CN208460085U - A kind of recognition of face lock based on distributed camera array - Google Patents
A kind of recognition of face lock based on distributed camera array Download PDFInfo
- Publication number
- CN208460085U CN208460085U CN201820849724.7U CN201820849724U CN208460085U CN 208460085 U CN208460085 U CN 208460085U CN 201820849724 U CN201820849724 U CN 201820849724U CN 208460085 U CN208460085 U CN 208460085U
- Authority
- CN
- China
- Prior art keywords
- recognition
- face
- lock
- camera array
- image
- 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.)
- Expired - Fee Related
Links
Landscapes
- Collating Specific Patterns (AREA)
- Lock And Its Accessories (AREA)
Abstract
The utility model describes a kind of recognition of face lock based on distributed camera array, including intelligent door lock and mobile terminal, intelligent door lock includes human body infrared induction module, face recognition module, wireless communication module, master controller and mechanical lock device, intelligent door lock is additionally provided with distributed camera array, the distribution camera array includes several cameras, and several cameras are along certain radian longitudinal arrangement.Due to using above-mentioned technology, the utility model installs multiple camera arrays by distributed in door lock, multiple camera array acquired images are passed through into high-precision stitching algorithm, obtain a secondary image comprising complete face, to carry out human face detection and recognition, so as to avoid existing human face recognition door lock for the limitation of different height users, and efficiently solve the problems, such as to cause recognition of face limited due to height, it has been finally reached effective expansion recognition of face range, met the purpose of wider height user demand.
Description
Technical field
The utility model relates to door lock technical field more particularly to a kind of recognitions of face based on distributed camera array
Lock.
Background technique
Face recognition technology is a kind of biological identification technology for carrying out identification based on facial feature information of people.With
Video camera or camera acquire image or video flowing containing face, and automatic detection and tracking face in the picture, and then right
The face that detects carries out a series of the relevant technologies of face, usually also referred to as Identification of Images, face recognition.
Currently, recognition of face has been widely used in numerous areas, such as in door lock technical field, usually in intelligence
One camera is set on the lock body of door lock, and the camera structure is as shown in Figure 1, to acquire the face of attacker by camera
Image, then judged by control units such as controllers, to complete the movement such as unlocking.
However, there is also certain deficiencies for existing recognition of face lock.For example, door lock is generally seated on door, height
Fixed during installation, for may cause for the user of different heights facial image capture difficult, recognition accuracy is low,
Classes are over is individually registered one's residence when going home with greater need for convenient and efficient scrubbing brush face for not fast enough the problem of recognition speed, especially children, and existing
Recognition of face lock be obviously unable to satisfy this demand, be badly in need of it is a kind of have a wide range of identification function recognition of face lock.
For how to realize large-scale identification, the prior art has also carried out certain improvement on original face door lock,
As Chinese patent (application number: 201520587480.6, the applying date: 2015-08-06) discloses a kind of recognition of face lock, the people
Face identification lock is by the regulating wheel that is mounted with to rotate upwardly and downwardly on CCD camera assembly, so as to be directed to the use of different heights
Family adjusts camera rotation to suitable angle.Although this kind of scheme has biggish identification range, but due to needing
It manually adjusts, so increasing the complexity of operation, is not well positioned to meet the actual demand of users.
Utility model content
In order to solve the problems such as above-mentioned identification range is small and complicated for operation, the utility model provides a kind of based on distribution
The recognition of face of camera array is locked, can be without realizing identification on a large scale under the premise of manually adjusting.
A kind of above-mentioned recognition of face lock based on distributed camera array, including intelligent door lock and mobile terminal, institute
Stating intelligent door lock includes human body infrared induction module, face recognition module, wireless communication module, master controller and mechanical lock dress
It sets, wherein the intelligent door lock is additionally provided with distributed camera array, and the distribution camera array is taken the photograph comprising several
As head, several described cameras are along certain radian longitudinal arrangement.
In above-mentioned apparatus, the distribution camera array includes the first camera set gradually from top to bottom, second
Camera and third camera.
In above-mentioned apparatus, the first camera in the distribution camera array is arranged in angle is looked up.
In above-mentioned apparatus, the second camera in the distribution camera array is arranged in level angle.
In above-mentioned apparatus, the third camera in the distribution camera array is arranged in depression angle.
In above-mentioned apparatus, the image transmitting that will take of distribution camera array to image mosaic module, and by
Image mosaic resume module obtains complete facial image.
In above-mentioned apparatus, described image splicing module includes image preprocessing, image registration and image co-registration.
In above-mentioned apparatus, the stitching algorithm of described image splicing module uses the calculation detected based on Harris corner feature
Method finds correctly registration angle point, and accurately and quickly complete image mosaic by using the algorithm of constraint registration pair.
In above-mentioned apparatus, the master controller judges whether according to the feedback of face recognition module with enabling permission, and
Control motor opens or closes door lock.
In above-mentioned apparatus, module carries out wireless communication the intelligent door lock with mobile terminal by wireless communication.
Advantages and beneficial effects of the utility model lie in that the utility model provides one kind based on distributed camera battle array
The recognition of face of column is locked, by the multiple camera arrays of installation distributed in door lock, multiple camera arrays are collected
Image obtains a secondary image comprising complete face by high-precision stitching algorithm, to carry out human face detection and recognition, thus
Existing human face recognition door lock is avoided for the limitation of different height users, and is efficiently solved since height causes face to be known
Not Shou Xian problem, be finally reached effective expansion recognition of face range, met the purpose of wider height user demand.
Detailed description of the invention
In order to illustrate the embodiment of the utility model or the technical proposal in the existing technology more clearly, below will be to embodiment
Or attached drawing needed to be used in the description of the prior art is briefly described, it should be apparent that, the accompanying drawings in the following description is only
It is some embodiments of the utility model, for those of ordinary skill in the art, before not making the creative labor property
It puts, is also possible to obtain other drawings based on these drawings.
Fig. 1 is the structural schematic diagram of existing camera；
Fig. 2 is the positive structure diagram of the recognition of face lock in the utility model based on distributed camera array；
Fig. 3 is the side structure schematic view of the recognition of face lock in the utility model based on distributed camera array；
Fig. 4 is the schematic perspective view of distributed camera array in the utility model；
Fig. 5 is the side structure schematic view of distributed camera array in the utility model；
Fig. 6 is the recognition effect figure of the recognition of face lock in the utility model based on distributed camera array.
Description of symbols:
1, intelligent door lock 2, distributed camera array 21, the first camera 22, second camera
23, third camera
Specific embodiment
With reference to the accompanying drawings and examples, specific embodiment of the present utility model is further described.Implement below
Example is only used for clearly illustrating the technical solution of the utility model, and cannot be used as a limitation the protection model of limitation the utility model
It encloses.
As Figure 2-3, the utility model describes a kind of recognition of face lock based on distributed camera array, mainly
Be made of intelligent door lock 1 and mobile terminal two parts, wherein intelligent door lock 1 in the utility model include for detecting whether
The close human body infrared induction module of someone, for the wireless communication module that intelligent door lock 1 and mobile terminal carry out wireless communication,
And face recognition module, master controller, mechanical lock device.Wherein:
Face recognition module is used for human face detection and tracing, its working principle is that: the characteristic point by extracting facial image,
And matched with the data stored in face database, the threshold value of a reflection similarity degree is obtained, is more than this threshold value, i.e.,
It is regarded as the same person, is identified successfully；It is on the contrary, then it is assumed that not have enabling permission, execute default is directed to stranger
Processing operation.
As for master controller then according to face recognition module feedback as a result, judge whether with enabling permission, if so,
Motor opening door lock is then controlled, is not otherwise opened the door, the corresponding operation for stranger is executed.
As illustrated in figures 4-5, the face in order to reach the technical effect for effectively expanding recognition of face range, in the utility model
Identification lock uses distributed camera array 2, which includes successively to set from top to bottom along certain radian
The first camera 21, second camera 22 and the third camera 23 set, and the first camera 21 is arranged in angle is looked up, the
Two cameras 22 are arranged in level angle, are then arranged in depression angle as third camera 23, so can be by three not
Camera with visual angle realizes larger range of recognition of face, recognition effect as shown in Figure 6, it can be achieved that away from ground 350mm~
The identification range of 2400mm, it is sufficient to meet the user demand of various heights.
Then, distributed camera array 2 is again by the image transmitting taken to image mosaic module, and by image mosaic
Resume module obtains complete facial image.Wherein, image mosaic module includes that image preprocessing, image registration and image melt
It closes, and its stitching algorithm uses the algorithm detected based on Harris corner feature, by using the algorithm of constraint registration pair, finds
Correctly registration angle point, can precisely and quickly complete image mosaic, obtain face camera and include the figure of complete face
Picture.
Wherein, Harris Corner Detection Algorithm is the basis of characteristic point detection, is proposed using vicinity points gray scale difference
It is worth concept, to carry out judging whether it is angle point, edge, smooth region.Harris Corner Detection principle is to utilize mobile window
Mouth calculates grey scale change value in the picture, and wherein critical workflow includes being converted into gray level image, calculating difference image, Gao Siping
Sliding, calculating local extremum, confirmation angle point, so that the features such as calculating is simple, stability is high and strong robustness is provided with, even if image
In the presence of the angle point that also can accurately find out image when rotation, grey scale change.The specific steps packet of its merging algorithm for images
It includes but is not limited to:
1. feature point extraction:
I.e. according to the multiple image of the collected different angle of camera array, reference picture and to be spliced is extracted respectively
The Harris angle point of image is as splicing feature point set.
2. Characteristic points match:
Corresponding registration point pair is found by correlative character registration, then two width figures are found to algorithm by constraining registration
The correct registration point as in.Specifically:
(1) associated window of a 2N+1*2N+1 size is taken centered on each characteristic point in two images, then with ginseng
The each angle point examined in image is reference point, and corresponding angle point is found in image to be spliced；
(2) realize that characteristics of image is registrated by calculating the related coefficient between characteristic point associated window；
(3) some characteristic point in reference picture and all characteristic points in image to be spliced carry out correlation calculations recklessly,
One group of related coefficient can be obtained, correlation is selected to be greater than the characteristic point of threshold value as candidate registration point pair.
3. transformation parameter estimation:
After finding registration point, using Perspective transformation model calculate image to be spliced to reference picture transformation parameter.
4. image co-registration:
Gradually go out method by using being fade-in, i.e., the second width is slowly transitioned by piece image in the overlapping region of stitching image
Image, to keep the consistency of stitching image visually.
The working principle that recognition of face is locked in the utility model are as follows:
Firstly, recognition of face lock when idle is in energy-efficient dormant state, when human body infrared induction module has detected
People close to when, triggering recognition of face lock starting work.
Secondly, distributed camera array 2 starts to acquire image, carried out by each camera of different location different perspectives
Man face image acquiring, then acquired image is sent to image mosaic module and carries out splicing, it is secondary comprising complete to obtain one
The image of face.
Then, extract the characteristic point of facial image by face recognition module, and with the data that are stored in face database
It is matched, obtains the threshold value of a reflection similarity degree, be more than such as this threshold value, that is, be regarded as the same person, be identified as
Function；On the contrary then recognition failures.
Finally, again by master controller then according to face recognition module feedback as a result, judging whether with enabling permission, such as
Fruit is the motor opening door lock for then controlling mechanical lock device, is not otherwise opened the door, and the corresponding operation for stranger is executed.
The above is only the preferred embodiment of the utility model only, is not intended to limit the utility model, all at this
Within the spirit and principle of utility model, any modification, equivalent replacement, improvement and so on should be included in the utility model
Protection scope within.
Claims (10)
1. a kind of recognition of face lock based on distributed camera array, including intelligent door lock and mobile terminal, the intelligent door
Lock includes human body infrared induction module, face recognition module, wireless communication module, master controller and mechanical lock device, spy
Sign is:
The intelligent door lock is additionally provided with distributed camera array, and the distribution camera array includes several camera shootings
Head, several described cameras are along certain radian longitudinal arrangement.
2. recognition of face according to claim 1 lock, which is characterized in that the distribution camera array include from up to
Under the first camera, second camera and the third camera that set gradually.
3. recognition of face lock according to claim 2, which is characterized in that first in the distribution camera array takes the photograph
As head is arranged in angle is looked up.
4. recognition of face lock according to claim 3, which is characterized in that second in the distribution camera array takes the photograph
As head is arranged in level angle.
5. recognition of face lock according to claim 4, which is characterized in that the third in the distribution camera array is taken the photograph
As head is arranged in depression angle.
6. recognition of face according to claim 1 lock, which is characterized in that the distribution camera array will take
Image transmitting obtains complete facial image by image mosaic resume module to image mosaic module.
7. recognition of face lock according to claim 6, which is characterized in that described image splicing module includes that image is located in advance
Reason, image registration and image co-registration.
8. recognition of face lock according to claim 7, which is characterized in that the stitching algorithm of described image splicing module uses
Based on the algorithm of Harris corner feature detection, by using the algorithm of constraint registration pair, correctly registration angle point, and essence are found
Really quickly complete image mosaic.
9. recognition of face lock according to claim 1, which is characterized in that the master controller is according to face recognition module
Feedback, judges whether with enabling permission, and control motor and open or close door lock.
10. recognition of face according to claim 1 lock, which is characterized in that intelligent door lock module by wireless communication
It is carried out wireless communication with mobile terminal.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201820849724.7U CN208460085U (en) | 2018-06-04 | 2018-06-04 | A kind of recognition of face lock based on distributed camera array |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201820849724.7U CN208460085U (en) | 2018-06-04 | 2018-06-04 | A kind of recognition of face lock based on distributed camera array |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN208460085U true CN208460085U (en) | 2019-02-01 |
Family
ID=65160170
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201820849724.7U Expired - Fee Related CN208460085U (en) | 2018-06-04 | 2018-06-04 | A kind of recognition of face lock based on distributed camera array |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN208460085U (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111050130A (en) * | 2019-12-12 | 2020-04-21 | 深圳市大拿科技有限公司 | Camera control method and device and storage medium |
-
2018
- 2018-06-04 CN CN201820849724.7U patent/CN208460085U/en not_active Expired - Fee Related
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111050130A (en) * | 2019-12-12 | 2020-04-21 | 深圳市大拿科技有限公司 | Camera control method and device and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104601964B (en) | Pedestrian target tracking and system in non-overlapping across the video camera room of the ken | |
| Mogelmose et al. | Tri-modal person re-identification with rgb, depth and thermal features | |
| Wu et al. | Face and facial feature extraction from color image | |
| US8787663B2 (en) | Tracking body parts by combined color image and depth processing | |
| Jordao et al. | Active face and feature tracking | |
| Kukharev et al. | Visitor identification-elaborating real time face recognition system | |
| CN101587485B (en) | Face information automatic login method based on face recognition technology | |
| CN103593598A (en) | User online authentication method and system based on living body detection and face recognition | |
| CN103605969A (en) | Method and device for face inputting | |
| CN107066969A (en) | A kind of face identification method | |
| CN102184016B (en) | Noncontact type mouse control method based on video sequence recognition | |
| CN102841354A (en) | Vision protection implementation method of electronic equipment with display screen | |
| CN109087411A (en) | A kind of recognition of face lock based on distributed camera array | |
| CN110569785B (en) | Face recognition method integrating tracking technology | |
| CN109086724A (en) | A kind of method for detecting human face and storage medium of acceleration | |
| CN105160331A (en) | Hidden Markov model based face geometrical feature identification method | |
| CN109726681A (en) | It is a kind of that location algorithm is identified based on the blind way of machine learning identification and image segmentation | |
| Harville | Stereo person tracking with short and long term plan-view appearance models of shape and color | |
| CN208460085U (en) | A kind of recognition of face lock based on distributed camera array | |
| Lee et al. | Hand gesture recognition using orientation histogram | |
| CN108446642A (en) | A kind of Distributive System of Face Recognition | |
| CN104573628A (en) | Three-dimensional face recognition method | |
| CN105844235A (en) | Visual saliency-based complex environment face detection method | |
| CN108509025A (en) | A kind of crane intelligent Lift-on/Lift-off System based on limb action identification | |
| Guan | Robust eye detection from facial image based on multi-cue facial information |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20190201 Termination date: 20190604 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |