CN103268477A - Three-dimensional face recognition system based on embedded platform - Google Patents
Three-dimensional face recognition system based on embedded platform Download PDFInfo
- Publication number
- CN103268477A CN103268477A CN2013101774217A CN201310177421A CN103268477A CN 103268477 A CN103268477 A CN 103268477A CN 2013101774217 A CN2013101774217 A CN 2013101774217A CN 201310177421 A CN201310177421 A CN 201310177421A CN 103268477 A CN103268477 A CN 103268477A
- Authority
- CN
- China
- Prior art keywords
- layer module
- face recognition
- recognition system
- dimensional face
- algorithm
- 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
Landscapes
- Image Analysis (AREA)
Abstract
The invention provides a three-dimensional face recognition system based on an embedded platform. The three-dimensional face recognition system is characterized in that an application layer module, an algorithm layer module and a middle communication layer are included, the application layer module is used for driving and managing peripheral equipment, acquiring face data information collected by the peripheral equipment and preprocessing the data information, the algorithm layer module is used for concrete realization of algorithm modules such as face detecting, feature point locating, feature extracting and feature comparing which are used in the face data information, compared results are returned to the application layer module, the middle communication layer is used for completing data and parameter transmission between the application layer module and the algorithm layer module. According to the three-dimensional face recognition system based on the embedded platform, an application layer is achieved on an ARM core, an algorithm layer is achieved on a DSP core, an ARM+DSP double-core framework processing platform is used for achieving three-dimensional face recognition, so that the processing advantage of the ARM core and the processing advantage of the DSP core are achieved, and operation efficiency of the whole system is improved.
Description
Technical field
The present invention relates to a kind of three-dimensional face recognition system based on embedded platform, belong to three-dimensional face recognition technology field.
Background technology
In present three-dimensional face identification market, most three-dimensional face recognition systems all is based on the PC platform and realizes, gate control system as company and enterprise, three-dimensional face identification identity authorization system in the computer etc., well-known, because PC plateform system less stable, cost of goods manifactured is higher, bulky, relatively harsher to the requirement of working environment, therefore limited the application of this system in a lot of fields.This year, some are also arranged based on research and the application of the automatic three-dimensional face recognition system on the embedded platform, generally all be to use the little process chip of monokaryon MPU, but these are based on the system of the embedded platform of monokaryon MPU chip, often just realized the part of functions of automatic three-dimensional face recognition system, using the little processing of embedded platform more widely at present mainly is ARM and dsp chip, when the algorithm of three-dimensional face identification after ARM or dsp chip are realized, the application layer program (graphic user interface etc.) of these systems need depend on PC toward contact and realize, thereby total system seems very too fat to move.
Processor based on ARM nuclear is carrying out human-computer interaction device's management, is supporting real time operating system aspects such as (RTOS) to have apparent in view advantage.The shortcoming of ARM is digital Signal Processing ability and speed limited, and in the middle of the three-dimensional face recognition system of a practicality, must relate to the great amount of images data and be stored in transmission, the processing of high complexity data and application, therefore, in the application in this field, it is particularly outstanding that the bottleneck of arm processor seems.And DSP nuclear carries out framework and the design of chip from the angle that adapts to the complicated digital signal processing, therefore has primary processing advantage, especially be fit to various multi-media signals, carry out such as complicated algorithm computings such as coding, decoding, conversions as sound, image etc.But at aspects such as member user interactions interface routines, it is unable to do what one wishes then to seem.Based on ARM nuclear and DSP nuclear relative merits separately,
The advantage of ARM nuclear and DSP nuclear is fully utilized, realize that the embedded processing platform of ARM+DSP double-core framework has important practical significance.
Summary of the invention
At above-mentioned defective of the prior art, the invention provides a kind of three-dimensional face recognition system.
A kind of three-dimensional face recognition system based on embedded platform is characterized in that: comprise
Application layer module is used for driving and management peripherals, obtains people's face data message that the peripherals collection is come in, and data message is carried out pre-service;
Algorithm layer module is used for the specific implementation of algoritic module of the detection of people's face, positioning feature point, feature extraction and the feature contrast etc. of people's face data message, and comparison result is returned to described application layer module;
Middle Communication Layer is used for finishing data between described application layer module and the described algorithm layer module and the transmission of parameter.
Preferably, described application layer module is deployed on the ARM platform, and described algorithm layer module is deployed on the DSP platform, and total system adopts the double-core framework.
Preferably, described middle Communication Layer is equipped with the sending and receiving end at ARM platform and DSP platform.
Preferably, the pre-service of described people's face data message comprises: extract the multistage Gaussian derivative filtering feature in the facial image, convert thereof into the single frames data, in the storer of Communication Layer, wait for that the algorithm layer calls in the middle of being stored in.
Preferably, described people's face data message comprises: the shape facility of eyes, nose and face and relative seat feature.
A kind of three-dimensional face recognition system based on embedded platform of the present invention, realize application layer at ARM nuclear, implementation algorithm layer on DSP nuclear, the facial image data that the peripheral collecting device of ARM nuclear management obtains, after pre-service, send into DSP nuclear and carry out people's face detection and Identification processing, and the result of algorithm computing is sent back to the peripheral display device of ARM nuclear, this ARM+DSP double-core framework processing platform that utilizes is realized the three-dimensional face recognition system, ARM nuclear and DSP nuclear processing advantage have separately been brought into play, improve the operational efficiency of total system, taken into account considering of efficient and practicality in the real system.
Description of drawings
The present invention is further detailed explanation below in conjunction with the drawings and specific embodiments.
Fig. 1 is the software systems hierarchical diagram of the embodiment of the invention.
Embodiment
In order to make those skilled in the art person understand the present invention program better, and above-mentioned purpose of the present invention, feature and advantage can be become apparent more, the present invention is further detailed explanation below in conjunction with embodiment and embodiment accompanying drawing.
Fig. 1 is the software systems hierarchical diagram of the embodiment of the invention, and the whole software system is divided into application layer module, algorithm layer module, middle Communication Layer.Total system adopts dual-core architecture.
Application layer module is deployed on the ARM platform, is used for driving and management peripherals, and the peripherals here comprises camera, display, mouse, keyboard.Take people face video information by camera, therefrom obtain shape facility and the relative seat feature of eyes, nose and face, and then extract multistage Gaussian derivative filtering feature, convert thereof into the single frames data, in the storer of Communication Layer, wait for that the algorithm layer calls in the middle of being stored in.
Algorithm layer module is deployed on the DSP nuclear, comprises 5 algoritic modules, the detection of people's face, positioning feature point, the standardization of people's face, feature extraction, feature contrast.As follows with regard to each module analysis:
The workflow of people's face detection algorithm is: at first obtain the single frames data that pre-service obtains from middle Communication Layer, and the image parameter that passes over of application layer, move the Adaboost algorithm then and seek people's face at image, return search result behind the algorithm end of run: if find people's face, then return position and the size of people's face, if do not find, then return null pointer.
Positioning feature point is chosen eyes as unique point, eyes positioned, and with this benchmark as people's face normalizing operations such as follow-up rotation, scalings.The course of work is: the positive negative sample of end user's eye pattern sheet carries out the training of human eye cascade classifier, after obtaining the human eye cascade classifier, in human face region, mark the candidate region of human eye, in the candidate region, carry out the detection of human eye, detect the rectangular window of human eye in the candidate region after, we are with the center of the detected eye areas coordinate position as eyes.
The standardization of people's face, the course of work: according to the positioning result of eyes unique point, facial image is done necessary rotation, scaling, then the facial image of unified size is done histogram equalization, image mask process.By histogram equalization, picture contrast is strengthened, sharpness improves, and in order further to eliminate the influence of disturbing, we eliminate the influence of ground unrest to the image mask film covering, finally obtain standardized facial image.
Feature extraction is carried out the LBP Feature Extraction to standardized people's face.
The feature contrast, after extracting feature, go into library module for the new person, be that the feature that will extract writes database with the name of waiting the personnel of putting in storage, for automatic identification module, be the feature that will extract with database in the identification of comparing of existing feature, and the most close personnel's of output name, the threshold values rule of foundation setting provides recognition result.
The comparison identification module result that will compare by middle Communication Layer returns to application layer, and application layer provides the result of identification according to the size of the current threshold values that sets, and shows at display.
Middle Communication Layer is used for finishing data between application module and the algorithm layer module and the transmission of parameter, is equipped with the sending and receiving end at AEM platform and DSP platform.
Above-described embodiment is preferred implementation of the present invention; but embodiments of the present invention are not limited by the examples; other any do not deviate from change, the modification done under spiritual essence of the present invention and the principle, substitutes, combination, simplify the substitute mode that all should be equivalence, is included within protection scope of the present invention.Therefore, protection scope of the present invention should be as the criterion with the protection domain that claim was defined.
Claims (5)
1. the three-dimensional face recognition system based on embedded platform is characterized in that: comprise
Application layer module is used for driving and management peripherals, obtains people's face data message that the peripherals collection is come in, and data message is carried out pre-service;
Algorithm layer module is used for the specific implementation of algoritic module of the detection of people's face, positioning feature point, feature extraction and the feature contrast etc. of people's face data message, and comparison result is returned to described application layer module;
Middle Communication Layer is used for finishing data between described application layer module and the described algorithm layer module and the transmission of parameter.
2. a kind of three-dimensional face recognition system based on embedded platform according to claim 1, it is characterized in that: described application layer module is deployed on the ARM platform, and described algorithm layer module is deployed on the DSP platform, and total system adopts the double-core framework.
3. a kind of three-dimensional face recognition system based on embedded platform according to claim 2 is characterized in that: Communication Layer is equipped with the sending and receiving end at ARM platform and DSP platform in the middle of described.
4. a kind of three-dimensional face recognition system based on embedded platform according to claim 1, it is characterized in that: the pre-service of described people's face data message comprises: extract the multistage Gaussian derivative filtering feature in the facial image, convert thereof into the single frames data, in the storer of Communication Layer, wait for that the algorithm layer calls in the middle of being stored in.
5. a kind of three-dimensional face recognition system based on embedded platform according to claim 4, it is characterized in that: described people's face data message comprises: the shape facility of eyes, nose and face and relative seat feature.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2013101774217A CN103268477A (en) | 2013-05-15 | 2013-05-15 | Three-dimensional face recognition system based on embedded platform |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2013101774217A CN103268477A (en) | 2013-05-15 | 2013-05-15 | Three-dimensional face recognition system based on embedded platform |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN103268477A true CN103268477A (en) | 2013-08-28 |
Family
ID=49012104
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2013101774217A Pending CN103268477A (en) | 2013-05-15 | 2013-05-15 | Three-dimensional face recognition system based on embedded platform |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103268477A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103985172A (en) * | 2014-05-14 | 2014-08-13 | 南京国安光电科技有限公司 | An access control system based on three-dimensional face identification |
| CN104536389A (en) * | 2014-11-27 | 2015-04-22 | 苏州福丰科技有限公司 | 3D face identification technology based intelligent household system and realization method thereof |
| CN108470167A (en) * | 2018-04-12 | 2018-08-31 | 苏州万店掌网络科技有限公司 | Face recognition chip, method and camera |
| CN109690542A (en) * | 2016-07-14 | 2019-04-26 | 金勇尚 | Use the user authen method and apparatus for the method for face recognition |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN201033077Y (en) * | 2006-11-20 | 2008-03-12 | 中山大学 | Built-in type human face detecting and recognizing device |
| CN101526997A (en) * | 2009-04-22 | 2009-09-09 | 无锡名鹰科技发展有限公司 | Embedded infrared face image identifying method and identifying device |
| CN102208015A (en) * | 2011-03-29 | 2011-10-05 | 上海洪剑智能科技有限公司 | Face identification system based on davinci platform and face identification method thereof |
| CN102509138A (en) * | 2011-11-18 | 2012-06-20 | 山东神思电子技术股份有限公司 | Authentication system based on second-generation ID card and human face feature recognition and working method therefor |
-
2013
- 2013-05-15 CN CN2013101774217A patent/CN103268477A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN201033077Y (en) * | 2006-11-20 | 2008-03-12 | 中山大学 | Built-in type human face detecting and recognizing device |
| CN101526997A (en) * | 2009-04-22 | 2009-09-09 | 无锡名鹰科技发展有限公司 | Embedded infrared face image identifying method and identifying device |
| CN102208015A (en) * | 2011-03-29 | 2011-10-05 | 上海洪剑智能科技有限公司 | Face identification system based on davinci platform and face identification method thereof |
| CN102509138A (en) * | 2011-11-18 | 2012-06-20 | 山东神思电子技术股份有限公司 | Authentication system based on second-generation ID card and human face feature recognition and working method therefor |
Non-Patent Citations (1)
| Title |
|---|
| 闫娟等: "人脸识别的技术研究与发展概况", 《电视技术》 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103985172A (en) * | 2014-05-14 | 2014-08-13 | 南京国安光电科技有限公司 | An access control system based on three-dimensional face identification |
| CN104536389A (en) * | 2014-11-27 | 2015-04-22 | 苏州福丰科技有限公司 | 3D face identification technology based intelligent household system and realization method thereof |
| CN109690542A (en) * | 2016-07-14 | 2019-04-26 | 金勇尚 | Use the user authen method and apparatus for the method for face recognition |
| CN108470167A (en) * | 2018-04-12 | 2018-08-31 | 苏州万店掌网络科技有限公司 | Face recognition chip, method and camera |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10509985B2 (en) | Method and apparatus for security inspection | |
| CN111723786B (en) | Method and device for detecting wearing of safety helmet based on single model prediction | |
| CN102375970B (en) | A kind of identity identifying method based on face and authenticate device | |
| EP3437019A1 (en) | Optical character recognition in structured documents | |
| CN110175528B (en) | Human body tracking method and device, computer equipment and readable medium | |
| CN105590097A (en) | Security system and method for recognizing face in real time with cooperation of double cameras on dark condition | |
| CN107610177B (en) | The method and apparatus of characteristic point is determined in a kind of synchronous superposition | |
| WO2011065952A1 (en) | Face recognition apparatus and methods | |
| CN110610127B (en) | Face recognition method and device, storage medium and electronic equipment | |
| CN104143086A (en) | Application technology of portrait comparison to mobile terminal operating system | |
| CN113490947A (en) | Detection model training method and device, detection model using method and storage medium | |
| CN103268477A (en) | Three-dimensional face recognition system based on embedded platform | |
| CN103996033A (en) | Human identification method based on tracking of human face five-sense-organ coordinates | |
| Aggarwal et al. | Online handwriting recognition using depth sensors | |
| Beg et al. | Text writing in the air | |
| CN105184236A (en) | Robot-based face identification system | |
| Wang et al. | Research on face recognition algorithm based on deep learning | |
| Baig et al. | Text writing in the air | |
| CN202694370U (en) | Multi-face recognition system based on digital image processing | |
| Chang et al. | Automatic Hand-Pose Trajectory Tracking System Using Video Sequences | |
| Ji et al. | Design of human machine interactive system based on hand gesture recognition | |
| CN201111085Y (en) | Human face automatic recognition system | |
| CN111414884A (en) | Facial expression recognition method based on edge calculation | |
| Alashbi et al. | Occluded face detection, face in niqab dataset | |
| Bhatt et al. | Text Extraction & Recognition from Visiting Cards |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20130828 |
|
| RJ01 | Rejection of invention patent application after publication |