CN107386909A - A kind of antitheft door with warning function - Google Patents
A kind of antitheft door with warning function Download PDFInfo
- Publication number
- CN107386909A CN107386909A CN201710558373.4A CN201710558373A CN107386909A CN 107386909 A CN107386909 A CN 107386909A CN 201710558373 A CN201710558373 A CN 201710558373A CN 107386909 A CN107386909 A CN 107386909A
- Authority
- CN
- China
- Prior art keywords
- module
- image
- personnel
- door body
- warning function
- 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
-
- E—FIXED CONSTRUCTIONS
- E06—DOORS, WINDOWS, SHUTTERS, OR ROLLER BLINDS IN GENERAL; LADDERS
- E06B—FIXED OR MOVABLE CLOSURES FOR OPENINGS IN BUILDINGS, VEHICLES, FENCES OR LIKE ENCLOSURES IN GENERAL, e.g. DOORS, WINDOWS, BLINDS, GATES
- E06B5/00—Doors, windows, or like closures for special purposes; Border constructions therefor
- E06B5/10—Doors, windows, or like closures for special purposes; Border constructions therefor for protection against air-raid or other war-like action; for other protective purposes
- E06B5/11—Doors, windows, or like closures for special purposes; Border constructions therefor for protection against air-raid or other war-like action; for other protective purposes against burglary
-
- E—FIXED CONSTRUCTIONS
- E06—DOORS, WINDOWS, SHUTTERS, OR ROLLER BLINDS IN GENERAL; LADDERS
- E06B—FIXED OR MOVABLE CLOSURES FOR OPENINGS IN BUILDINGS, VEHICLES, FENCES OR LIKE ENCLOSURES IN GENERAL, e.g. DOORS, WINDOWS, BLINDS, GATES
- E06B7/00—Special arrangements or measures in connection with doors or windows
- E06B7/28—Other arrangements on doors or windows, e.g. door-plates, windows adapted to carry plants, hooks for window cleaners
-
- G—PHYSICS
- G07—CHECKING-DEVICES
- G07C—TIME OR ATTENDANCE REGISTERS; REGISTERING OR INDICATING THE WORKING OF MACHINES; GENERATING RANDOM NUMBERS; VOTING OR LOTTERY APPARATUS; ARRANGEMENTS, SYSTEMS OR APPARATUS FOR CHECKING NOT PROVIDED FOR ELSEWHERE
- G07C9/00—Individual registration on entry or exit
- G07C9/30—Individual registration on entry or exit not involving the use of a pass
- G07C9/32—Individual registration on entry or exit not involving the use of a pass in combination with an identity check
Abstract
The invention provides a kind of antitheft door with warning function, holdee can be identified and be alarmed, the identifying system and early warning system being connected including door body, with door body, the identifying system include acquisition module, pretreatment module, characteristic extracting module, again training module, identification module and evaluation module.The present invention realizes the warning function of antitheft door.
Description
Technical field
The present invention relates to antitheft door field, and in particular to a kind of antitheft door with warning function.
Background technology
The antitheft door that we make, is generally used for antitheft, does not have warning function.
The content of the invention
A kind of in view of the above-mentioned problems, the present invention is intended to provide antitheft door with warning function.
The purpose of the present invention is realized using following technical scheme:
A kind of antitheft door with warning function is provided, holdee can be identified and be alarmed, including door body,
The identifying system and early warning system being connected with door body, the identifying system include acquisition module, pretreatment module, feature extraction mould
Block, again training module, identification module and evaluation module;The acquisition module includes two towards different cameras, is used for
Video image acquisition, the pretreatment module are used to determine the personnel positions in pedestrian image, obtain the rectangle region for including personnel
Domain;The characteristic extracting module, for carrying out external appearance characteristic extraction in the rectangular area comprising personnel;The training module is used
In training multiple cross-module state projection models, two projection functions are included in each cross-module state projection model, they respectively will not
Hold to levy with the image in video camera and be mapped in common feature space and complete Similarity Measure;The identification module again, use
Whether contain the pedestrian image consistent with personnel query in identification database and confirm personnel query identity;The evaluation module
For assessing systematic function.
Beneficial effects of the present invention are:Realize the warning function of antitheft door.
Brief description of the drawings
Using accompanying drawing, the invention will be further described, but the embodiment in accompanying drawing does not form any limit to the present invention
System, for one of ordinary skill in the art, on the premise of not paying creative work, can also be obtained according to the following drawings
Other accompanying drawings.
Fig. 1 is the structure connection diagram of the present invention.
Reference:
Door body 1, identifying system 2, early warning system 3.
Embodiment
The invention will be further described with the following Examples.
Referring to Fig. 1, a kind of antitheft door with warning function of the present embodiment, holdee can be identified and be reported
It is alert, including door body 1, the identifying system 2 and early warning system 3 that are connected with door body 1, the identifying system 2 include acquisition module, locate in advance
Manage module, characteristic extracting module, again training module, identification module, evaluation module;The acquisition module includes two directions not
Same camera, for video image acquisition, the pretreatment module is used to determine the personnel positions in pedestrian image, obtains bag
Rectangular area containing personnel;The characteristic extracting module, for carrying out external appearance characteristic extraction in the rectangular area comprising personnel;
The training module is used to train multiple cross-module state projection models, and two projection letters are included in each cross-module state projection model
Number, the image in different cameras is held to levy respectively and is mapped in common feature space and completes Similarity Measure by they;Institute
Identification module again is stated, for whether containing the pedestrian image consistent with personnel query in identification database and confirming personnel query body
Part;The evaluation module is used to assess systematic function.
Preferably, door body 1 has positive and negative two top layer, and internal structure is provided with metallic support.
Originally select embodiment antitheft door more light.
Preferably, between positive and negative two top layer of door body 1, and at each space of internal structure metallic support, filling is full
There are foam or asbestos, and Men Bianchu is provided with angle bar.
Originally have and select embodiment to realize the heat-insulated of antitheft door.
Preferably, the pretreatment module includes image fusion unit, and described image integrated unit is used for separate sources
Pedestrian image carry out fusion treatment, preferably to obtain the comprehensive character of personnel, including:
(1) the two width source images merged to needs carry out wavelet decomposition with bi-orthogonal wavelet transformation respectively, it is determined that after decomposing
The wavelet coefficient of image;
(2) wavelet coefficient of image after decomposing is chosen in the ratio of setting to low frequency coefficient, forms the small echo of fused images
Low frequency coefficient matrix;
(3) local edge to high frequency coefficient using texture homogeneity measure analysis specific region difference low-and high-frequency coefficient,
The texture homogeneity measure of image-region is calculated, and the high-frequency wavelet coefficient matrix of fused images is determined according to predetermined rule,
The calculation formula of the texture homogeneity measure in described image region is defined as:
In formula, GF (x) represents image-region x texture homogeneity measure, GFlRepresent image-region x each high fdrequency component
The texture homogeneity measure of image in the horizontal direction, GFcRepresent image-region x each high fdrequency component image in vertical direction
Texture homogeneity measure, GFdRepresent that the texture homogeneity of image-region x each high fdrequency component image in the diagonal directions is surveyed
Degree;
(4) the high-frequency wavelet coefficient matrix of the wavelet low frequency coefficient matrix of the fused images, the fused images is entered
The discrete biorthogonal wavelet inverse transformation of row, finally obtains fused images.
This preferred embodiment sets image fusion unit, and the puppet of image can be preferably told according to texture homogeneity measure
Edge, make detailed information more abundant and true while overall visual effect is ensured;Define the texture one of image-region
The calculation formula that cause property is estimated, accelerate the speed of image co-registration.
Preferably, the predetermined rule includes:If there is more than 78% pixel value that there is larger texture in image-region
Homogeneity measure, it is marginal zone to define the image-region, chooses the maximum high frequency imaging of corresponding Edge texture homogeneity measure
Wavelet coefficient forms the high-frequency wavelet coefficient matrix of the fused images;If have in image-region more than 78% pixel value have compared with
Small texture homogeneity measure, it is smooth area to define the image-region, calculates energy of the two width source images in the image-region respectively
Amount and matching degree, according to energy and matching degree determine the wavelet coefficients of two width source images in fused images wavelet coefficient it is shared
Proportion, the high-frequency wavelet coefficient matrix of the fused images is determined according to following formula:
RG=βARA+(1-βA)RB
In formula, RGRepresent the high-frequency wavelet coefficient matrix of fused images, RA、βAThe wavelet systems of a secondary source images are represented respectively
Number, the wavelet coefficient proportion shared in fused images wavelet coefficient, RB、1-βAThe small echo of another secondary source images is represented respectively
Coefficient, the wavelet coefficient proportion shared in fused images wavelet coefficient.
This preferred embodiment determines the high-frequency wavelet coefficient matrix of fused images according to predetermined rule, improves fusion
Effect and the speed of fusion.
Preferably, the progress external appearance characteristic extraction in the rectangular area comprising personnel, including:Carry out the illumination of image
Normalized, it is specially:Image is set first as I, image I is transformed into log-domain using LOG logarithms, is filtered using difference Gauss
Ripple device is smoothed to image I, then carries out global contrast equalization processing to image I;Carry out picture size normalizing
Change is handled;Image block is carried out, for each image block, carries out characteristic vector pickup;The characteristic vector of all image blocks is entered
Row series connection, then carries out PCA Feature Dimension Reductions to the image after series connection.
This preferred embodiment sets characteristic extracting module, and first unitary of illumination processing is carried out to image before feature is extracted,
Reduce because of illumination variation and caused by scalloping, make the extraction of feature more accurate.
Preferably, the training module includes sample classification unit and cross-module state projection model unit;The sample
Taxon specifically performs:
If two video camera C1And C2Corresponding feature space is respectivelyWithd1And d2Two are represented respectively
The dimension in individual camera feature space, it is assumed that training dataset is combined into K to across camera review feature
sk=s (xk, yk) ∈ { -1 ,+1 } represent sample pair class label, -1 represent foreign peoples ,+1 represent it is similar, according to class label will
Training set is divided into negative sample setWith positive sample set|F1|+|F2|=K;
The cross-module state projection model unit specifically performs:
If cross-module state projection model set H=[h1h2..., hL], L submodel is used to handle L kind data differences, each
Individual submodel is made up of a pair of projection functions, hl=[pXl(x), pYl(y) footnote l, projection function p], are omittedXAnd p (x)Y(y) will
X ∈ X and y ∈ Y projections are to common feature space:
In formula,Projection vector is represented, a, b ∈ R are deviation from linearity, pXAnd p (x)Y(y) by original spy
Sign is projected in { -1 ,+1 } space;
Projection function q be present simultaneouslyXAnd q (x)Y(y) by x ∈ X and y ∈ Y projections to another common feature space:
The relation established between data category and common trait space, objective function:
In formula, E represents it is expected,Represent the importance balance index of similar sample pair and foreign peoples's sample pair;
In formula, wkRepresent sample to { xk, ykThis submodel study in sample weights,
sk=s (xk, yk) ∈ { -1 ,+1 } represent sample pair class label,
By minimizing object function come learning parameter { u, v, a, b }, corresponding projection function is obtained.
This preferred embodiment uses multiple cross-module state projection models, can fully tackle a variety of data distribution differences.
Preferably, the pedestrian image consistent with personnel query whether is contained in the identification database and confirms personnel query
Identity, including:
Assuming that being queried personnel's collection is combined into { fi, sta (fi), i=1,2 ..., N, fiRepresent to be queried personnel, sta i-th
(fi) i-th of identity for being queried personnel is represented, for personnel query set { gj, sta (gj), j=1,2 ..., M:
sta(gj)=sta (f)
gjAnd fiSimilarity A (gj, fi) be expressed as:
A(gj, fi)=sign (uTgj+a)·sign(vTfi+b)+||(uTgj+a)-(vTfi+b)||
Wealthy value T, T ∈ [1,2] are set, if A (gj, fi) < T, then it is queried in personnel in the absence of consistent with personnel query
Image;
If A (gj, fi) >=T, personnel will be queried and sorted from big to small according to similarity, come foremost and inquirer
Member has identical identity.
This preferred embodiment improves the accuracy of identification and efficiency of antitheft door personnel.
Preferably, it is described that identifying system performance is assessed, define evaluation function:
In formula, N represents inquiry times, ZnThe number of correct result can be found before expression in n positions, evaluation function value is got over
Greatly, then the recognition performance again of system is better, and tracking performance is stronger.
This preferred embodiment sets evaluation module, is advantageous to antitheft door to being improved.
One group of antitheft door recognition result of the present invention is as shown in the table:
N | The personal identification average used time | Alarm accuracy rate |
7 | 0.13s | 94.5% |
14 | 0.11s | 94.3% |
21 | 0.15s | 94.8% |
Finally it should be noted that the above embodiments are merely illustrative of the technical solutions of the present invention, rather than the present invention is protected
The limitation of scope is protected, although being explained with reference to preferred embodiment to the present invention, one of ordinary skill in the art should
Work as understanding, technical scheme can be modified or equivalent substitution, without departing from the reality of technical solution of the present invention
Matter and scope.
Claims (3)
1. a kind of antitheft door with warning function, it is characterized in that, holdee can be identified and be alarmed, including door
Body, the identifying system and early warning system being connected with door body, the identifying system include acquisition module, pretreatment module, feature and carried
Modulus block, again training module, identification module and evaluation module;The acquisition module includes two towards different cameras,
For video image acquisition, the pretreatment module is used to determine the personnel positions in pedestrian image, obtains the square for including personnel
Shape region;The characteristic extracting module, for carrying out external appearance characteristic extraction in the rectangular area comprising personnel;The training mould
Block is used to train multiple cross-module state projection models, and two projection functions are included in each cross-module state projection model, and they distinguish
Image in different cameras is held to levy and is mapped in common feature space and completes Similarity Measure;It is described to identify mould again
Block, for whether containing the pedestrian image consistent with personnel query in identification database and confirming personnel query identity;Institute's commentary
Valency module is used to assess systematic function.
2. a kind of antitheft door with warning function according to claim 1, it is characterized in that, door body has positive and negative two top layer,
Internal structure is provided with metallic support.
3. a kind of antitheft door with warning function according to claim 2, it is characterized in that, positive and negative two top layer of door body it
Between, and at each space of internal structure metallic support, fill out and be filled with foam or asbestos, and Men Bianchu is provided with angle bar.
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN2016110633495 | 2016-11-24 | ||
CN201611063349 | 2016-11-24 |
Publications (1)
Publication Number | Publication Date |
---|---|
CN107386909A true CN107386909A (en) | 2017-11-24 |
Family
ID=60339023
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710558373.4A Pending CN107386909A (en) | 2016-11-24 | 2017-07-10 | A kind of antitheft door with warning function |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107386909A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107993331A (en) * | 2018-01-30 | 2018-05-04 | 深圳众厉电力科技有限公司 | A kind of safe automation access control system |
Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN205532105U (en) * | 2016-01-30 | 2016-08-31 | 李刚 | Intelligent burglary -resisting door of multi -functional phonotape and videotape |
-
2017
- 2017-07-10 CN CN201710558373.4A patent/CN107386909A/en active Pending
Patent Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN205532105U (en) * | 2016-01-30 | 2016-08-31 | 李刚 | Intelligent burglary -resisting door of multi -functional phonotape and videotape |
Non-Patent Citations (2)
Title |
---|
刘凯: ""无交叠多摄像机网络中的人员再辨识"", 《中国博士学位论文全文数据库-信息科技辑》 * |
杨小义: ""图像特征识别算法及其在聋人视觉识别中的应用研究"", 《中国博士学位论文全文数据库-信息科技辑》 * |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107993331A (en) * | 2018-01-30 | 2018-05-04 | 深圳众厉电力科技有限公司 | A kind of safe automation access control system |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US10198623B2 (en) | Three-dimensional facial recognition method and system | |
US10872269B2 (en) | Methods and systems for the fast estimation of three-dimensional bounding boxes and drivable surfaces using LIDAR point clouds | |
CN106446150B (en) | A kind of method and device of vehicle precise search | |
CN104599275B (en) | The RGB-D scene understanding methods of imparametrization based on probability graph model | |
Chen et al. | Feature detection and description for image matching: from hand-crafted design to deep learning | |
CN105447441B (en) | Face authentication method and device | |
CN108875600A (en) | A kind of information of vehicles detection and tracking method, apparatus and computer storage medium based on YOLO | |
WO2021082635A1 (en) | Region of interest detection method and apparatus, readable storage medium and terminal device | |
CN110532970B (en) | Age and gender attribute analysis method, system, equipment and medium for 2D images of human faces | |
CN105809651B (en) | Image significance detection method based on the comparison of edge non-similarity | |
CN102024152B (en) | Method for recognizing traffic sings based on sparse expression and dictionary study | |
WO2021072886A1 (en) | Method and apparatus for image style transfer, device and storage medium | |
CN101770578B (en) | Image characteristic extraction method | |
CN110232387B (en) | Different-source image matching method based on KAZE-HOG algorithm | |
CN111104867A (en) | Recognition model training and vehicle heavy recognition method and device based on component segmentation | |
CN104615998B (en) | A kind of vehicle retrieval method based on various visual angles | |
CN106649747A (en) | Scenic spot identification method and system | |
Taverriti et al. | Real-time wearable computer vision system for improved museum experience | |
CN104463962B (en) | Three-dimensional scene reconstruction method based on GPS information video | |
CN106886754B (en) | Object identification method and system under a kind of three-dimensional scenic based on tri patch | |
CN106203446B (en) | Three dimensional object recognition positioning method for augmented reality auxiliary maintaining system | |
Cao et al. | A two-stage level set evolution scheme for man-made objects detection in aerial images | |
CN107386909A (en) | A kind of antitheft door with warning function | |
CN106557756A (en) | A kind of personnel identifying system again | |
CN109740405B (en) | Method for detecting front window difference information of non-aligned similar vehicles |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20171124 |