CN105513043B - The coal mine underground operators method of counting of view-based access control model - Google Patents
The coal mine underground operators method of counting of view-based access control model Download PDFInfo
- Publication number
- CN105513043B CN105513043B CN201510801099.XA CN201510801099A CN105513043B CN 105513043 B CN105513043 B CN 105513043B CN 201510801099 A CN201510801099 A CN 201510801099A CN 105513043 B CN105513043 B CN 105513043B
- Authority
- CN
- China
- Prior art keywords
- oval
- image
- coal mine
- random process
- represent
- 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
- 238000000034 method Methods 0.000 title claims abstract description 28
- 239000003245 coal Substances 0.000 title claims abstract description 18
- 238000005457 optimization Methods 0.000 claims abstract description 5
- 239000000463 material Substances 0.000 claims abstract description 4
- 238000005381 potential energy Methods 0.000 claims description 5
- 230000003287 optical effect Effects 0.000 claims description 2
- 238000012544 monitoring process Methods 0.000 abstract description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000002922 simulated annealing Methods 0.000 description 2
- 241000464908 Elliptica Species 0.000 description 1
- 241000135309 Processus Species 0.000 description 1
- 238000001816 cooling Methods 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30196—Human being; Person
Landscapes
- Complex Calculations (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a kind of coal mine underground operators method of counting of view-based access control model, one piece of elliptic region in personnel targets image in image characterizes, picture material is expressed with stochastic model, oval appearance is then a random process in image, and the random process is described i.e. with Gibbs densityThe oval configuration for making Gibbs density maximum is found, that is, solves optimization problemStatisticsIn oval number, as total number of persons's mesh.This method can count automatically to coal mine operation number, be cooperated with one's own initiative without personnel, realize noiseless monitoring in real time.
Description
Technical field
The present invention relates to coal mine underground operators method of counting, more particularly to a kind of coal mine operation of view-based access control model
Personnel's method of counting.
Background technology
In order to prevent colliery spy, major accident occurs, and colliery forbids overdetermination person to produce, and monitors coal mine operation people in real time
Number is one of effective measures for preventing overdetermination person from producing.At present, coal mine operation number is to carry to identify by personnel in the pit
The more card phenomenons of people one of identification card system None- identified, there is skip often come what is counted in card, thus is unable to accurate counting coal mine
Lower operating personnel.
Need a kind of coal mine operation people for solving or at least improving one or more problems intrinsic in the prior art
Member's method of counting.
The content of the invention
It is an object of the invention to provide a kind of coal mine underground operators method of counting of view-based access control model, this method can be with
Coal mine operation number is counted automatically, cooperated with one's own initiative without personnel, realizes noiseless monitoring.
According to a kind of embodiment form, there is provided a kind of coal mine underground operators method of counting of view-based access control model, its feature
It is:One piece of elliptic region in personnel targets image in image characterizes, and defines elliptic space χ=[0, X in imageM]
× [0, YM]×[am, aM]×[bm, bM] × [0, π], wherein, XMAnd YMThe width and height of image, (a are represented respectivelym, aM) and
(bm, bM) minimum value and maximum of expression transverse and short axle, θ ∈ [0, π] represent oval direction respectively;With random mould
Type expresses picture material, and oval appearance is then a random process in image, and the random process is described i.e. with Gibbs densityWherein, U (o) represents the potential energy of oval random process, o={ o1=(x1, m1) ..., on=(xn,
mn) ∈ χ represent a kind of configuration of oval random process,For normaliztion constant;Searching makes Gibbs
The maximum oval configuration of density, that is, solve optimization problemStatisticsIn oval number, as total number of persons
Mesh.
Brief description of the drawings
By following explanation, accompanying drawing embodiment becomes aobvious and seen, it is only preferred with least one being described in conjunction with the accompanying
But the way of example of non-limiting example provides.
Fig. 1 is the method for the invention flow chart.
Fig. 2 is that the method for the invention personnel targets characterize schematic diagram.
Fig. 3 is the method for the invention personnel targets candidate region and background schematic diagram.
Embodiment
Fig. 1 is the flow chart of the inventive method, the flow of compares figure 1, is described.
Coal mine underground operators must safe wearing cap, safety cap is in subcircular, and usually yellow, with gray background
Contrast is obvious.When underground coal mine is installed and catches operation area scene camera, because underground coal mine is tunnel, camera optical axis
Generally there is the angle for being less than 90 more than 0 with ground, therefore, the safety cap that operating personnel is worn is in the picture then near oval
Shape, 2 points based on more than, the personnel targets in image are characterized with one piece of elliptic region in image.Personnel in the pit is dispersed in one
Determine operation in scope, they are different with the distance of video camera and visual angle, therefore safety cap position in the picture and form
Also it is different, therefore, space χ=K × M=[0, X belonging to definition ellipse in the pictureM] × [0, YM]×[am, aM]×[bm, bM]×
[0, π], K represent oval locational space, and M represents oval attribute space, XMAnd YMThe width and height of image, (a are represented respectivelym,
aM) and (bm, bM) minimum value and maximum of expression transverse and short axle, θ ∈ [0, π] represent oval direction, elliptical modes respectively
Type is as shown in Figure 2.Picture material is expressed with stochastic model, oval appearance is then a random process in image, close with Gibbs
Degree describes this random processU (o) represents the potential energy of oval random process, o={ o1=(x1,
m1) ..., on=(xn, mn) ∈ χ represent a kind of configuration of oval random process,For normalization
Constant;The oval configuration for making Gibbs density maximum is found, that is, solves optimization problemStatisticsIn ellipse
Number, as total number of persons's mesh.
Potential-energy function U (o)=Up(o)+Ud(o), Up(o) priori energy, U are representedd(o) data capacity is represented.
Mutually blocked sometimes in view of underground work area personnel, then show as ellipse in the picture and have overlapping, Up(o)
Punished, defined according to each oval Maximum overlap with neighbouring ellipseA(oi,
oj) ∈ [0,1] expression overlap coefficients,C is normal number, and μ () is ellipse area, μ
(oi∩oj) represent oval oiWith ojOverlapping area, γpThe weight to overlapping relation punishment is represented, its size is between ellipse
The size of overlapping area and change, overlapping area is big, then assigns greater weight, and corresponding priori energy is also bigger.
Data item Ud(o) confidence level that candidate's elliptic region in image is personnel is expressed, safety cap is in the picture
It is yellow ellipse, and background is grey, with both color characteristic difference sizes come to characterize candidate's elliptic region be personnel
Confidence level, i.e.,
oiRepresent candidate's elliptic region, F (oi) represent candidate's area elliptica
The neighborhood in domain, as shown in Figure 3.dB(oi, F (oi)) represent candidate's elliptic region distribution of color and the Pasteur of its neighborhood distribution of color
Distance:
Wherein, pj() and qj() represents candidate's elliptic region respectively, and each bin is corresponding with its neighborhood color histogram
Probability distribution, L represents color histogram bin number, normalized to [- 1,1] scope,
Wherein, d0Oval personnel targets region and the average of Pasteur's distance of its neighborhood distribution of color are represented, D is a yardstick
Parameter, od(dB) ∈ [- 1,1] have evaluated the similarity of candidate's elliptic region and its neighborhood distribution of color, for less Pasteur away from
From od(dB) it is on the occasion of expression candidate's elliptic region is that the confidence level of personnel is big, and negative value represents that candidate's elliptic region is personnel
Confidence level is low.
Solved using the simulated annealing (SA) based on birth and death processOptimization problem, algorithm include
Following steps:
E1. temperature parameter is initializedTime discretization step-length δ=δ0;
E2. raw step:For each pixel s ∈ I, if existed without target, increase by one at s with probability δ B (s)
Individual target, wherein,Z is
To a parameter;
E3. sequence step:Calculate configuration target oiData item ud(oi), the order successively decreased according to data capacity sorts;
E4. the step of going out:To each target o according to this orderiCalculate the death rateWherein
aβ(oi)=exp (- β U (oi)), then target is with probability d (oi) be destroyed;
E5. convergence test:If all targets are destroyed in step E4 just added by step E2, convergence, algorithm are realized
Terminate, otherwise, reduce temperature T (n+1)=kT (n) and discretization step-length δ (n+1)=δ (n)-Δ δ, be then return to step E2, its
Middle n represents cooling number, and k < 1 are constant, and Δ δ is constant.
Claims (1)
- A kind of 1. method of counting of the coal mine underground operators of view-based access control model, it is characterised in that:Underground coal mine is tunnel, shooting Machine optical axis generally has the angle for being less than 90 ° more than 0 ° with ground, and the safety cap that operating personnel is worn is in the picture then near ellipse Circle, one piece of elliptic region characterizes in the personnel targets image in image, defines elliptic space χ=[0, X in imageM]× [0, YM]×[am, aM]×[bm, bM] × [0, π], wherein, XMAnd YMThe width and height of image, a are represented respectivelym, aMTable respectively Show the minimum value and maximum of transverse, bm, bMThe minimum value and maximum of ellipse short shaft are represented respectively, and θ ∈ [0, π] are represented Oval direction;Picture material is expressed with stochastic model, oval appearance is then a random process in image, close with Gibbs Degree describes the random processWherein, U (o) represents the potential energy of oval random process, o={ o1 =(x1, m1) ..., oi=(xi, mi) ..., on=(xn, mn)) ∈ χ represent a kind of configuration of oval random process,For normaliztion constant;The oval configuration for making Gibbs density maximum is found, that is, solves optimization problemStatisticsIn oval number, as total number of persons's mesh;Potential-energy function U (o)=Up(o)+Ud(o), Up(o) Represent priori energy, Ud(o) data capacity is represented;Up(o) punished according to each oval with close oval Maximum overlap, DefinitionA(oi, oj) ∈ [0,1] expression overlap coefficients,C For normal number, μ () is ellipse area, μ (oi∩oj) represent oiAnd ojOverlapping area, γpRepresent to overlapping relation punishment Weight, its size change with the size of overlapping area between ellipse, and overlapping area is big, then assigns larger weight, accordingly Priori energy is also bigger.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510801099.XA CN105513043B (en) | 2015-11-20 | 2015-11-20 | The coal mine underground operators method of counting of view-based access control model |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510801099.XA CN105513043B (en) | 2015-11-20 | 2015-11-20 | The coal mine underground operators method of counting of view-based access control model |
Publications (2)
Publication Number | Publication Date |
---|---|
CN105513043A CN105513043A (en) | 2016-04-20 |
CN105513043B true CN105513043B (en) | 2018-04-03 |
Family
ID=55721000
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201510801099.XA Expired - Fee Related CN105513043B (en) | 2015-11-20 | 2015-11-20 | The coal mine underground operators method of counting of view-based access control model |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN105513043B (en) |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106355723B (en) * | 2016-10-19 | 2018-08-14 | 中国矿业大学(北京) | Mine operation personnel method of counting based on image |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103218598A (en) * | 2013-03-26 | 2013-07-24 | 中国科学院电子学研究所 | Method for automatically detecting remote sensing ground object target based on stochastic geometry model |
CN104732552A (en) * | 2015-04-09 | 2015-06-24 | 西安电子科技大学 | SAR image segmentation method based on nonstationary condition field |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP1916624B1 (en) * | 2006-10-25 | 2016-11-23 | Agfa HealthCare NV | Method for segmenting a digital medical image. |
-
2015
- 2015-11-20 CN CN201510801099.XA patent/CN105513043B/en not_active Expired - Fee Related
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103218598A (en) * | 2013-03-26 | 2013-07-24 | 中国科学院电子学研究所 | Method for automatically detecting remote sensing ground object target based on stochastic geometry model |
CN104732552A (en) * | 2015-04-09 | 2015-06-24 | 西安电子科技大学 | SAR image segmentation method based on nonstationary condition field |
Non-Patent Citations (3)
Title |
---|
一种基于双椭圆模型的视频人数统计方法;张继法 等;《计算机科学》;20120630;第39卷(第6A期);第499-500页 * |
一种新的基于吉布斯随机场的视频运动对象分割算法;刘龙 等;《自动化学报》;20070630;第33卷(第6期);第609、611页 * |
非参数吉布斯模型和多波段遥感影像纹理分割方法研究;龚衍 等;《武汉大学学报信息科学版》;20070731;第32卷(第7期);第581-584页 * |
Also Published As
Publication number | Publication date |
---|---|
CN105513043A (en) | 2016-04-20 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Huang et al. | Detection algorithm of safety helmet wearing based on deep learning | |
CN102831618B (en) | Hough forest-based video target tracking method | |
CN110188807A (en) | Tunnel pedestrian target detection method based on cascade super-resolution network and improvement Faster R-CNN | |
CN106446926A (en) | Transformer station worker helmet wear detection method based on video analysis | |
CN114627140B (en) | Coal mine ventilator intelligent adjusting method based on high-voltage frequency converter | |
WO2016077521A1 (en) | System and method for measuring characteristics of cuttings and fluid front location during drilling operations with computer vision | |
CN103065325B (en) | A kind of method for tracking target based on the polymerization of color Distance geometry Iamge Segmentation | |
CN104166861A (en) | Pedestrian detection method | |
CN104392464B (en) | A kind of artificial intrusion detection method based on color video frequency image | |
CN104408745A (en) | Real-time smog scene detection method based on video image | |
CN112149761A (en) | Electric power intelligent construction site violation detection method based on YOLOv4 improved algorithm | |
CN107909044B (en) | People counting method combining convolutional neural network and track prediction | |
CN105469069A (en) | Safety helmet video detection method for production line data acquisition terminal | |
CN105426820A (en) | Multi-person abnormal behavior detection method based on security monitoring video data | |
CN104732236B (en) | A kind of crowd's abnormal behaviour intelligent detecting method based on layered shaping | |
CN103049765A (en) | Method for judging crowd density and number of people based on fish eye camera | |
CN105513043B (en) | The coal mine underground operators method of counting of view-based access control model | |
CN104050677A (en) | Hyper spectrum image segmentation method based on multilayer neural network | |
Wiliem et al. | Detecting uncommon trajectories | |
CN104732216A (en) | Expression recognition method based on key points and local characteristics | |
Wu et al. | Automated visual helmet identification based on deep convolutional neural networks | |
CN115035088A (en) | Helmet wearing detection method based on yolov5 and posture estimation | |
CN105306909B (en) | The overcrowding warning system of coal mine underground operators of view-based access control model | |
CN102521646A (en) | Complex scene people counting algorithm based on depth information cluster | |
CN104537636A (en) | Image detection method for 'black hole effect' at tunnel entrance |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20180403 Termination date: 20191120 |
|
CF01 | Termination of patent right due to non-payment of annual fee |