CN103093234A - Identity recognition method based on ground reactive force during walking - Google Patents
Identity recognition method based on ground reactive force during walking Download PDFInfo
- Publication number
- CN103093234A CN103093234A CN2012105596657A CN201210559665A CN103093234A CN 103093234 A CN103093234 A CN 103093234A CN 2012105596657 A CN2012105596657 A CN 2012105596657A CN 201210559665 A CN201210559665 A CN 201210559665A CN 103093234 A CN103093234 A CN 103093234A
- Authority
- CN
- China
- Prior art keywords
- gait
- force
- wavelet
- ground reaction
- reaction force
- 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
Images
Landscapes
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
Abstract
The invention discloses an identity recognition method based on ground reactive force during walking. The identity recognition method is characterized in that a gait channel composed of a three-dimensional force detecting platform hidden on the ground or under a floor is used for collecting gait information during walking, the gait information is subjected to data preprocessing, a wavelet packet decomposition coefficient, wavelet packet energy, a wavelet packet decomposition mean value and wavelet packet decomposition variance are extracted through a wavelet packet decomposition algorithm to represent gait features, a feature selecting algorithm and a support vector machine classifier are used in a training process to establish a trained gait feature template and a trained classifying template, the trained gait feature template is used in a recognition process for feature mapping of the gait features of an extracted to-be-recognized object, the trained classifying template is used for template matching and recognition of the gait feathers, and recognition results are output. According to the identity recognition method, the gait information collecting process and the gait information recognition process have invisibility, perceiving and human right issue are avoided, and identity recognition and safety protective requirements in safety sensitive places in the high-technology era can be well met.
Description
Technical field
The invention belongs to mode identification technology, the personal identification method of ground reaction force during particularly based on walking.
Background technology
The world today faces the safe difficult problems such as the severe attack of terrorism and finance theft, and identification or authentication become more and more important, especially on the airport, the security sensitive such as customs, museum, party scene, bank and national treasury place.Biometrics identification technology is to utilize the intrinsic physiological characteristic of human body and behavioural characteristic to realize individual identity identification or authentication, because selected biological characteristic is that human body is intrinsic, can carry, do not need memory, have uniqueness and be difficult to the characteristics such as replicability, more be applicable to identification or authentication under new demand for security.At present, be used for the existing kind more than ten of biological characteristic of identification, wherein, physiological characteristic mainly contains fingerprint, people's face, iris, palmmprint, palm shape, people's ear, retina, DNA and smell etc., and behavioural characteristic mainly contains signature, voice, keystroke and gait etc.The existing comparatively ripe product of the recognition technologies such as fingerprint, palmmprint, palm shape, iris, people's face, DNA, voice and signature, they have all played positive effect in application separately.But regrettably, the biometrics identification technologies such as fingerprint, people's face, iris, palmmprint, palm shape, people's ear, retina, DNA, smell, signature, voice and keystroke all need the checked object close fit just can complete information acquisition when practical application, do not possess disguise, easily cause discovering and then taking the counterreconnaissance means of offender.Meanwhile, physiological characteristic used is all static nature, is easy to forge; This three behaviors feature of keystroke, voice and signature is also easily imitated; Except senior recognition of face, above-mentioned biometrics identification technology is all contact collection and closely identification, easily causes human rights dispute and diplomatic disturbance, not too is applicable to the anti-terrorism safety check demand of entry and exit port.Although DNA authentication technique reliability is high, what sampling neither hidden, and its analytic process is longer, requires high-end instrument and equipment, only be confined at present that the laboratory is used and the law evaluation in.That is to say, be identity recognizing technology closely based on the biometrics identification technology of these physiological characteristics and this three behaviors feature of keystroke, voice and signature, can not adapt to well the identification requirement under the current safety situation.Therefore, need to develop and use remote identity recognizing technology.
At present, the biological characteristic that can be used for remote identification has people's face and gait.What still adopt due to remote recognition of face is the human face image information that video camera or camera obtain, its identification and closely recognition of face blocked with background influence by illumination, people or thing larger, and a great problem especially of the face tracking under crowded crowd or complex environment.The posture that Gait Recognition is walked by the people and/or the footprint that stays are identified and are authenticated identity, people can have such cognitive experience: we can be in the crowd by observing his/her far familiar people that just recognizes that walks, perhaps do not hear that walking sound just can recognize familiar people in the situation that see its people, this cognitive experience is also by the experimental verification of psychophysics family expenses, they point out: even in very poor environment of observation, the mankind are the capable identity of distinguishing walker according to gait also.At computer vision field, Gait Recognition has more advantage with respect to recognition of face, and it utilizes people's walking behavioural characteristic, easily unlike people's face just can get by under false pretences with means such as face mould or cosmetics; Also can play a role in some special event, such as in the events such as bank raid and museum's historical relic theft, the suspect puts on face shield or cap escapes from the tracking of video camera with regard to hiding the face picture, but it is visible that gait remains, or the discriminating that can carry out gait judges whether it is real criminal, and precondition is that the suspect does not deliberately change its gait.In fact, people can on the vision pattern, others' gait be imitated the spitting image of, produce gait image information about the same, thereby cause the gait recognition method on computer vision correctly to distinguish.Therefore, need research and development based on the personal identification method of other gait information.
Chinese patent ZL01144157.7 discloses a kind of method of identifying personnel by the gait image sequence of human locomotion, the method information acquisition has the advantages such as untouchable and disguised, taked that space profiles is cut apart, the methods such as principal component analysis (PCA) and personalized corporal characteristic auxiliary examination improve discrimination and reduce calculation cost, but this method based on gait image is owing to using common camera to take gait image, can't ideally solve complex background, block the problem with noise, also there is no to consider deliberately to imitate the problem of gait.The disclosed a kind of multiaxis power plane matrix that is consisted of by four six axle power platforms of Chinese patent ZL200410014352.9, and the method for dynamic gait information when obtaining human body walking by this plane matrix, this plane matrix can be used for test and the analysis of balanced ability of human body, the coordination ability and nervous function, but does not relate to the method for Method of Gait Feature Extraction and Gait Recognition.disclosed a kind of " the gait detecting system of US Patent No. 2002/0107649A1, pick-up unit, equipment and gait detection method " (Gait detection system, gaitdetection apparatus, device, and gait detection method), sound oscillation signal when detecting walking also is used as personal identification system, this system requirements is placed microphone (acoustic-electric converter) on human body, microphone obtains the electric signal subset by the sound oscillation energy that gathers in the walking movement process, detect walking period with this electric signal subset as an index that goes on foot, the signal subset of the sound oscillation when collection expression pin lands, system extracts the specific gait feature of human body based on this signal subset, and use it for the individual and identify.This personal identification system has two deficiencies: the one, and the sound oscillation when landing is the placement position on human body and changing along with microphone; The 2nd, due to the major effect that is subjected to microphone ambient noise and electrical noise, can not determine exactly the sound oscillation electric signal when the expression pin lands, also just be difficult to extract exactly the feature of gait waveform.patented claim international publication number WO/2004/040501 disclosed " gait waveform feature extracting method and personal identification system " (Gait waveform feature extracting method and individual identification system) with the waveform peak amplitude in allocated frequency band as the reference index, and determine the gait waveform in a step in the electric signal that detects and extract method and the personal identification system of gait waveform feature from the electric field displacement detector, can not be subjected to a left side, electric charge between right leg is interfered and the installation position impact of sensor on health, but still be not suitable in the environment that has complicated electromagnetic interference (EMI).And, in the technical scheme that US Patent No. 2002/0107649A1 and international publication number WO/2004/040501 open, its gait information remains and relies on the sensor collection that is placed on health, remain the contact type measurement mode, the measured can perceive and obtains, thereby may cause the offender to take the counterreconnaissance means to get by under false pretences because perceiveing, even may be stung to fury so that the event that wakes a snake in the crowd.
Summary of the invention
The objective of the invention is to propose a kind of personal identification method of ground reaction force during based on walking, make identifying be difficult for being realized with hidden and non-contact measurement method, avoid being identified object and get by under false pretences, the identification in the security sensitive such as customs, airport, museum, party are on-the-spot to be applicable to, bank and national treasury place.
The personal identification method of ground reaction force when the present invention is based on walking is characterized in that comprising training process and identifying;
Described training process comprises: utilize the gait information when being hidden in the gait passage collection walking that ground or underfloor three-dimensional force force plate/platform form and set up the gait data storehouse, carry out data pre-service, Method of Gait Feature Extraction, gait feature selection and sorter training, finally obtain gait feature template and the classification model of having trained;
Described identifying comprises: obtain the real-time gait information of object to be identified to obtain test sample book, test sample book is carried out data pre-service, Method of Gait Feature Extraction, utilize the gait feature template of having trained to carry out Feature Mapping to the gait feature that object to be identified extracts, the classification model that recycling has been trained carries out template matches and identification according to nearest neighbouring rule to test sample book, the output recognition result;
Described gait information is the ground reaction force by the single step of three-dimensional force force plate/platform collection, or the ground reaction force of continuous a plurality of single steps cumulative ground reaction force that obtains synthesizing on sequential;
The ground reaction force of described single step or synthetic ground reaction force all are comprised of these three components of left and right directions shearing force, fore-and-aft direction shearing force and vertical direction anchorage force;
Described data pre-service comprises that denoising and effective sample select; Described denoising refers to adopt the Wavelet Transform Threshold method that the ground reaction force data are decomposed and reconstruction processing; Whether described effective sample is selected the data dimension and the peak point numerical value that refer to according to vertical direction anchorage force in the ground reaction force after denoising and is judged in effective range, selects ground reaction force data in effective range as effective sample; Described effective range refers to the numerical range relevant with sample frequency Fp to the number Np of the three-dimensional force force plate/platform that adopts, ground reaction force for single step, if wherein the data dimension of vertical direction anchorage force is lower than 0.4Fp or higher than 0.8Fp, or the ordinate value of the crest of vertical direction anchorage force and trough differs over 300, namely be judged to be invalid sample, otherwise be judged to be effective sample; For the synthetic ground reaction force of the continuous a plurality of single steps that gathered by Np three-dimensional force force plate/platform, if wherein the data dimension of vertical direction anchorage force is lower than 0.4Fp * Np or higher than 0.8Fp * Np, or the number of the trough of vertical direction anchorage force is less than Np, or the maximin of the ordinate value of all troughs of vertical direction anchorage force differs over 300, namely be judged to be invalid sample, otherwise be judged to be effective sample;
Described Method of Gait Feature Extraction refers to that the average and the variance that adopt Wavelet Packet Transform Method to extract WAVELET PACKET DECOMPOSITION coefficient, wavelet-packet energy, WAVELET PACKET DECOMPOSITION coefficient from ground reaction force characterize gait feature, and step is:
At first adopt the piecewise linear interpolation algorithm to carrying out dimension normalization through data pretreated ground reaction force data, the data dimension of effective sample is normalized to same value;
Effective sample after adopting again L layer wavelet packet decomposition algorithm to denoising and dimension normalization carries out WAVELET PACKET DECOMPOSITION, decomposes to obtain 2
L+1-2 sub-wavelet packets characterize gait feature with average and the variance of the WAVELET PACKET DECOMPOSITION coefficient of every sub-wavelet packet, wavelet-packet energy, WAVELET PACKET DECOMPOSITION coefficient;
Then adopt feature selecting algorithm that the WAVELET PACKET DECOMPOSITION coefficient is selected, pick out optimal wavelet bag coefficient of dissociation, with average and the variance combination of optimal wavelet bag coefficient of dissociation and wavelet-packet energy, WAVELET PACKET DECOMPOSITION coefficient, the gait feature template that obtains the gait feature collection and trained;
Described sorter training refers to adopt support vector machine classifier that training sample is trained, the classification model that obtains having trained; This sorter training also can first adopt maximum-minimum requirement method to carry out normalization to the gait feature of inputting support vector machine on amplitude before described employing support vector machine classifier is trained training sample.
When compared with prior art, the present invention is based on walking there be the remarkable advantage of the personal identification method of ground reaction force:
one, the ground reaction force that when the gait information that is used for identification in the present invention is human body walking, foot and ground interaction produce, can be convenient by being hidden in ground or underfloor force plate/platform, get exactly, with respect to sensor being placed in measured's method of obtaining walking sound oscillation signal and walking electric field signal with it, because the gait information collecting device in the present invention can be hidden under ground or floor, be not placed in it the measured, gait information acquisition process is hidden to the measured fully, on naturally walking without any impact of measured.That is to say, gait information acquisition of the present invention is noncontact and hidden measuring method, can obtain the true gait information when being identified the walking of object nature, be identified object and can not realize identification apparatus is arranged and take measures to get by under false pretences, can also can monitor round the clock identifying in far place at the scene, can not cause human rights dispute yet, can satisfy better that cyberage airport, customs, museum, party are on-the-spot, the identity in the security sensitive such as bank and national treasury place differentiates and the safety precaution demand.
They are two years old, due to gait information used in the present invention ground reaction force that to be the human body walking time foot that obtains of force plate/platform produce with ground interaction, with respect to obtaining the method for gait image in computer vision field by video camera, recognition result is not subjected to that complex background, clothes and health block etc. to be affected; What more have superiority is a bit, situation when the present invention can tackle mimicker's gait deliberately, and computer vision field can't better be tackled based on the personal identification method of gait, a because people mimicker's the gait posture that often others walks from vision pattern learning, can imitate the spitting image of, but because the gait mechanical information of foot and ground interaction can't observe and record others and walk the time, also just can't learn, imitate mechanical information, produce surely similar gait mechanical information even if similar walking posture also differs.
Its three, adopt the inventive method, both can use single force plate/platform to obtain the ground reaction force of single step, the ground reaction force of single step is used for identification; Also can use a plurality of force plate/platforms to obtain the ground reaction force of continuous a plurality of single steps, the ground reaction force that then adds up and obtain synthesizing on sequential, more synthetic ground reaction force is used for identification; The feature extraction of adopting and sorting algorithm can be distinguished and process well both of these case, facilitate the application of different places.
Its four, the gait mechanical information acquisition method, Method of Gait Feature Extraction and the identification algorithm that adopt in the present invention can promote the use of a lot of fields.For example, in the medical rehabilitation field, by with the contrast of normal person's gait feature, can the evaluating patient skeletal muscle etc. the degree that suffers damage of tissue and the ability of controlling balance, the patient is classified and classification, determine the therapy rehabilitation scheme, and assess the rehabilitation degree can verify curative effect, treatment in therapeutic process after; In addition, this gait mechanical information can also be applied to the fields such as the design of insurance assessment, Claims Resolution (insurance company is used for disability assessment, therapy rehabilitation assessment) and human lives's means of production (furniture, shoes, artificial limb, office accommodations etc.) and body-building, rehabilitation material and measures of effectiveness.
Description of drawings
The algorithm flow chart of the identification of ground reaction force when Fig. 1 is based on walking;
Fig. 2 is the uniformly-spaced gait access diagram of staggered assembly unit of ground reaction force when being hidden in subsurface collection walking;
Fig. 3 is the gait access diagram of the tight concordant assembly unit of ground reaction force when being hidden in subsurface collection walking;
Left and right directions shearing force Fx, fore-and-aft direction shearing force Fy when Fig. 4 is walking in the ground reaction force of single step and the force-time curve of vertical direction anchorage force Fz;
Left and right directions shearing force Fhx, fore-and-aft direction shearing force Fhy in the ground reaction force that when Fig. 5 is walking, continuous a plurality of single steps are synthesized and the force-time curve of vertical direction anchorage force Fhz;
Fig. 6, Fig. 7 and Fig. 8 are respectively left and right directions shearing force curve, fore-and-aft direction shearing force curve and the vertical direction anchorage force curves in the ground reaction force of the single step before denoising;
Fig. 9, Figure 10 and Figure 11 are respectively left and right directions shearing force curve, fore-and-aft direction shearing force curve and the vertical direction anchorage force curves that adopts in the ground reaction force of the single step after the denoising of Wavelet Transform Threshold method;
Figure 12 is left and right directions shearing force Fhxd, fore-and-aft direction shearing force Fhyd in synthetic ground reaction force after denoising and dimension normalization and the force-time curve of vertical direction anchorage force Fhzd;
Figure 13 is through the vertical direction anchorage force Fwz curve after denoising, weight standard and dimension normalized;
Figure 14 carries out 4 layers of second layer metric space after WAVELET PACKET DECOMPOSITION to vertical direction anchorage force Fwz;
Figure 15 carries out 4 layers of second layer wavelet space after WAVELET PACKET DECOMPOSITION to vertical direction anchorage force Fwz.
Embodiment
Embodiment 1:
Remote biometrics identification technology is high-tech crime epoch protection public safety and personal security's important identity recognizing technology means; can be used at present the biological characteristic of remote identification except people's face, gait is human body behavior biological characteristic with the largest potentiality at last.The present invention is take people's gait as starting point, the ground reaction force that when employing is hidden in ground or underfloor force plate/platform collection people walking, foot and ground interaction produce, replace the gait image sequence of easily imitating with the gait mechanical information that can not imitate, and proposed the obtaining of gait mechanical information, disposal route and personal identification method.The present invention is further illustrated below in conjunction with drawings and Examples, it is pointed out that described embodiment is for the ease of understanding the present invention, the present invention not being played any restriction effect.
The present invention propose based on walking the time ground counteractive personal identification method comprise training process and identifying, algorithm flow is as shown in Figure 1.In the present embodiment, no matter be the gait information of training process or identifying, be all the ground reaction force that adopts when being hidden in the walking that gait passage that ground or underfloor three-dimensional force force plate/platform form collects.The present invention utilizes the gait passage that is comprised of force plate/platform to gather gait information, set up the gait data storehouse, then the gait information that collects is carried out denoising and effective sample and the data pre-service such as select, the design feature extraction algorithm extracts gait feature again, adopt feature selecting algorithm to pick out optimum gait feature in training process, the gait feature template that obtains having trained, select simultaneously suitable sorter that training sample is trained, the classification model that obtains having trained, the algorithm flow of training process is as shown in the first half in Fig. 1; Adopt in identifying the feature extraction algorithm the same with training process from Real-time Obtaining to the gait information of object to be identified extract gait feature, utilize the gait feature template of having trained that obtains in training process to carry out Feature Mapping to the gait feature of object to be identified, the classification model of having trained that obtains in the recycling training process carries out template matches and identification according to nearest neighbouring rule to test sample book, export at last recognition result, the algorithm flow of identifying is as shown in Fig. 1 the latter half.
The below is described in more detail gait information acquisition of the present invention, Method of Gait Feature Extraction and recognition methods.
1. gait information acquisition
From present technology realizability, the gait information that can be used for Gait Recognition has gait image/video information, gait acoustic information, gait electromagnetic field signal and gait mechanical information, gait image/video information has had a lot of researchs in Gait Recognition is used, but still exists and blocked and the problem such as illumination effect; Gait acoustic information and the collection of gait electromagnetic field signal need to be placed in sensor with it the measured, remain the contact type measurement mode, and the measured can perceive and obtains, and are easy to occur discriminated one's identification and take the means such as imitation to get by under false pretences because perceive.The present invention is directed to Gait Recognition these deficiencies on Information Level, propose the gait mechanical information, i.e. during walking, foot is used for identification with the ground reaction force that ground interaction produces.Gait mechanical information as much as possible when naturally walking in order to collect object, the present invention is arranged on a plurality of three-dimensional force force plate/platforms under ground or floor snugly by certain erection method, form a gait passage, the three-dimensional ground reaction force when gathering human body walking by the gait passage.Adopting China Patent No. in the present invention is that six axle power platforms described in ZL200410014352.9 " a kind of multiaxis power plane matrix and human body walking gait information getting method " gather three-dimensional force, is referred to as in the present invention the three-dimensional force force plate/platform.In specific implementation process, with place size and the application demand of a plurality of three-dimensional force force plate/platforms according to application places, adopt different erection methods to form the gait passage, Fig. 2 is the uniformly-spaced gait access diagram of staggered assembly unit of ground reaction force when being hidden in subsurface and gathering walking, and a plurality of three-dimensional force force plate/platform Plt are assemblied to form a gait passage alternately with one the first from left right side, identical interval; Fig. 3 is the gait access diagram of the tight concordant assembly unit of ground reaction force when being hidden in subsurface collection walking, and a plurality of three-dimensional force force plate/platform Plt closely evenly are assemblied to form a gait passage one by one.the uniformly-spaced gait passage of staggered assembly unit shown in Figure 2 can be distinguished the ground reaction force of left and right pin simply with respect to the gait passage of tight concordant assembly unit shown in Figure 3, but require measured's left foot Lf and right crus of diaphragm Rf correctly to set foot on corresponding force plate/platform by the left and right corresponded manner of figure indicating, be a three-dimensional force force plate/platform Plt below the zone that in Fig. 2, each black box indicates, on direct of travel, require measured's right crus of diaphragm Rf to step in the black box on dotted line right side in the drawings, left foot Lf steps in the black box in the left side of dotted line in the drawings, so just can guarantee that the measured correctly sets foot on corresponding force plate/platform, there is no this requirement on gait passage shown in Figure 3.
In the present invention, the ground reaction force that gathers is three-dimensional ground reaction force, comprise left and right directions shearing force Fx, fore-and-aft direction shearing force Fy and vertical direction anchorage force Fz, left and right directions shearing force Fx, fore-and-aft direction shearing force Fy when Fig. 4 is walking in the ground reaction force of single step and the force-time curve of vertical direction anchorage force Fz, vertical direction anchorage force Fz mainly reflects the load-bearing process of walking, its force-time curve generally is horse-saddle, namely generally has two crests and clips a trough; Shearing force mainly reflects the process of friction catch, driving and balance health.Three-dimensional ground reaction force when therefore, adopting walking will intuitively characterize from multi-angle the pattern feature of walking.
2. gait data pre-service
In order to make more accurate stable of the gait feature that extracts, the gait data that requires to be used for feature extraction has higher quality, and the data that collect often contain noise, and some data even may be because the invalid data of acquisition operations generation lack of standardization.Therefore, before carrying out Method of Gait Feature Extraction, need to carry out pre-service to the gait data that collects, comprise that denoising and effective sample select.
For the noises such as association interference between the power frequency interference that exists in data acquisition circuit, circuit, adopted the hardware filtering method to carry out preliminary denoising on circuit, in order to eliminate various noises as far as possible, adopt again the software filtering method to carry out further denoising on host computer.Be described as the multiresolution analysis characteristic of " school microscop " because wavelet transformation has, make it have band-pass filtering function, Wavelet noise-eliminating method is used more and more wider.Wavelet noise-eliminating method carries out signal wavelet decomposition under different scale, just the mixed signal that comprises the multi-frequency composition can be decomposed different frequency range, then presses frequency band according to the different characteristic of each seed signal on frequency domain and processes.Ground reaction force signal during walking be mainly frequency at 40Hz with interior low frequency signal, and be the 50Hz industrial frequency noise of working power introducing and the electromagnetic interference (EMI) in Acquisition Circuit by the overriding noise in gait passage acquisition system, the electromagnetic interference (EMI) major part is high frequency noise.In addition, the singular point of the ground reaction force that collects is more, a lot of flex points can occur on its force-time curve; And based on wavelet modulus maxima method and Wavelet Transform Threshold method relatively are fit to carry out denoising to containing the more signal of singular point or data, again because the Wavelet Transform Threshold method is applicable to the denoising of most of signals, and computing velocity is fast, is very suitable for the ground reaction force data in the present invention are carried out the requirement that real-time de-noising is processed.Therefore, the present invention selects the Wavelet Transform Threshold method to carry out noise suppression preprocessing to the ground reaction force data.Wavelet Denoising Method comprises wavelet decomposition → to the operation of high frequency wavelet coefficient processing → wavelet reconstruction three basic, for time-domain signal f (t), first with its discrete f that turns to
k, make c
0, k=f
kThe gait data that collects in the present invention can be considered as discretize data, i.e. f
k=f (t), the orthogonal wavelet decomposition formula of signal f (t) is:
The wavelet reconstruction formula is:
In formula, c
j,kBe scale coefficient, d
j,kBe wavelet coefficient, h, g are pair of orthogonal mirror filter group, and j is for decomposing the number of plies, and N is discrete sampling count (data length).
For the Wavelet Transform Threshold method, estimate wavelet coefficient c according to threshold value λ
j,kNormal adopt two kinds of functions: soft-threshold function and hard-threshold function are called wavelet transformation soft-threshold method and wavelet transformation hard threshold method based on the Wavelet Transform Threshold method of these two kinds of functions.Carry out signal after wavelet reconstruction can obtain denoising by the wavelet coefficient of estimating.The soft-threshold function is:
The hard-threshold function is:
In formula, sgn () is sign function, and λ is threshold value.The choosing method of threshold value λ is more, and is larger on the denoising effect impact, should select according to pretreated signal.Unified threshold value by Donoho and Johnstone proposition
More with getting, σ is the standard deviation of noise.
By analyzing and experimental verification, determine to adopt 5 layers of Daubechies wavelet function as the wavelet basis function of Wavelet Transform Threshold method, gait data to be carried out denoising in the present embodiment.Fig. 6, Fig. 7 and Fig. 8 are respectively the examples of left and right directions shearing force curve, fore-and-aft direction shearing force curve and vertical direction anchorage force curve in the ground reaction force of the single step before denoising.Wherein, Fig. 6 is the force-time curve of the left and right directions shearing force before denoising, and Fig. 7 is the force-time curve of the fore-and-aft direction shearing force before denoising, and Fig. 8 is the force-time curve of the vertical direction anchorage force before denoising; Fig. 9, Figure 10 and Figure 11 adopt the Wavelet Transform Threshold method to carry out force-time curve example after denoising to the ground reaction force data of Fig. 6, Fig. 7 and single step shown in Figure 8.Wherein, Fig. 9 is the force-time curve of the left and right directions shearing force after denoising, and Figure 10 is the force-time curve of the fore-and-aft direction shearing force after denoising, and Figure 11 is the force-time curve of the vertical direction anchorage force after denoising.Can find by the contrast to these curves, the Wavelet Transform Threshold method is being fallen the low noise while, the unique point of original signal has all been preserved, and do not changed details and the configuration feature of original signal, proves that its denoising effect is better.
Effective sample in the present invention selects according to the data dimension of the vertical direction anchorage force in the ground reaction force after denoising and peak point numerical value whether judge in effective range whether the gait data that gathers can be used as effective sample, selects gait data in effective range as effective sample according to term of reference.Described effective range refers to the numerical range relevant with sample frequency Fp to the number Np of the three-dimensional force force plate/platform that adopts, ground reaction force for single step, if wherein the data dimension of vertical direction anchorage force is lower than 0.4Fp or higher than 0.8Fp, or the ordinate value of the crest of vertical direction anchorage force and trough differs over 300, namely be judged to be invalid sample, otherwise be judged to be effective sample; For the synthetic ground reaction force of the continuous a plurality of single steps that gathered by Np three-dimensional force force plate/platform, if wherein the data dimension of vertical direction anchorage force is lower than 0.4Fp * Np or higher than 0.8Fp * Np, or the number of the trough of vertical direction anchorage force is less than Np, or the maximin of the ordinate value of all troughs of vertical direction anchorage force differs over 300, namely be judged to be invalid sample, otherwise be judged to be effective sample.In the present embodiment, the sample frequency Fp of three-dimensional force force plate/platform is 800Hz.with a people by 5 three-dimensional force force plate/platforms by uniformly-spaced up the going as example of gait passage of staggered assembly unit shown in Figure 2, when the ground reaction force with single step is used for identification, if the data dimension of the vertical direction anchorage force Fz of ground reacting force is lower than 320 or higher than 640 as shown in Figure 4, or the ordinate value of the crest of vertical direction anchorage force Fz and trough differs over 300, namely be judged to be invalid sample, otherwise be judged to be effective sample, when being used for identification with the synthetic ground reaction force of continuous a plurality of single steps, if in synthetic ground reaction force curve as shown in Figure 5, the data dimension of vertical direction anchorage force Fhz is lower than 1600 or higher than 3200, or the number of the trough of vertical direction anchorage force Fhz (as these points of V that cross curve indicates in Fig. 5) is less than 5, or the maximin of the ordinate value of all troughs of vertical direction anchorage force Fhz (as these points of V that cross curve indicates in Fig. 5) differs over 300, as long as occur a kind of in these three kinds of situations, this gait data is invalid sample, opposite, be effective sample, the gait data of effective sample is set up the gait data storehouse.
3. Method of Gait Feature Extraction and selection
Gait feature is the basis of gait analysis and Gait Recognition, directly affect the quality of classification results based on the quality of the gait feature that is used for classification in the identification of gait, therefore, the good gait feature of feature extraction algorithm Extraction and discrimination performance that need to design, the design feature selection algorithm reduces intrinsic dimensionality again, improves efficiency of algorithm.
At first, as the scheme of alternative, described ground reaction force for training and identification can be the ground reaction force of single step, or the ground reaction force of continuous a plurality of single steps cumulative synthetic ground reaction force that obtains on sequential.Comprise abundant gait information in ground reaction force during walking, except the time domain charactreristic parameter of the reflection gait cycle that can see from the ground reaction force curve and globality, the feature of more reflection walking frequency spectrums and local detail is hidden on frequency domain.Frequency-domain analysis method is the analytical approach of signal common, can go out to excavate the feature that signal can't be observed on time domain.Wavelet transformation is as a kind of outstanding signal time frequency analyzing tool, the characteristic of its multiresolution analysis makes our have the ability to characterize in the time-frequency two territories local feature of gaits, and wavelet package transforms more is better than time of wavelet transformation and frequency resolution and more can be competent at our demand with it.WAVELET PACKET DECOMPOSITION has been widely used in signal decomposition, de-noising, coding and compression etc., also is used for extracting the cancer classification feature from gene expression profile data.Therefore, adopt wavelet packet decomposition algorithm to carry out multiscale analysis to ground reaction force in the present invention, with the average of the WAVELET PACKET DECOMPOSITION coefficient of every sub-wavelet packet, wavelet-packet energy, WAVELET PACKET DECOMPOSITION coefficient and variance not only integral body but also characterize subtly gait feature from the time-frequency domain, and they are used for identification.
The step of Method of Gait Feature Extraction comprises:
(1) at first adopt the piecewise linear interpolation algorithm to carry out dimension normalization to selecting pretreated ground reaction force data through denoising and effective sample, the data dimension of effective sample is normalized to same value.When forming a gait passage by Np three-dimensional force force plate/platform, the same value that in the present embodiment, the gait data dimension of the ground reaction force of single step normalizes to is calculated by 500, and the same value that the gait data dimension of synthetic ground reaction force normalizes to is calculated by 500 * Np.Take the gait passage that formed by 5 three-dimensional force force plate/platforms as example, Figure 12 is left and right directions shearing force Fhxd, fore-and-aft direction shearing force Fhyd in synthetic ground reaction force after denoising and dimension normalization and the force-time curve of vertical direction anchorage force Fhzd, be to adopt the piecewise linear interpolation algorithm to carry out interpolation to the synthetic ground reaction force data of Fig. 5, data dimension is normalized to force-time curve after 2500 dimensions.
(2) ground reaction force after adopting again wavelet packet decomposition algorithm to dimension normalization carries out L layer WAVELET PACKET DECOMPOSITION, decomposes to obtain 2
L+1-2 sub-wavelet packets characterize gait feature with average and the variance of the WAVELET PACKET DECOMPOSITION coefficient of every sub-wavelet packet, wavelet-packet energy, WAVELET PACKET DECOMPOSITION coefficient.When carrying out WAVELET PACKET DECOMPOSITION, determine in advance wavelet basis function and decompose the number of plies.The selection of wavelet basis function usually needs by virtue of experience to determine with experiment, and the wavelet basis function commonly used that not only can be used for continuous wavelet transform at present but also can be used for wavelet transform has five kinds of Haar, Daubechies, Biorthgonal, Coiflets and Symlets.Adopt the inventive method, found through experiments, when selecting the Coif1 small echo to carry out WAVELET PACKET DECOMPOSITION, discrimination is higher.Decompose the number of plies and usually determined by signal low-limit frequency and highest frequency, the highest frequency of the ground reaction force signal that collects is in 400Hz, and the eigenfrequency of the ground reaction force of gait is decomposed the number of plies in the 30Hz left and right
Can select 4 layers of WAVELET PACKET DECOMPOSITION.Ground reaction force for each sample in the present embodiment carries out 4 layers of WAVELET PACKET DECOMPOSITION successively, obtains 2
L+1-2 sub-wavelet packets are preserved the WAVELET PACKET DECOMPOSITION coefficient of all sub-wavelet packets, with the average of the WAVELET PACKET DECOMPOSITION coefficient of each sub-wavelet packet, wavelet-packet energy, WAVELET PACKET DECOMPOSITION coefficient and variance as initial gait feature collection.Figure 13 is through the vertical direction anchorage force Fwz curve after denoising, weight standard and dimension normalized, Figure 14 carries out 4 layers of second layer metric space after WAVELET PACKET DECOMPOSITION to vertical direction anchorage force Fwz, and Figure 15 carries out 4 layers of second layer wavelet space after WAVELET PACKET DECOMPOSITION to vertical direction anchorage force Fwz.Wherein, Figure 14 and Figure 15 have showed the example that adopts the vertical direction anchorage force extraction gait feature of 4 layers of WAVELET PACKET DECOMPOSITION method from synthetic ground reaction force.First with the wavelet transformation hard threshold method, original gait data is carried out denoising, again data are carried out standardization and data dimension is normalized to 2500 dimensions divided by body weight, the force-time curve of the vertical direction anchorage force Fwz after these pretreatment operation of process as shown in figure 13; And then it is carried out 4 layers of WAVELET PACKET DECOMPOSITION, vertical direction anchorage force Fwz is resolved into metric space and wavelet space; Metric space is the compression of original signal, comprises the low-frequency information of original signal, and Figure 14 carries out 4 layers of second layer metric space after WAVELET PACKET DECOMPOSITION to vertical direction anchorage force Fwz; Wavelet space is the complementary space of metric space, comprises the high-frequency information of original signal, and Figure 15 carries out 4 layers of second layer wavelet space after WAVELET PACKET DECOMPOSITION to vertical direction anchorage force Fwz; Each space is a sub-wavelet packet, the coefficient of every sub-wavelet packet is the WAVELET PACKET DECOMPOSITION coefficient, again the WAVELET PACKET DECOMPOSITION coefficient of every sub-wavelet packet is carried out integration on frequency band and obtain wavelet-packet energy, ask average and the variance of the WAVELET PACKET DECOMPOSITION coefficient of every sub-wavelet packet, at last with the average of the WAVELET PACKET DECOMPOSITION coefficient of all sub-wavelet packets and wavelet-packet energy, WAVELET PACKET DECOMPOSITION coefficient and variance as initial gait feature collection.
(3) then in training process, the WAVELET PACKET DECOMPOSITION coefficient that adopts FCM Algorithms that initial gait feature is concentrated carries out Feature Dimension Reduction, pick out the optimal wavelet bag coefficient of dissociation that redundancy is few and classification capacity is strong, preserve with optimal wavelet bag coefficient of dissociation and corresponding to the sequence number (call number) of initial gait feature collection, and the wavelet-packet energy of every sub-wavelet packet, the average of WAVELET PACKET DECOMPOSITION coefficient and the gait feature template that the variance combination obtains the gait feature collection and trained;
(4) according to the call number in the gait feature template of having trained, the initial gait feature collection of test sample book is carried out Feature Mapping in identifying, pick out the gait feature for identification.
4. the training of sorter
In pattern recognition system, feature extraction and selection are the bases, and the selection of sorter and training are also and very crucial.Traditional statistical pattern recognition method is to set up disaggregated model under abundant sample, and when only having sample number to be tending towards infinity, its classification performance just can have theoretic guarantee, and in practical application, sample number is but limited retrievable.Being based upon the learning algorithm on structuring principle of minimization risk basis---support vector machine (SVM:Support Vector Machines) has outstanding performance in solving small sample, non-linear and higher-dimension pattern classification problem, therefore, the present invention adopts support vector machine as sorter.
Support vector machine proposes for two minutes problems, for multi-class classification problem, need to carry out PROBLEM DECOMPOSITION and re-use support vector machine and classify.Suppose to have in the present invention M class training sample, the present invention adopts the one-to-many combined strategy to process many classification problems, is about to 1 pair of M-1 combination of M class training sample, and training obtains N support vector machine respectively, determines classification results according to nearest neighbouring rule.Suppose to have registered n class (individual) gait sample in the gait data storehouse, these samples input support vector machine classifiers are trained, obtain output and a classification model corresponding to 1～n.In identifying, the real-time gait sample that gets is input in trained support vector machine classifier, namely mates with the classification model of having trained, if output valve is between 1～n, assert that unknown object is the competent person who registered, otherwise be judged to be unregistered personnel.
The present invention is adopting before support vector machine classifier trains training sample, first adopts maximum-minimum requirement method to carry out normalization on amplitude to the gait feature of input support vector machine.The unique distinction of support vector machine just is to adopt the kernel function of Nonlinear Mapping the lower dimensional space of linearly inseparable to be mapped to the higher dimensional space of linear separability, should select suitable kernel function according to problem, and kernel function commonly used has three kinds at present:
Polynomial kernel function K (x, x')=((xx')+1)
d, d〉and 1, need to determine in advance (5)
Radial basis kernel function K (x, x')=exp (γ || x-x'||
2), γ〉0, need to determine in advance (6)
Sigmoid core K (x, x')=tanh (β
0(xx')+β
1) (7)
When the d in the polynomial kernel function was 1, the polynomial kernel function became linear kernel function.With Sigmoid nuclear phase ratio, the parameter that polynomial kernel function and radial basis kernel function need to be inputted is few.For Nonlinear Classification, the radial basis kernel function is best selection, choice for use radial basis kernel function of the present invention.In addition, when selecting the radial basis kernel function, the selection of a pair of parameter of kernel function---penalty factor and kernel functional parameter γ also has certain influence to classification performance.Penalty factor is used for controlling model and mistake is divided the punishment degree of sample, decision model complexity and generalization ability.Kernel functional parameter γ determines the complexity (dimension) of the rear feature space of mapping, thereby affects the minimum experience error of classifying face.Therefore, the parameter of radial basis kernel function need to be adjusted to suitable value to (C, γ) during use, the present invention adopts the grid search algorithm to determine that the kernel functional parameter of support vector machine is to (C, γ).
5. the identification of unknown object and recognition result
After completing the training of support vector machine classifier, complete the identification to unknown object in identifying.In identifying, the real-time gait sample of the unknown object that gets is input in trained support vector machine classifier, namely mate with the classification model of having trained, if output valve is between the training result 1～n of registered sample, assert that unknown object is the competent person who registered, otherwise be judged to be unregistered personnel.
In order to test the validity of institute of the present invention extracting method, ground reaction force data when having gathered 103 people's self-selected speeds in the present embodiment experiment and wearing footwear and naturally walk, everyone gathers 12 effective samples, amounts to 1236 effective samples, has set up 103 people's gait data storehouse.For estimating recognition performance, use correct recognition rata CRR(Correct Recognition Rate) as evaluation index.Extract gait feature from synthetic ground reaction force, adopt respectively the support vector machine of different IPs function as sorter, carry out 4 repeated tests, when find adopting support vector machine based on the radial basis kernel function as sorter, correct recognition rata CRR is the highest, reaches 95.3% average correct recognition rata.Therefore, the present invention is feasible and effective.What is more important, the present invention is disguised and anti-pretends to be ability good, recognition result is not subjected to that complex background, clothes and health block etc., and disturbing factor affects, can be the well integrated identity that is applied to cyberage security sensitive place differentiate and security access control system in.
Embodiment 2:
In order to further describe the specific embodiment of the present invention, the present embodiment is applied as example with the entrance guard management of certain bank vault and is illustrated.
Bank vault is the focused protection place that a designated person just has the turnover authority; also that the offender thinks into the place of stealing very much; even as describe in film, they are by snatching password and key, disguising oneself or the means such as fingerprint, iris of copying enter national treasury and steal.Therefore, can by tight concordant erection method shown in Figure 3, the gait passage of 3 meters long that is comprised of 6 three-dimensional strength measurement platforms, identification and the entrance guard management work of ground reaction force when utilizing the present invention to carry out based on walking be installed by the subsurface before national treasury entrance antitheft door.If the visitor walks the out-of-date personnel that enter of allowing that can be identified as appointment from the gait passage, the antitheft door of encrypted pinning is opened automatically, allows the visitor to enter; Otherwise, will be denied access to, and superior management and control unit gives the alarm, automatically snap simultaneously and transmit image scene to higher level management and control unit.
In the entrance guard management of bank vault is used, at first need the sample training with the designated person, the gait feature template and the classification model that obtain having trained, identifying is identified the visitor by template again, whether identify visiting person and be and allow the people that enters, whether recognition result is as order or the signal controlling anti-theft door for national treasury and open.
For training process, at first on-demand creation designated person's gait data storehouse, utilization is arranged on the ground reaction force of the some when underground gait passage collection designated person repeatedly walks before bank vault entrance antitheft door on passage, each ground reaction force data that gather everyone continuous a plurality of single step walkings on the gait passage, obtain as shown in Figure 5 synthetic ground reaction force by superposition calculation, and as the gait data sample; For each designated person, utilize above-mentioned effective sample selection method to pick out effective gait data sample from the gait data sample that collects, with one of all designated persons' gait data Sample Establishing gait data storehouse, the later stage regularly or irregularly upgrades or increases the gait data storehouse.For the gait data that has been collected, the ground reaction force data of effectively synthesizing of namely picking out, at first adopt the Wavelet Transform Threshold method to carry out noise suppression preprocessing to it, then adopt above-mentioned gait feature abstracting method to extract gait feature, adopt again FCM Algorithms that gait feature is selected, the gait feature template that obtains having trained; Adopt again support vector machine classifier that training sample is trained, the classification model that obtains having trained, adopting before support vector machine classifier trains training sample, can first adopt maximum-minimum requirement method to carry out normalization on amplitude to the gait feature of input support vector machine.
For identifying, at first utilize and be arranged on before bank vault entrance antitheft door underground gait passage collection visitor and walk out-of-date ground reaction force from the gait passage is up, by the quick synthetic ground reaction force that obtains as shown in Figure 5 of superposition calculation, and as test sample book; Adopt again said method to carry out effective sample and select and the data pre-service such as denoising, adopt said method to carry out Method of Gait Feature Extraction, utilize the gait feature template of having trained to carry out Feature Mapping (being feature selecting) to gait feature to be identified; Then adopt maximum-minimum requirement method to carry out normalization on amplitude to gait feature to be identified, adopt support vector machine test gait feature to be identified to mate with the classification model of having trained, if mate fully with certain already present classification model in the gait data storehouse, the visitor is the personnel that enter of allowing of appointment so, give an order or signal is opened the antitheft door of encrypted pinning automatically, allow the visitor to enter; If do not mate with all classification models that existed in the gait data storehouse, the visitor is considered to not allow the personnel that enter so, do not send order or the signal of opening antitheft door, refusing it enters, and superior management and control unit gives the alarm, and automatically snaps simultaneously and transmits image scene to higher level management and control unit.
In a word, the ground reaction force when the present invention is based on walking has proposed a kind of simple and effective remote hidden personal identification method.At first, utilization can free assembly unit and the ground reaction force can be hidden in gait passage under ground/floor and obtain walking the time, adopts the Wavelet Transform Threshold method to carry out noise suppression preprocessing to ground reaction force; Adopt again the feature extracting method that proposes to extract gait feature, and adopt feature selecting algorithm to select optimum gait feature; Then, adopt support vector machine training classification model on the gait data storehouse; At last, according to nearest neighbouring rule, test sample book is carried out template matches and identification in identifying, the output recognition result.Test result on 103 self-built people gait data storehouse has been verified feasibility and the algorithm complexity of the inventive method, and the present invention is convenient and easy, has robustness.Because the gait information of utilization of the present invention can be by being hidden in the force-sensing sensor collection under ground/floor, gatherer process has untouchable and disguised, be identified object and can not realize identification apparatus is arranged and take measures to get by under false pretences, also can not cause human rights dispute; Simultaneously, with respect to the gait identification method of computer vision field, recognition result is not subjected to that complex background, clothes and health block etc. to be affected, and the vola force information also hardly may be imitated; The present invention combines these advantages, can satisfy better that cyberage airport, customs, museum, party are on-the-spot, the identity in the security sensitive such as bank and national treasury place differentiates and the safety precaution demand.Simultaneously, the present invention can also provide supplementary and analytical approach for the apparatus Computer Aided Designs such as sports and scientific research, medical rehabilitation, productive life body-building and criminal investigation.
Claims (3)
1. the personal identification method of a ground reaction force during based on walking, is characterized in that comprising training process and identifying;
Described training process comprises: utilize the gait information when being hidden in the gait passage collection walking that ground or underfloor three-dimensional force force plate/platform form and set up the gait data storehouse, carry out data pre-service, Method of Gait Feature Extraction, gait feature selection and sorter training, finally obtain gait feature template and the classification model of having trained;
Described identifying comprises: obtain the real-time gait information of object to be identified to obtain test sample book, test sample book is carried out data pre-service, Method of Gait Feature Extraction, utilize the gait feature template of having trained to carry out Feature Mapping to the gait feature that object to be identified extracts, the classification model that recycling has been trained carries out template matches and identification according to nearest neighbouring rule to test sample book, the output recognition result;
Described gait information is the ground reaction force by the single step of three-dimensional force force plate/platform collection, or the ground reaction force of continuous a plurality of single steps cumulative ground reaction force that obtains synthesizing on sequential;
The ground reaction force of described single step or synthetic ground reaction force all are comprised of these three components of left and right directions shearing force, fore-and-aft direction shearing force and vertical direction anchorage force;
Described data pre-service comprises that denoising and effective sample select; Described denoising refers to adopt the Wavelet Transform Threshold method that the ground reaction force data are decomposed and reconstruction processing; Whether described effective sample is selected the data dimension and the peak point numerical value that refer to according to vertical direction anchorage force in the ground reaction force after denoising and is judged in effective range, selects ground reaction force data in effective range as effective sample; Described effective range refers to the numerical range relevant with sample frequency Fp to the number Np of the three-dimensional force force plate/platform that adopts, ground reaction force for single step, if wherein the data dimension of vertical direction anchorage force is lower than 0.4Fp or higher than 0.8Fp, or the ordinate value of the crest of vertical direction anchorage force and trough differs over 300, namely be judged to be invalid sample, otherwise be judged to be effective sample; For the synthetic ground reaction force of the continuous a plurality of single steps that gathered by Np three-dimensional force force plate/platform, if wherein the data dimension of vertical direction anchorage force is lower than 0.4Fp * Np or higher than 0.8Fp * Np, or the number of the trough of vertical direction anchorage force is less than Np, or the maximin of the ordinate value of all troughs of vertical direction anchorage force differs over 300, namely be judged to be invalid sample, otherwise be judged to be effective sample;
Described Method of Gait Feature Extraction refers to that the average and the variance that adopt Wavelet Packet Transform Method to extract WAVELET PACKET DECOMPOSITION coefficient, wavelet-packet energy, WAVELET PACKET DECOMPOSITION coefficient from ground reaction force characterize gait feature;
Described sorter training refers to adopt support vector machine classifier that training sample is trained, the classification model that obtains having trained.
2. by the described personal identification method of ground reaction force during based on walking of claim 1, be characterised in that the step of described Method of Gait Feature Extraction is:
At first adopt the piecewise linear interpolation algorithm to carrying out dimension normalization through data pretreated ground reaction force data, the data dimension of effective sample is normalized to same value;
Effective sample after adopting again L layer wavelet packet decomposition algorithm to denoising and dimension normalization carries out WAVELET PACKET DECOMPOSITION, decomposes to obtain 2
L+1-2 sub-wavelet packets characterize gait feature with average and the variance of the WAVELET PACKET DECOMPOSITION coefficient of every sub-wavelet packet, wavelet-packet energy, WAVELET PACKET DECOMPOSITION coefficient;
Then adopt feature selecting algorithm that the WAVELET PACKET DECOMPOSITION coefficient is selected, pick out optimal wavelet bag coefficient of dissociation, with average and the variance combination of optimal wavelet bag coefficient of dissociation and wavelet-packet energy, WAVELET PACKET DECOMPOSITION coefficient, the gait feature template that obtains the gait feature collection and trained.
3. by the described personal identification method of ground reaction force during based on walking of claim 1, be characterised in that the training of described sorter first adopts maximum-minimum requirement method to carry out normalization to the gait feature of input support vector machine on amplitude, adopt again support vector machine classifier that training sample is trained, the classification model that obtains having trained.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210559665.7A CN103093234B (en) | 2012-12-21 | 2012-12-21 | Based on the personal identification method of ground reaction force during walking |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210559665.7A CN103093234B (en) | 2012-12-21 | 2012-12-21 | Based on the personal identification method of ground reaction force during walking |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103093234A true CN103093234A (en) | 2013-05-08 |
| CN103093234B CN103093234B (en) | 2016-01-06 |
Family
ID=48205782
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201210559665.7A Active CN103093234B (en) | 2012-12-21 | 2012-12-21 | Based on the personal identification method of ground reaction force during walking |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103093234B (en) |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104434128A (en) * | 2014-12-25 | 2015-03-25 | 中国科学院合肥物质科学研究院 | Left and right foot dynamic recognition method based on plantar pressure distribution information |
| CN106681487A (en) * | 2015-11-05 | 2017-05-17 | 三星电子株式会社 | Walking assistance apparatus and method of controlling same |
| CN106953810A (en) * | 2017-03-13 | 2017-07-14 | 成都育芽科技有限公司 | The call method and device of template in a kind of WTC communication protocols |
| CN107028427A (en) * | 2017-06-08 | 2017-08-11 | 翼猫科技发展（上海）有限公司 | The method for sensing of the sleep system of new automated intelligent regulation |
| CN107066945A (en) * | 2017-03-10 | 2017-08-18 | 清华大学 | A kind of quick identity checking method of big flow clearance and system |
| CN107095677A (en) * | 2016-02-19 | 2017-08-29 | 财团法人资讯工业策进会 | gait analysis system and method |
| CN108053587A (en) * | 2018-01-08 | 2018-05-18 | 京东方科技集团股份有限公司 | Personal identification method and its device, carpet |
| CN109274930A (en) * | 2018-07-26 | 2019-01-25 | 华为技术有限公司 | A kind of method for safety monitoring and device based on Gait Recognition |
| CN109359580A (en) * | 2018-10-12 | 2019-02-19 | 中国科学院福建物质结构研究所 | Footprint based on deep learning identifies and gait detection method and its device |
| CN109447199A (en) * | 2018-10-16 | 2019-03-08 | 山东大学 | A kind of multi-modal criminal's recognition methods and system based on step information |
| CN109646011A (en) * | 2018-10-15 | 2019-04-19 | 王长贵 | A kind of voltage inductance answers compound tile |
| CN109740423A (en) * | 2018-11-22 | 2019-05-10 | 霍尔果斯奇妙软件科技有限公司 | Ethnic recognition methods and system based on face and wavelet packet analysis |
| CN112198222A (en) * | 2020-09-17 | 2021-01-08 | 江苏美的清洁电器股份有限公司 | Ground material identification method, system, equipment and storage medium |
| CN115553757A (en) * | 2022-12-06 | 2023-01-03 | 深圳市第二人民医院(深圳市转化医学研究院) | Method and equipment for intelligently screening effective ground reaction force in gait process |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070198194A1 (en) * | 2005-12-19 | 2007-08-23 | Analiza, Inc. | Systems and methods involving data patterns such as spectral biomarkers |
-
2012
- 2012-12-21 CN CN201210559665.7A patent/CN103093234B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070198194A1 (en) * | 2005-12-19 | 2007-08-23 | Analiza, Inc. | Systems and methods involving data patterns such as spectral biomarkers |
Non-Patent Citations (2)
| Title |
|---|
| 姚志明: "基于步态触觉信息的身份识别研究", 《中国博士学位论文全文数据库》, 30 June 2011 (2011-06-30) * |
| 林尔东，姚志明，郑重，周旭，孙向阳，孙怡宁: "一种改进的机遇地面反作用力的步态识别方法", 《模式识别与人工智能》, vol. 24, no. 3, 30 June 2011 (2011-06-30) * |
Cited By (21)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104434128A (en) * | 2014-12-25 | 2015-03-25 | 中国科学院合肥物质科学研究院 | Left and right foot dynamic recognition method based on plantar pressure distribution information |
| CN106681487A (en) * | 2015-11-05 | 2017-05-17 | 三星电子株式会社 | Walking assistance apparatus and method of controlling same |
| CN106681487B (en) * | 2015-11-05 | 2021-05-04 | 三星电子株式会社 | Walking assistance apparatus and method of controlling walking assistance apparatus |
| CN107095677A (en) * | 2016-02-19 | 2017-08-29 | 财团法人资讯工业策进会 | gait analysis system and method |
| CN107095677B (en) * | 2016-02-19 | 2019-12-03 | 财团法人资讯工业策进会 | Gait analysis system and method |
| CN107066945B (en) * | 2017-03-10 | 2019-06-18 | 清华大学 | A kind of quick identity checking method of big flow clearance and system |
| CN107066945A (en) * | 2017-03-10 | 2017-08-18 | 清华大学 | A kind of quick identity checking method of big flow clearance and system |
| CN106953810A (en) * | 2017-03-13 | 2017-07-14 | 成都育芽科技有限公司 | The call method and device of template in a kind of WTC communication protocols |
| CN107028427B (en) * | 2017-06-08 | 2020-02-21 | 翼猫科技发展（上海）有限公司 | Novel sensing method of sleep system capable of automatically and intelligently adjusting |
| CN107028427A (en) * | 2017-06-08 | 2017-08-11 | 翼猫科技发展（上海）有限公司 | The method for sensing of the sleep system of new automated intelligent regulation |
| CN108053587A (en) * | 2018-01-08 | 2018-05-18 | 京东方科技集团股份有限公司 | Personal identification method and its device, carpet |
| CN109274930A (en) * | 2018-07-26 | 2019-01-25 | 华为技术有限公司 | A kind of method for safety monitoring and device based on Gait Recognition |
| CN109274930B (en) * | 2018-07-26 | 2020-06-26 | 华为技术有限公司 | Safety monitoring method and device based on gait recognition |
| CN109359580A (en) * | 2018-10-12 | 2019-02-19 | 中国科学院福建物质结构研究所 | Footprint based on deep learning identifies and gait detection method and its device |
| CN109359580B (en) * | 2018-10-12 | 2021-05-11 | 中国科学院福建物质结构研究所 | Footprint identification and gait detection method and device based on deep learning |
| CN109646011A (en) * | 2018-10-15 | 2019-04-19 | 王长贵 | A kind of voltage inductance answers compound tile |
| CN109646010A (en) * | 2018-10-15 | 2019-04-19 | 王长贵 | A kind of gait detecting system and method based on compound tile |
| CN109447199A (en) * | 2018-10-16 | 2019-03-08 | 山东大学 | A kind of multi-modal criminal's recognition methods and system based on step information |
| CN109740423A (en) * | 2018-11-22 | 2019-05-10 | 霍尔果斯奇妙软件科技有限公司 | Ethnic recognition methods and system based on face and wavelet packet analysis |
| CN112198222A (en) * | 2020-09-17 | 2021-01-08 | 江苏美的清洁电器股份有限公司 | Ground material identification method, system, equipment and storage medium |
| CN115553757A (en) * | 2022-12-06 | 2023-01-03 | 深圳市第二人民医院(深圳市转化医学研究院) | Method and equipment for intelligently screening effective ground reaction force in gait process |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103093234B (en) | 2016-01-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103093234B (en) | Based on the personal identification method of ground reaction force during walking | |
| CN103049741A (en) | Foot-to-ground acting force-based gait feature extraction method and gait identification system | |
| Chelli et al. | A machine learning approach for fall detection and daily living activity recognition | |
| Doukas et al. | Advanced patient or elder fall detection based on movement and sound data | |
| Rong et al. | A wearable acceleration sensor system for gait recognition | |
| Wan et al. | A survey on gait recognition | |
| Kao et al. | Development of a portable activity detector for daily activity recognition | |
| Mubashir et al. | A survey on fall detection: Principles and approaches | |
| Rong et al. | Identification of individual walking patterns using gait acceleration | |
| Shan et al. | A wearable pre-impact fall detector using feature selection and support vector machine | |
| CA2835460C (en) | System and method for enabling continuous or instantaneous identity recognition based on physiological biometric signals | |
| CN104573458A (en) | Identity recognition method, device and system based on electrocardiogram signals | |
| CN111178155B (en) | Gait feature extraction and gait recognition method based on inertial sensor | |
| Sasikala et al. | Identification of individuals using electrocardiogram | |
| He | Activity recognition from accelerometer signals based on wavelet-ar model | |
| Hestbek et al. | Biometric gait recognition for mobile devices using wavelet transform and support vector machines | |
| CN107293301B (en) | Recognition methods and system based on dental articulation sound | |
| CN105139029A (en) | Activity recognition method and activity recognition device for persons serving sentences | |
| Nandy et al. | Novel features for intensive human activity recognition based on wearable and smartphone sensors | |
| CN110415152A (en) | A kind of safety monitoring system | |
| CN113609977A (en) | Pedestrian gait recognition method based on channel state information quotient distance | |
| Guelta et al. | An Improved Behavioral Biometric System based on Gait and ECG signals. | |
| Lu et al. | An EMG-based personal identification method using continuous wavelet transform and convolutional neural networks | |
| Karthikeyan et al. | Enhancement of multi-modal biometric authentication based on iris and brain neuro image coding | |
| CN106971203A (en) | Personal identification method based on characteristic on foot |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant |