10)
1., based on a Morse code character input system for finger motion pattern, it is characterized in that, comprising:
Signal gathering unit, comprise the biopotential sensor being arranged on wrist and the acceleration transducer be arranged on finger, described biopotential sensor gathers bio-electrical information during user's gesture motion, and described acceleration transducer gathers movement locus and the acceleration of thumb;
Signal Pretreatment unit, the bioelectrical signals obtain described signal gathering unit and acceleration signal carry out noise reduction filtering process, and carry out analog to digital conversion;
Signal segmentation unit, carries out dividing processing by the bioelectrical signals after Signal Pretreatment cell processing and acceleration signal, obtains gesture active segment;
Feature extraction unit, carries out feature extraction for the bioelectrical signals in gesture active segment and acceleration signal respectively;
Fusion Features analytic unit, is undertaken merging and analyzing by the eigenwert of described bioelectrical signals and acceleration signal, identify thumb and other contact actions pointed and duration of contact length, be converted into the basic composition unit of corresponding Morse code;
Symbol maps unit, according to the corresponding relation of Morse code character and Morse code basic composition unit, obtains gesture motion and does corresponding character.
2. the Morse code character input system based on finger motion pattern according to claim 1 and 2, is characterized in that, also comprise transmission unit, by wireless transmission method, the character after identifying is sent to external unit.
3. the Morse code character input system based on finger motion pattern according to claim 1 and 2, it is characterized in that, described biopotential sensor is placed on the muscle group surface of user's wrist, obtain the bioelectrical signals of user gesture motion, described bioelectrical signals comprises the impedance transformation signal of electromyographic signal under skin and skin surface.
4., based on a Morse code characters input method for finger motion pattern, it is characterized in that, comprise the steps:
Be arranged on bio-electrical information during biopotential sensor collection user's gesture motion of wrist, the acceleration transducer be arranged on thumb gathers movement locus and the acceleration of thumb;
Noise reduction filtering process is carried out to described bioelectrical signals and acceleration signal, and carries out analog to digital conversion;
Bioelectrical signals after process and acceleration signal are carried out dividing processing, obtains gesture active segment; Feature extraction is carried out respectively for the bioelectrical signals in gesture active segment and acceleration signal;
The eigenwert of described bioelectrical signals and acceleration signal is carried out merging and analyzing, identify contact action between finger and duration of contact length, be converted into the basic composition unit of corresponding Morse code;
Symbol maps unit, according to the corresponding relation of Morse code character and Morse code basic composition unit, obtains gesture motion and does corresponding character.
5. input method according to claim 4, is characterized in that, the process of described " bioelectrical signals after process and acceleration signal are carried out dividing processing, obtains gesture active segment ", comprising:
By bioelectrical signals, first obtained the baseline of signal by average filter, then use original signal to deduct baseline, be eliminated the stationary signal of low frequency wonder; Then, service time, window obtained the window self-energy of bioelectrical signals; Afterwards this energy magnitude is normalized, absolute figure by signal becomes the relative value between 0-1, a threshold value is set in 0-1, if signal energy has several sampled points continuously higher than this threshold value, then think that signal is the starting point of gesture active segment herein, after this be that active segment is inner, until several sampled points lower than threshold value, then think that signal is the terminal of gesture active segment herein continuously.
6. the input method according to claim 4 or 5, is characterized in that, the process of described " carrying out feature extraction for the bioelectrical signals in gesture active segment and acceleration signal respectively ", comprising:
By the bioelectrical signals in an active segment, by the energy magnitude of each sensor composition multi-C vector, as bioelectricity eigenwert;
By the acceleration signal in an active segment, calculate the change of the acceleration on three-dimensional respectively, as acceleration signature value.
7. according to the arbitrary described input method of claim 4-6, it is characterized in that, described " eigenwert of described bioelectrical signals and acceleration signal is carried out merging and analyzing; identify contact action between finger and duration of contact length; be converted into the basic composition unit of corresponding Morse code " process, comprising:
Judge it is contact or separate between finger according to bioelectricity eigenwert;
The length of finger Contact time is judged according to acceleration signature value.
8. according to the arbitrary described input method of claim 4-6, it is characterized in that, judge that according to acceleration signature value the process of the length pointing the Contact time comprises: what acceleration change was fast is short contact, slow the contacting for length of acceleration change.
9., according to the arbitrary described input method of claim 4-6, it is characterized in that, described in when being converted into the basic composition unit of corresponding Morse code, comprise
Short contact, long contact and finger unclamp the code in short-term of corresponding Morse code respectively, long time-code and space.
10., according to the arbitrary described input method of claim 4-6, it is characterized in that, also comprise the process by wireless transmission method, the character after identification being sent to external unit.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410289073.7A CN105204611B (en) | 2014-06-24 | 2014-06-24 | A kind of Morse code character input system and method based on finger motion pattern |
| US14/748,557 US9619024B2 (en) | 2014-06-24 | 2015-06-24 | Virtual input device and virtual input method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410289073.7A CN105204611B (en) | 2014-06-24 | 2014-06-24 | A kind of Morse code character input system and method based on finger motion pattern |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105204611A true CN105204611A (en) | 2015-12-30 |
| CN105204611B CN105204611B (en) | 2018-01-09 |
Family
ID=54952349
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410289073.7A Active CN105204611B (en) | 2014-06-24 | 2014-06-24 | A kind of Morse code character input system and method based on finger motion pattern |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105204611B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107632699A (en) * | 2017-08-01 | 2018-01-26 | 东南大学 | Natural human-machine interaction system based on the fusion of more perception datas |
| CN109656358A (en) * | 2018-11-23 | 2019-04-19 | 南京麦丝特精密仪器有限公司 | A kind of multidimensional sign Language Recognition Method |
| CN110585684A (en) * | 2019-07-05 | 2019-12-20 | 华东理工大学 | VR interaction method and system based on motion recognition |
| CN113407041A (en) * | 2018-08-01 | 2021-09-17 | 神讯电脑(昆山)有限公司 | Moss code input method and portable electronic device thereof |
| CN117421540A (en) * | 2023-12-19 | 2024-01-19 | 北京中电慧声科技有限公司 | Automatic Morse code identification method and device |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6380923B1 (en) * | 1993-08-31 | 2002-04-30 | Nippon Telegraph And Telephone Corporation | Full-time wearable information managing device and method for the same |
| CN101639730A (en) * | 2008-07-31 | 2010-02-03 | 鸿富锦精密工业（深圳）有限公司 | Information input method and handheld device using same |
| CN103760979A (en) * | 2014-01-20 | 2014-04-30 | 联想(北京)有限公司 | Information processing method and electronic equipment |
| WO2014081927A1 (en) * | 2012-11-24 | 2014-05-30 | Keller Eric Jeffrey | Computing interface system |
-
2014
- 2014-06-24 CN CN201410289073.7A patent/CN105204611B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6380923B1 (en) * | 1993-08-31 | 2002-04-30 | Nippon Telegraph And Telephone Corporation | Full-time wearable information managing device and method for the same |
| CN101639730A (en) * | 2008-07-31 | 2010-02-03 | 鸿富锦精密工业（深圳）有限公司 | Information input method and handheld device using same |
| WO2014081927A1 (en) * | 2012-11-24 | 2014-05-30 | Keller Eric Jeffrey | Computing interface system |
| CN103760979A (en) * | 2014-01-20 | 2014-04-30 | 联想(北京)有限公司 | Information processing method and electronic equipment |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107632699A (en) * | 2017-08-01 | 2018-01-26 | 东南大学 | Natural human-machine interaction system based on the fusion of more perception datas |
| CN107632699B (en) * | 2017-08-01 | 2019-10-11 | 东南大学 | Human-computer natural interaction system based on multi-sensory data fusion |
| CN113407041A (en) * | 2018-08-01 | 2021-09-17 | 神讯电脑(昆山)有限公司 | Moss code input method and portable electronic device thereof |
| CN113407041B (en) * | 2018-08-01 | 2023-09-05 | 神讯电脑(昆山)有限公司 | Moss code input method and portable electronic device thereof |
| CN109656358A (en) * | 2018-11-23 | 2019-04-19 | 南京麦丝特精密仪器有限公司 | A kind of multidimensional sign Language Recognition Method |
| CN110585684A (en) * | 2019-07-05 | 2019-12-20 | 华东理工大学 | VR interaction method and system based on motion recognition |
| CN117421540A (en) * | 2023-12-19 | 2024-01-19 | 北京中电慧声科技有限公司 | Automatic Morse code identification method and device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105204611B (en) | 2018-01-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105446461A (en) | Gesture recognition method, palm virtual keyboard using same, and input method | |
| CN103246349B (en) | Input method of wearable finer-sensing wireless communication, and input device using input method | |
| CN105204611A (en) | Finger movement mode based Morse code character inputting system and method | |
| CN105573536B (en) | Processing method, the device and system of touch-control interaction | |
| US9619024B2 (en) | Virtual input device and virtual input method | |
| CN106527716A (en) | Wearable equipment based on electromyographic signals and interactive method between wearable equipment and terminal | |
| CN105190578A (en) | Methods and devices that combine muscle activity sensor signals and inertial sensor signals for gesture-based control | |
| CN105204640A (en) | Wearable mouse | |
| JP2010204724A (en) | Input system and electric equipment | |
| CN103126657A (en) | Device and method for monitoring finger and wrist fatigue and working time | |
| Li et al. | Hand gesture recognition and real-time game control based on a wearable band with 6-axis sensors | |
| CN106200988A (en) | A kind of wearable hand language recognition device and sign language interpretation method | |
| CN101118471B (en) | System and method for implementing function by user-defined hand-written symbol | |
| CN107622235B (en) | Fingerprint unlocking method and related product | |
| CN105118356B (en) | A kind of gesture language-voice conversion method and device | |
| CN103971108A (en) | Wireless communication-based human body posture recognition method and device | |
| CN103176651B (en) | A kind of hand-written information Quick Acquisition method | |
| Manware et al. | Smart gloves as a communication tool for the speech impaired and hearing impaired | |
| Quiapo et al. | Development of a sign language translator using simplified tilt, flex and contact sensor modules | |
| CN109478115A (en) | Noise sensing circuit and touch device | |
| CN101907924B (en) | Impulse-based information input device | |
| Mankar et al. | Review on hand gesture based mobile control application | |
| CN105988695B (en) | Intelligent device and operation response method thereof | |
| KR101640898B1 (en) | Character recognizable wearable touch system and method thereof through emg signals | |
| CN203217496U (en) | Input device for wearable finger-sensing wireless communication |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |