CN106203391A - Face identification method based on intelligent glasses - Google Patents
Face identification method based on intelligent glasses Download PDFInfo
- Publication number
- CN106203391A CN106203391A CN201610586532.7A CN201610586532A CN106203391A CN 106203391 A CN106203391 A CN 106203391A CN 201610586532 A CN201610586532 A CN 201610586532A CN 106203391 A CN106203391 A CN 106203391A
- Authority
- CN
- China
- Prior art keywords
- face
- intelligent glasses
- image
- method based
- identification method
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/168—Feature extraction; Face representation
- G06V40/171—Local features and components; Facial parts ; Occluding parts, e.g. glasses; Geometrical relationships
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/172—Classification, e.g. identification
Abstract
The invention discloses a kind of face identification method based on intelligent glasses, comprise the steps: a) to arrange on intelligent glasses photo module, microprocessor and wireless communication module；B) obtained the face-image of user by the photo module on intelligent glasses, the microprocessor on recycling intelligent glasses carries out recognition of face to the face-image obtained；C) by wireless communication module, face recognition result is uploaded to central database compare, it is judged that the true identity of user, and returns the relevant information of this user.The face identification method based on intelligent glasses that the present invention provides; by glasses user in real image and be identified, then connect network and carry out the mutual comparison of data information with given server, prevent demographic data's information from flowing out; protection personnel's privacy, the problem effectively solving information leakage.
Description
Technical field
The present invention relates to a kind of face identification method, particularly relate to a kind of face identification method based on intelligent glasses.
Background technology
The modes such as traditional personal verification means such as password, certificate, IC-card, due to the separability with identity people,
Cause forgery, usurp, the phenomenon such as decoding happens occasionally, can not meet that modern social economy is movable and social safety has been taken precautions against
Need.Living things feature recognition includes fingerprint, palmmprint, voice, face, iris, gait, vena metacarpea etc..Biometrics identification technology
First putting into wide variety of is fingerprint, palmmprint scanning recognition technology, but the most usually because by skin texture and degree of drying
Deng conditionality, erroneous judgement occurs, cause unnecessary trouble, the demand of people can not be met the most far away.Along with science and technology not
Disconnected development, and society is for the highest requirement of identification, biometrics identification technology gradually in diversified development, example
Such as face recognition technology.
Along with improving constantly of electronic information degree, public security department is increasingly utilized informationization to process
Confidential information, strengthen communication, convenient link up while add information illegally divulge a secret and go beyond one's commission the security risks such as use in inside.
Conditional electronic document is little affected by any authority and limits, and deposits in plain text, arbitrarily reads, revises, replicates, prints or distribution etc., this
A little main causes causing information-leakage just.
Summary of the invention
The technical problem to be solved is to provide a kind of face identification method based on intelligent glasses, it is possible to pass through
Glasses user in real image, and connect network and given server and carry out the mutual comparison of data information, prevent demographic data
Information flows out, and protects personnel's privacy, the problem effectively solving information leakage.
The present invention solves that above-mentioned technical problem employed technical scheme comprise that a kind of face based on intelligent glasses of offer
Recognition methods, comprises the steps: a) to arrange on intelligent glasses photo module, microprocessor and wireless communication module；B) logical
Cross the photo module on intelligent glasses and obtain the face-image of user, the face to obtaining of the microprocessor on recycling intelligent glasses
Portion's image carries out recognition of face；C) by wireless communication module, face recognition result is uploaded to central database compare,
Judge the true identity of user, and return the relevant information of this user.
Above-mentioned face identification method based on intelligent glasses, wherein, before described photo module is for being suspended at intelligent glasses
One photographic head of side, described microprocessor and wireless communication module are positioned at picture frame left and right sides, on the eyeglass of described intelligent glasses
It is equipped with one and wears the type display screen that declines.
Above-mentioned face identification method based on intelligent glasses, wherein, described intelligent glasses also sets up GPS module,
While obtaining user related information, also this customer location is positioned, and show electronics on wear-type miniature display screen
Map and the track of this customer location change.
Above-mentioned face identification method based on intelligent glasses, wherein, described wireless communication module include WiFi module and/
Or bluetooth module, the mirror holder of described intelligent glasses arranges switch switch, takes pictures, image two-pass button.
Above-mentioned face identification method based on intelligent glasses, wherein, described step b) also includes the face figure obtained
As carrying out following pretreatment: coloured image is first converted to gray level image, and each pixel of described coloured image is made by index
With unified palette；Then use rectangular histogram that the gray level image after conversion is carried out equilibrium treatment, enhancing contrast ratio；Then adopt
With field averaging method, the gray level image after conversion is carried out mean filter；Median filtering method is finally used to remove in gray level image
Isolated noise point.
Above-mentioned face identification method based on intelligent glasses, wherein, described step b) uses based on local characteristic region
One training sample carry out recognition of face, comprise the steps: b1) differentiate in piece image whether there is face, if there is people
It is face region that face then defines regional area；B2) local feature region of face region is extracted, and by local feature region
Classify by face position, shape of face and the angle of face, it is thus achieved that face local feature region；B3) each face local feature region
A corresponding grader, uses the assembled classifier face local feature region ballot to extracting or linear weighted function mode to carry out phase
Seemingly spend identification.
Above-mentioned face identification method based on intelligent glasses, wherein, the local feature region of described face region
Extraction process is as follows: obtain the facial image information of measurement space；Use principal component method, high dimension vector is passed through feature
Vector matrix projects in the vector space of low-dimensional；By the projection coefficient obtained in low gt and known facial image
Model comparision, thus obtain testing result.
Above-mentioned face identification method based on intelligent glasses, wherein, the face-image in described step b) is people from side
Face image, described step b1) utilize integral projection method that pretreated binaryzation Side Face Image is carried out human face region
Location；During side face characteristic extracts, integral projection method and face priori is utilized to obtain each principal character of face
Essential information, and by these information with one group of geometric properties vector representation, represent side face with this；Knowledge at side face
During Bie, the sorting technique of minimum distance classifier and the method for RBF artificial neural network is used to carry out judging to identify.
The present invention contrasts prior art a following beneficial effect: the recognition of face based on intelligent glasses that the present invention provides
Method, by glasses user in real image and be identified, then connects network and carries out data information with given server
Mutual comparison, prevents demographic data's information from flowing out, and protects personnel's privacy, the problem effectively solving information leakage.
Accompanying drawing explanation
Fig. 1 is present invention recognition of face based on intelligent glasses schematic flow sheet.
Detailed description of the invention
The invention will be further described with embodiment below in conjunction with the accompanying drawings.
Fig. 1 is present invention recognition of face based on intelligent glasses schematic flow sheet.
Refer to Fig. 1, the face identification method based on intelligent glasses that the present invention provides, comprise the steps:
Step S1: photo module, microprocessor and wireless communication module are set on intelligent glasses；
Step S2: obtained the face-image of user by the photo module on intelligent glasses, on recycling intelligent glasses
Microprocessor carries out recognition of face to the face-image obtained；
Step S3: face recognition result is uploaded to central database by wireless communication module and compares, it is judged that
The true identity of user, and return the relevant information of this user.
The present invention use intelligent glasses specifically include that glasses front suspension a photographic head and one be positioned at picture frame
The computer processor device of the wide strip on right side, the photographic head pixel of outfit is 8,000,000 pixel SONY sensors, 1080p video
Video recording and broadcasting.Being equipped with one on eyeglass and wear the type display screen that declines, it can be by above data projection to user's right eye
In the small screen.Display effect is as watched 35 cun of high definition screens at 2.5 meters.Use switch switch, take pictures, image two-pass by
Key, mirror holder is collapsible.Battery can be supported the stand-by time normal use of 150 hours or record 1080P HD video one continuously
Individual hour, charging can use Micro USB interface or custom-designed charger, and it is many that Micro USB fills electric data/earphone
Unification interface.According to ambient sound display distance and direction on screen, two pieces of eyepieces show map and navigation letter respectively
The product of breath technology.
The weight of the intelligent glasses that the present invention uses only has tens grams, inside saves as 1GB, and operating system and the software of use are
An-droid 4.4, LLVision customize Intelligent wearable UI.Network connects supports WiFi802.11b/g/n, 2.4GHz WIFI
Connect with Bluetooth BT4.0 low-power consumption bluetooth.Total memory capacity is 16GB, about 1000 photos or 90 minutes 1080p
Video.4 core 1.2G Cortex A7 processors, have automatic Load administrative mechanism, accomplish the perfectly balanced of performance and power consumption；In
Put PowerVR hardware GPU, masterly to complicated video algorithms and image procossing.
For the photo module of intelligent glasses, present invention optimizes form conversion and the pressure of the image that intelligent glasses collects
Compression method.The image that image capture device gathers generally preserves with bmp or jpg form.As a example by bmp formatted file, bmp literary composition
Part is generally divided into four parts: header file, information header, palette, data division.One secondary 24 rgb image each pixel by
Three byte representations, the image so storing a pair 640 3 480 is accomplished by taking 921600 bytes, no matter for memory capacity
Not ideal, accordingly, it would be desirable to image is compressed with the time of process.The method used is that coloured image is converted to ash
Degree image, is defined in the cromogram of rgb space, and the color of each of which pixel is together decided on by tri-components of R, G, B.Each point
Amount has together decided on picture depth in the figure place shared by internal memory, the byte number shared by the most each pixel.Its file memory format
For BITMAPFILEHEADER+BITMAPINFOHEADER, immediately following being probably below: a) if 24 true coloured pictures, the most often
Individual point is to be represented R/G/B respectively by three bytes, so the most directly following the color information of image；B) if 8 (256
Color), 4 (16 color), 1 (monochromatic) figure, then closelying follow is palette data below, the array of a RGBQUAD type, and it is long
Degree is determined by BITMAPINFOHEADER.biClrUsed.Then followed by the view data that is only (24 bitmaps are true
View data, other are then the index datas of palette).
Gray-scale map refers to containing only monochrome information, without the image of color information, the black-and-white photograph as seeing at ordinary times: brightness
By secretly to bright, change is continuous print.Therefore, gray-scale map to be represented, it is necessary to brightness value is quantified.It is commonly divided into 0 to arrive
255 totally 256 ranks, wherein 0 the darkest (completely black), 255 the brightest (the whitest).In the method representing color, in addition to RGB, also
There is a kind of method for expressing being YUV, apply the most a lot.In TV signal is exactly a kind of color side of expression being similar to YUV
Method.In this method for expressing, the physical meaning of Y-component is exactly brightness, and Y-component contains all information of gray-scale map, only uses Y
Component just can be fully able to represent a width gray-scale map.
Y conversion formula from RGB to yuv space is: Y=R30.299+G30.587+B30.114, wherein Y is referred to as ash
Angle value.So, each pixel of coloured image uses unified palette by index, saves substantial amounts of memory space and place
The reason time.
The image processing techniques being mainly concerned with in intelligent glasses recognition of face of the present invention is Image semantic classification.Pretreatment
Purpose is to improve picture quality, strengthening useful information.Conventional pretreatment has attitude rectification, illumination compensation, size to return
One change, denoising, border strengthen, improve contrast etc..The present invention use histogram equalization technology carry out the pre-place of facial image
Reason, main purpose is enhancing contrast ratio, improves picture quality.Theoretical foundation is as follows:
A) rectangular histogram is a kind of expression of image.To a width gray level image, its gray-scale statistical rectangular histogram reflects in this figure
The statistical conditions that different grey-scale occurs.It is defined as: p (sk)=nk/N.Wherein nk represents the kth level gray value of image, N table
Show sum of all pixels.Rectangular histogram can be given the tonal range of the whole description of this image, such as image, the frequency of each gray level and
Intensity profile, the average light and shade of entire image and contrast.
B) histogram equalization is also histogram, it is simply that the histogram distribution of given image is changed over and uniformly divides
The rectangular histogram of cloth, it is a kind of conventional grayscale enhancing method.
C) mean filter is also referred to as linear filtering, and its main method used is neighborhood averaging, the pixel that transfinites smoothing techniques
With weighted mean filter method.Neighborhood averaging is employed herein, and averaging method its ultimate principle in field is to substitute former by average
Each pixel value in image, i.e. to pending current pixel point (x, all pixels in y) selecting a template to seek template
Average, then this average give current pixel point (x, y), as image gray scale g at that point after processing (x, y).Template is transported
The basic step calculated is: b1) template is roamed in the picture, and template center is overlapped with certain location of pixels in image；
B2) pixel that coefficient in template is corresponding with under template is multiplied；B3) all products are added up；B4) will be assigned in image right
Answer the pixel of template center position.
D) typical method of nonlinear smoothing filtering is medium filtering.The ultimate principle of median filter is digital picture
Or the value of any in Serial No. Mesophyticum of each point value in a neighborhood of this point replaces, allow surrounding pixel gray value
The bigger pixel of difference changes to take the value close with the pixel value of surrounding, such that it is able to cancel isolated noise spot.
The present invention provide face identification method based on intelligent glasses, described intelligent glasses will collection smart mobile phone, GPS,
, it has function that is similar with Android but that cross, such as map etc. for spectacles for optimized to camera.Put on this
Intelligent glasses user can also use camera function to shoot face full face, realizes people by photo and information bank comparing
Face identification function, identifies the information of personnel fast and accurately and feeds back timely.Obtaining the same of personnel's essential information
Time, moreover it is possible to personnel positions positioned and stores location data, map showing the track that personnel positions changes, when first
Between grasp position and the trend of personnel.In compatibility, with arbitrary money, this intelligent glasses can support that the smart mobile phone of bluetooth synchronizes.
The face identification method based on intelligent glasses that the present invention provides, obtains the face figure of user by intelligent glasses
Picture, the algorithm of recycling core carries out computational analysis, and then and its data storehouse to face position, shape of face and the angle of its face
In existing model compare, after judge the true identity of user.Intelligent glasses face recognition technology is based on local feature
The single training image per person method in region.The first step, needs localized region to be defined；Second step, face partial zones
The extraction of characteristic of field, according to the transformation matrix that obtains after sample training by facial image DUAL PROBLEMS OF VECTOR MAPPING be face characteristic to
Amount；3rd step, local feature selects (optional)；A rear step is by classification.The form of grader many employings assembled classifier, often
The corresponding grader of individual local feature, can use the modes such as ballot or linear weighted function to obtain whole recognition result afterwards.Intelligent glasses people
The face identification integrated use multiple technologies such as digital picture/Video processing, pattern recognition, computer vision, core technology is people
Face recognizer.The algorithm of at present recognition of face has 4 kinds: recognizer based on human face characteristic point, based on view picture facial image
Recognizer, recognizer based on template, the algorithm that utilizes neutral net to be identified.
As the first step of recognition of face, the work that intelligent glasses Face datection is carried out be by face from image background
Detect, made Face datection become one owing to affecting by factors such as the head poses of image background, brightness flop and people
Complicated research contents.Detection and location: detection is to differentiate whether to there is face in piece image, location is then to provide face at image
In position.The facial image information obtained behind location is the pattern of measurement space, work to be identified, and first has to measure
Data in space are mapped in feature space.Using principal component method, principle is by a high dimension vector, by a spy
Different eigenvectors matrix, projects in the vector space of a low-dimensional, is characterized as a low dimensional vector, and only loses one
A little secondary information.Image dimension can be reduced by the facial image through detecting and positioned is carried out feature extraction operation
Number, identifies amount of calculation such that it is able to reduce, and improves the effect of accuracy of identification.Intelligent glasses face identification system uses feature based
The PCA (PCA) of face, constructs principal component subspace according to lineup's face training sample, during detection, is thrown by test image
Shadow in principal component space, obtains one group of projection coefficient, then with each known facial image model comparision, thus obtain detection knot
Really.
In view of in the case of improper shooting based on intelligent glasses, Side Face Image often can only be obtained.The present invention
During Side Face Image detects, utilize integral projection method that pretreated binaryzation Side Face Image is carried out face
The location in region.During side face characteristic extracts, the present invention utilizes integral projection method and face priori to obtain face
The essential information of each principal character in portion, and by these information with one group of geometric properties vector representation, represent side face with this.?
During the identification of side face, the sorting technique of minimum distance classifier and the method for RBF artificial neural network is used to carry out
Judging to identify, the sorting technique of minimum distance classifier utilizes range formula to measure the similarity degree of two width facial images, and selects
Take suitable threshold value to carry out judging to identify；Facial image feature, by the study to training sample, is passed through net by RBF neural
Hidden node data center and connection weights within network are remembered, and when running into the input with certain face feature similarity, network can
This face is identified voluntarily according to memory.
The present invention relates generally to minimum distance classifier technology in the face recognition process of intelligent glasses side and RBF is artificial
Nerual network technique.Minimum distance classification, refers to obtain unknown categorization vector to identify representation vector central point of all categories
Distance, belongs to a kind of image classification method apart from a minimum class by unknown categorization vector.Minimum distance classification is according to mould
The distance of formula and all kinds of representative samples carries out a kind of statistical recognition method of pattern classification.In this approach, identified pattern
Minimum with the distance of affiliated pattern class sample.Assuming that c classification represents characteristic vector R1 of pattern ..., Rc represents, x is
The characteristic vector of identified pattern, | x-Ri | be x Yu Ri (i=1,2 ..., c) between distance, if | x-Ri | is minimum, then
X is divided into the i-th class.Can with all kinds of representative sample set, and be not only with a sample conduct in more complicated cases
The basis of minimum distance classification.Carry out minimum distance classification first have to for each classification determine it representative pattern feature to
Amount, this is the key carrying out classifying quality quality in this way.A kind of distance metric determined of its secondary selection is to calculate quilt
Recognition mode and all kinds of distance represented between pattern character vector.RBF neural is a kind of Learning Algorithm, entirely
Claim artificial neural network based on error backpropagation algorithm.Single hidden layer feedforward network of topological structure, before commonly referred to as three layers
Feedback net or three layer perceptron, it may be assumed that input layer, intermediate layer (also referred to as hidden layer) and output layer.The structure of RBF neural is as follows: the
One layer is input layer, is made up of signal source node.The second layer is hidden layer, and the transforming function transformation function of hidden unit is a kind of local distribution
Non-negative nonlinear function, he is to central point radial symmetric and decay.The unit number of hidden layer by described problem it needs to be determined that.
Third layer is output layer, and the output of network is the linear weighted function of hidden unit output.The input space of RBF network is to hidden layer space
Conversion be nonlinear, and be linear from hidden layer space to the conversion in output layer space.
The principle of intelligent glasses face recognition algorithms: system input is usually one or a series of containing not determining one's identity
Facial image, and the facial image of the some known identities in face database or encode accordingly, its output is then
It is a series of similarity scores, shows the identity of face to be identified.
Wear with policeman intelligent glasses equipment on duty time, can carry out easily character face identify and find suspicious people
Thing, just as the thing that the contact lens of the leading man of " mission spy 4 " is accomplished.Meanwhile, the whole enforcing law of police also will
Recorded and supervised, it is achieved real law enforcement is transparent.Intelligent glasses can pretend naturally, can easily carry, and then outside
Information acquisition is carried out, it is also possible to being directed at specific face and search out related data, policeman holds to specification in the case of people is unwitting
Method has facilitation.
As a example by the search system application of public security photo: the difficult problem that public security system faces is cannot to make full use of at hand
The photo resources that ready-made (identity card, temporary residency card etc.) is millions of, taking a photo during investigating into a case but cannot have
Its identity of location of effect, artificial carries out the work that photo comparison has been practically impossible to one by one, can only spend substantial amounts of police
Power and time are investigated.Use face recognition algorithms to realize fast face retrieval, fully demonstrate the strong alert power of science and technology.
The identification step of intelligent glasses face is as follows:
(1) the image surface archives of face are initially set up.I.e. form image surface file with the certificate photograph such as identity card, temporary residency card, and
These image surface file generated faces stricture of vagina (Faceprint) coding is stored up.
(2) current human body image surface is obtained.I.e. by the image surface of the current a suspect of cameras capture, or it is defeated to take photo
Enter, and by current image surface file generated face stricture of vagina coding.
(3) with the comparison of current face stricture of vagina coding with archives stock.The face stricture of vagina coding of image surface that will be current and file store
Face stricture of vagina in depositing is coded into line retrieval comparison.Above-mentioned " face stricture of vagina coding " mode is the substitutive characteristics according to face face and beginning
Carry out work.This stricture of vagina coding can resist the change of light, skin color, facial hair, hair style, glasses, expression and attitude
Change, there is powerful reliability, so that it can accurately recognize someone from million people.The identification process of face,
Utilize the common image processing equipment just can automatically, continuously, complete in real time.
Although the present invention discloses as above with preferred embodiment, so it is not limited to the present invention, any this area skill
Art personnel, without departing from the spirit and scope of the present invention, when making a little amendment and perfect, the therefore protection model of the present invention
Enclose when with being as the criterion that claims are defined.
Claims (8)
1. a face identification method based on intelligent glasses, it is characterised in that comprise the steps:
A) photo module, microprocessor and wireless communication module are set on intelligent glasses；
B) face-image of user, the microprocessor pair on recycling intelligent glasses is obtained by the photo module on intelligent glasses
The face-image obtained carries out recognition of face；
C) by wireless communication module, face recognition result is uploaded to central database compare, it is judged that user's is true
Identity, and return the relevant information of this user.
2. face identification method based on intelligent glasses as claimed in claim 1, it is characterised in that described photo module is outstanding
Putting a photographic head in intelligent glasses front, described microprocessor and wireless communication module are positioned at picture frame left and right sides, described intelligence
It is equipped with one on the eyeglass of energy glasses and wears the type display screen that declines.
3. face identification method based on intelligent glasses as claimed in claim 2, it is characterised in that on described intelligent glasses also
GPS module is set, while obtaining user related information, also this customer location is positioned, and miniature aobvious in wear-type
Show screen display electronic chart and the track of this customer location change.
4. face identification method based on intelligent glasses as claimed in claim 2, it is characterised in that described wireless communication module
Including WiFi module and/or bluetooth module, the mirror holder of described intelligent glasses is arranged switch switch, take pictures, image two-pass by
Key.
5. face identification method based on intelligent glasses as claimed in claim 1, it is characterised in that described step b) also includes
The face-image obtained is carried out following pretreatment:
Coloured image is first converted to gray level image, and each pixel of described coloured image uses unified toning by index
Plate；
Then use rectangular histogram that the gray level image after conversion is carried out equilibrium treatment, enhancing contrast ratio；
Then use field averaging method that the gray level image after conversion is carried out mean filter；
Median filtering method is finally used to remove the isolated noise point in gray level image.
6. face identification method based on intelligent glasses as claimed in claim 5, it is characterised in that described step b) uses base
One training sample in local characteristic region carries out recognition of face, comprises the steps:
B1) differentiating in piece image whether there is face, if there is face, defining regional area is face region；
B2) extract the local feature region of face region, and local feature is pressed face position, shape of face and the angle of face
Classify, it is thus achieved that face local feature region；
B3) the corresponding grader of each face local feature region, uses the assembled classifier face local feature region to extracting
Similarity identification is carried out with ballot or linear weighted function mode.
7. face identification method based on intelligent glasses as claimed in claim 6, it is characterised in that described face region
The extraction process of local feature region as follows: obtain the facial image information of measurement space；Use principal component method, by height
Dimensional vector is projected to by eigenvectors matrix in the vector space of low-dimensional；By the projection coefficient that obtains in low gt with
Known facial image model comparision, thus obtain testing result.
8. face identification method based on intelligent glasses as claimed in claim 7, it is characterised in that the face in described step b)
Portion's image is Side Face Image, described step b1) utilize integral projection method to pretreated binaryzation Side Face Image
Carry out the location of human face region；
During side face characteristic extracts, integral projection method and face priori is utilized to obtain each principal character of face
Essential information, and by these information with one group of geometric properties vector representation, represent side face with this；
During the identification of side face, use sorting technique and the side of RBF artificial neural network of minimum distance classifier
Method carries out judging to identify.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610586532.7A CN106203391A (en) | 2016-07-25 | 2016-07-25 | Face identification method based on intelligent glasses |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610586532.7A CN106203391A (en) | 2016-07-25 | 2016-07-25 | Face identification method based on intelligent glasses |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106203391A true CN106203391A (en) | 2016-12-07 |
Family
ID=57491752
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610586532.7A Pending CN106203391A (en) | 2016-07-25 | 2016-07-25 | Face identification method based on intelligent glasses |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106203391A (en) |
Cited By (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106650359A (en) * | 2016-12-30 | 2017-05-10 | 中广热点云科技有限公司 | System and method for collecting object information and matching information |
| CN107194817A (en) * | 2017-03-29 | 2017-09-22 | 腾讯科技（深圳）有限公司 | Methods of exhibiting, device and the computer equipment of user social contact information |
| CN107463910A (en) * | 2017-08-09 | 2017-12-12 | 中国电子科技集团公司第二十八研究所 | Target identification augmented reality system based on GoogleGlass |
| CN107918771A (en) * | 2017-12-07 | 2018-04-17 | 河北工业大学 | Character recognition method and Worn type person recognition system |
| CN108038468A (en) * | 2017-12-26 | 2018-05-15 | 北斗七星（重庆）物联网技术有限公司 | A kind of security terminal based on recognition of face |
| CN108566536A (en) * | 2018-05-15 | 2018-09-21 | 新疆习美人工智能科技有限公司 | Collection apparatus analysis system based on wearable device |
| CN109165597A (en) * | 2018-08-24 | 2019-01-08 | 福建铁工机智能机器人有限公司 | A kind of visiting method based on wisdom rural area AI system |
| CN109583462A (en) * | 2017-09-28 | 2019-04-05 | 幻视互动（北京）科技有限公司 | Data flow processing method, apparatus and system based on deep neural network |
| CN109905757A (en) * | 2019-02-13 | 2019-06-18 | 安徽励图信息科技股份有限公司 | The method that video caption broadcasts is controlled by recognition of face |
| CN110287780A (en) * | 2019-05-17 | 2019-09-27 | 长安大学 | A kind of illumination human face image characteristic extracting method |
| CN111444886A (en) * | 2020-04-27 | 2020-07-24 | 无锡职业技术学院 | Intelligent face recognition system based on wearable equipment |
| CN111538401A (en) * | 2019-01-31 | 2020-08-14 | 创新服务股份有限公司 | Method for positioning human face and eyes by using microscope |
| CN111596764A (en) * | 2020-05-18 | 2020-08-28 | 周庆博 | AR glasses face recognition system and method based on android system |
| CN111652166A (en) * | 2020-06-08 | 2020-09-11 | 重庆科技学院 | Palm print and face recognition method based on cellular neural network different association memory model |
| CN112037373A (en) * | 2020-08-10 | 2020-12-04 | 国网上海市电力公司 | Five-prevention safety auxiliary device based on face recognition |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101763504A (en) * | 2009-12-30 | 2010-06-30 | 邱怀骏 | Human head identification method under complex scene |
| CN104880835A (en) * | 2015-05-13 | 2015-09-02 | 浙江吉利控股集团有限公司 | Intelligent glasses |
| CN105357421A (en) * | 2015-10-27 | 2016-02-24 | 南京泓众电子科技有限公司 | Smart glasses system |
| CN105450929A (en) * | 2015-12-09 | 2016-03-30 | 安徽海聚信息科技有限责任公司 | Information identification and communication method for smart wearable devices |
-
2016
- 2016-07-25 CN CN201610586532.7A patent/CN106203391A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101763504A (en) * | 2009-12-30 | 2010-06-30 | 邱怀骏 | Human head identification method under complex scene |
| CN104880835A (en) * | 2015-05-13 | 2015-09-02 | 浙江吉利控股集团有限公司 | Intelligent glasses |
| CN105357421A (en) * | 2015-10-27 | 2016-02-24 | 南京泓众电子科技有限公司 | Smart glasses system |
| CN105450929A (en) * | 2015-12-09 | 2016-03-30 | 安徽海聚信息科技有限责任公司 | Information identification and communication method for smart wearable devices |
Cited By (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106650359A (en) * | 2016-12-30 | 2017-05-10 | 中广热点云科技有限公司 | System and method for collecting object information and matching information |
| CN107194817A (en) * | 2017-03-29 | 2017-09-22 | 腾讯科技（深圳）有限公司 | Methods of exhibiting, device and the computer equipment of user social contact information |
| CN107463910A (en) * | 2017-08-09 | 2017-12-12 | 中国电子科技集团公司第二十八研究所 | Target identification augmented reality system based on GoogleGlass |
| CN109583462A (en) * | 2017-09-28 | 2019-04-05 | 幻视互动（北京）科技有限公司 | Data flow processing method, apparatus and system based on deep neural network |
| CN107918771A (en) * | 2017-12-07 | 2018-04-17 | 河北工业大学 | Character recognition method and Worn type person recognition system |
| CN107918771B (en) * | 2017-12-07 | 2023-11-24 | 河北工业大学 | Person identification method and wearable person identification system |
| CN108038468A (en) * | 2017-12-26 | 2018-05-15 | 北斗七星（重庆）物联网技术有限公司 | A kind of security terminal based on recognition of face |
| CN108566536A (en) * | 2018-05-15 | 2018-09-21 | 新疆习美人工智能科技有限公司 | Collection apparatus analysis system based on wearable device |
| CN108566536B (en) * | 2018-05-15 | 2021-02-02 | 新疆习美人工智能科技有限公司 | Feature acquisition and analysis system based on wearable equipment |
| CN109165597A (en) * | 2018-08-24 | 2019-01-08 | 福建铁工机智能机器人有限公司 | A kind of visiting method based on wisdom rural area AI system |
| CN111538401A (en) * | 2019-01-31 | 2020-08-14 | 创新服务股份有限公司 | Method for positioning human face and eyes by using microscope |
| CN109905757A (en) * | 2019-02-13 | 2019-06-18 | 安徽励图信息科技股份有限公司 | The method that video caption broadcasts is controlled by recognition of face |
| CN110287780B (en) * | 2019-05-17 | 2021-08-03 | 长安大学 | Method for extracting facial image features under illumination |
| CN110287780A (en) * | 2019-05-17 | 2019-09-27 | 长安大学 | A kind of illumination human face image characteristic extracting method |
| CN111444886A (en) * | 2020-04-27 | 2020-07-24 | 无锡职业技术学院 | Intelligent face recognition system based on wearable equipment |
| CN111596764A (en) * | 2020-05-18 | 2020-08-28 | 周庆博 | AR glasses face recognition system and method based on android system |
| CN111652166A (en) * | 2020-06-08 | 2020-09-11 | 重庆科技学院 | Palm print and face recognition method based on cellular neural network different association memory model |
| CN111652166B (en) * | 2020-06-08 | 2022-08-30 | 重庆科技学院 | Palm print and face recognition method based on cellular neural network different association memory model |
| CN112037373A (en) * | 2020-08-10 | 2020-12-04 | 国网上海市电力公司 | Five-prevention safety auxiliary device based on face recognition |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106203391A (en) | Face identification method based on intelligent glasses | |
| CN101558431B (en) | Face authentication device | |
| CN110889312B (en) | Living body detection method and apparatus, electronic device, computer-readable storage medium | |
| Kashem et al. | Face recognition system based on principal component analysis (PCA) with back propagation neural networks (BPNN) | |
| KR100944247B1 (en) | System and method for face recognition | |
| CN110008813B (en) | Face recognition method and system based on living body detection technology | |
| CN107438854A (en) | The system and method that the image captured using mobile device performs the user authentication based on fingerprint | |
| US8515127B2 (en) | Multispectral detection of personal attributes for video surveillance | |
| KR20130037734A (en) | A system for real-time recognizing a face using radial basis function neural network algorithms | |
| de Souza et al. | On the learning of deep local features for robust face spoofing detection | |
| Choi et al. | Shadow compensation in 2D images for face recognition | |
| CN112215180A (en) | Living body detection method and device | |
| Wasnik et al. | Presentation attack detection for smartphone based fingerphoto recognition using second order local structures | |
| Galiyawala et al. | Person retrieval in surveillance using textual query: a review | |
| Pauca et al. | Challenging ocular image recognition | |
| Zou et al. | Application of facial symmetrical characteristic to transfer learning | |
| Peng et al. | Presentation attack detection based on two-stream vision transformers with self-attention fusion | |
| Ravi et al. | A study on face recognition technique based on Eigenface | |
| CN109145855A (en) | A kind of method for detecting human face and device | |
| Li et al. | Intelligent terminal face spoofing detection algorithm based on deep belief network | |
| CN107368811B (en) | LBP-based face feature extraction method under infrared and non-infrared illumination | |
| CN109635860B (en) | Image classification method and system | |
| CN117351579B (en) | Iris living body detection method and device based on multi-source information fusion | |
| Yu et al. | Effect of Facial Shape Information Reflected on Learned Features in Face Spoofing Detection | |
| Mothwa | Smart attendance monitoring system using computer vision. |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20161207 |