CN105869184A - Forest fire smoke image detection method based on path analysis - Google Patents
Forest fire smoke image detection method based on path analysis Download PDFInfo
- Publication number
- CN105869184A CN105869184A CN201610206104.7A CN201610206104A CN105869184A CN 105869184 A CN105869184 A CN 105869184A CN 201610206104 A CN201610206104 A CN 201610206104A CN 105869184 A CN105869184 A CN 105869184A
- Authority
- CN
- China
- Prior art keywords
- image
- region
- smog
- forest
- computer
- 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.)
- Granted
Links
- 238000001514 detection method Methods 0.000 title abstract description 13
- 239000000779 smoke Substances 0.000 title abstract description 6
- 238000000034 method Methods 0.000 claims abstract description 22
- 238000009792 diffusion process Methods 0.000 claims abstract description 4
- 230000000694 effects Effects 0.000 claims description 3
- 239000007787 solid Substances 0.000 claims 1
- 238000012544 monitoring process Methods 0.000 abstract description 11
- 238000012545 processing Methods 0.000 abstract description 5
- 238000005516 engineering process Methods 0.000 description 4
- 238000006243 chemical reaction Methods 0.000 description 2
- 230000009466 transformation Effects 0.000 description 2
- 238000004891 communication Methods 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- DMBHHRLKUKUOEG-UHFFFAOYSA-N diphenylamine Chemical compound C=1C=CC=CC=1NC1=CC=CC=C1 DMBHHRLKUKUOEG-UHFFFAOYSA-N 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 230000001788 irregular Effects 0.000 description 1
- 238000004321 preservation Methods 0.000 description 1
- 238000012549 training Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
-
- 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
- G06T2207/20—Special algorithmic details
- G06T2207/20212—Image combination
- G06T2207/20224—Image subtraction
-
- 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
- G06T2207/30232—Surveillance
Landscapes
- Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Geometry (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Image Analysis (AREA)
- Fire-Detection Mechanisms (AREA)
Abstract
The invention discloses a forest fire smoke image detection method based on path analysis. According to different motion paths of different objects under the influence of the atmosphere and inherent features of smoke at a diffusion initial source, through image processing of a background computer, a scope of a motion path angle is obtained, accordingly, whether a fire hazard takes place is determined, such that objectivity of object motion is ensured, image data is converted into mathematic formulas, features of data processing of the computer is fully utilized, and the response speed of the computer is fast. The detection method provided by the invention does not have to additionally process image data of other portions, only processes and analyzes a dynamic area, reduces the computation amount and improves the monitoring efficiency. The method is adapted to interfaces and images of the computer, enables the response of the computer to be more rapid and is applied to long-distance non-contact monitoring of a forest area.
Description
Technical field
The invention belongs to technical field of image detection, relate to one and utilize ccd video camera that forest fires smog is carried out
Detection method, is particularly suitable for the monitoring of remote forest or the monitoring of forest district.
Background technology
In forest fire is investigated, the monitoring of fire is a very important content.Traditional fire detection
Often relying on forest ranger, Watchtower carries out manual detection, and its detection range is limited, it is impossible to detect at a distance, and
And limited personnel, efficiency comparison is low.The method using video monitoring carries out the noncontact monitoring energy of forest fire
The detection region enough making up tradition monitoring method is little, and shortage of manpower the most increasingly comes into one's own.
University of Electronic Science and Technology proposes to judge doubtful smog district with according to color characteristic, and extracts doubtful smog district shape
Shape is irregular, area increases, grader is trained by K blurred background feature, differentiates smog with grader
(Zhou Xue, Zou Jianxiao, Xu Hongbing, Deng Lin, video smoke detection method based on multi-feature fusion, China
Patent: 201510170159.2,2015-4-10.), the characteristic vector that this method calculates is many, and data are loaded down with trivial details,
And big to the performance quality dependency of grader, forest is the region of an opening, is affected greatly by geographical conditions,
And there is the feature of miniclimate, the generation of smog is also randomness, therefore the training radix of grader big and
Substantial amounts of data are difficult to obtain, and can cause data redundancy, are not suitable for detecting the feature of labile state.
Wuxi the Big Dipper communication breath Science and Technology Ltd. proposes a kind of forest smoke region detecting system, this detection
System is positioned on unmanned plane, and this system carries out forest detection by accepting the instruction on ground, although can be quick
Determine smoke region, but capital expenditure is relatively big, and forest can not aloft be examined in the moment by unmanned plane
Survey, and easily ignore narrow regions and the region of special geological surrounding.
Fujitsu Ltd. propose a kind of multiframe to image to be detected carry out feature extraction, it is thus achieved that one or
Multiple candidate regions, then analyze the area change in multiple frame, the direction of motion and change of shape, sentence
Whether disconnected candidate region exists smog, relates to data various in the method, and processing speed is relatively slow, wherein by institute
Select the barycenter of candidate region in the highest direction of motion of the frequency of occurrences as direction of primary motion, but forest environment
Changing unpredictably, depend on air motion, the discriminant accuracy of direction of primary motion is indefinite, increases rate of false alarm.
Summary of the invention
Instant invention overcomes the deficiency of above-mentioned technology, it is provided that a kind of forest fires smog image based on path analysis is examined
Survey method, its object is to efficient, sensitive, remote, noncontact monitoring forest zone.
The forest fires smog image detecting method based on path analysis of the present invention, is arranged on the CCD of forest zone eminence
The video camera moment monitors whole forest farm, and when fire occurs to produce smog, the change in this region is taken the photograph by CCD
Camera captures, and gathers the smog image in a period of time, according to the affiliated scope of smog movement path angle
Determining whether this region has fire to produce, the method is carried out according to the following steps,
(1) ccd video camera being positioned at forest zone eminence captures the dynamic change in certain region,
(2) ccd video camera gathers the view data of this region a period of time, and is sent to background computer,
Carry out Image semantic classification, image be converted into binary image,
(3) image in this region is partitioned into dynamic region of variation by background computer, and is recorded as A,
(4) the coordinate position A of each barycenter in the connected domain of region A is calculated1(x1, y1), A2(x2, y2) depend on
Secondary record also preserves,
(5) relatively each center-of-mass coordinate size, and remember that the maximum center-of-mass coordinate of coordinate figure is B0(x0, y0), sit
The center-of-mass coordinate of scale value minimum is B1(x1, y1), coordinate figure be second largest center-of-mass coordinate be B2(x2, y2), take
B0With B1Between barycenter, line is θ, B with the angle of horizontal direction0With B2Line and the angle of horizontal direction between barycenter
For α,
(6) value of θ and α is calculated,
(7) judge the size of θ and α, and determine scope,
(8) if θ and α is between 0 °~90 °, then background computer window display result: be, otherwise,
Show no.
Described forest fires smog image detecting method based on path analysis, utilizes ccd video camera to gather dynamically
Area image, different according to different objects motion path under atmospheric effect, and smog is initial in diffusion
Feature intrinsic at source, through the image procossing of background computer, draws the scope at motion path angle, thus
Determine whether there is fire to produce.
Present invention foundation different objects motion path under atmospheric effect is different, and smog is initial in diffusion
Feature intrinsic at source, through the image procossing of background computer, draws the scope at motion path angle, thus
Determine whether there is fire to produce, it is ensured that the objectivity of thing movement, view data be converted into mathematical formulae,
Make full use of the feature of computer digital animation, make computer fast response time.Detection provided by the present invention
Method, it is not necessary to the view data of other parts of extra process, just for Treatment Analysis dynamic area, reduces
Operand, improves monitoring efficiency.The present invention adapts with interfaceization and the image conversion of computer so that
Computer response is rapider, it is adaptable to remote noncontact monitoring forest zone.
Detailed description of the invention
Forest fires smog image detecting method based on path analysis, is arranged on the ccd video camera moment of forest zone eminence
Monitor whole forest farm, when fire occurs to produce smog, the change in this region is captured by ccd video camera,
And gather the smog image in a period of time, determine that this region is according to the affiliated scope of smog movement path angle
No have fire to produce, and the method is carried out according to the following steps,
(1) ccd video camera being positioned at forest zone eminence captures the dynamic change in certain region,
Double frame differences obtain difference diagram: Dn(i, j)=| In(i, j)-In-1(i, j) |, In, In-1Represent phase in image sequence
Two adjacent two field pictures, carry out thresholding and obtain sport foreground region difference image:
(2) ccd video camera gathers the view data of this region a period of time, and is sent to background computer,
Carry out Image semantic classification, image be converted into binary image,
Image is carried out gray processing process, and carries out greyscale transformation, make the contrast of image increase and be prone to two-value
Change processes:
Gray processing: f (i, j)=0.03R (i, j)+0.59G (i, j)+0.11B (i, j)
Greyscale transformation: DB=f (DA)=fDA+fB
By iterative method, image being carried out binary conversion treatment, its step is as follows:
First obtain maximum gradation value and the minimum gradation value of image, be designated as R respectivelymaxAnd Rmin, make threshold value
T=(Rmax+Rmin)/2, are divided into two groups of R further according to threshold value T by the average gray value of image1And R2, distinguish afterwards
Obtain the average gray value μ of two groups1And μ2, finally obtain new threshold value T=(μ1+μ2)/2, i.e. set T as the overall situation
The data of image are divided into two parts with T: the pixel group more than T and the pixel group less than T by threshold value, its
In the pixel value of pixel group more than T be set as white, the pixel value of the pixel group less than T is set as black,
Wherein white portion is dynamic region of variation,
(3) image in this region is partitioned into dynamic region of variation by background computer, and is recorded as A,
(4) the coordinate position A of each barycenter in the connected domain of region A is calculated1(x1, y1), A2(x2, y2) etc.
And record preservation successively,
Barycenter mathematic(al) representation:
(5) relatively each center-of-mass coordinate size, and remember that the maximum center-of-mass coordinate of coordinate figure is B0(x0, y0), sit
The center-of-mass coordinate of scale value minimum is B1(x1, y1), coordinate figure be second largest center-of-mass coordinate be B2(x2, y2), take
B0With B1Between barycenter, line is θ, B with the angle of horizontal direction0With B2Line and the angle of horizontal direction between barycenter
For α,
(6) calculating the value of θ and α, unit employing degree represents,
(7) judge the size of θ and α, and determine its scope,
(8) if θ and α is between 0 °~90 °, then background computer window display result: be, otherwise,
Show no.
Claims (2)
1. forest fires smog image detecting method based on path analysis, is arranged on the ccd video camera of forest zone eminence
Moment monitors whole forest farm, and when fire occurs to produce smog, the change in this region is caught by ccd video camera
Grasp, and gather the smog image in a period of time, determine this according to the affiliated scope of smog movement path angle
Whether region has fire to produce, and the method is carried out according to the following steps,
(1) ccd video camera being positioned at forest zone eminence captures the dynamic change in certain region,
(2) ccd video camera gathers the view data of this region a period of time, and is sent to background computer,
Carry out Image semantic classification, image be converted into binary image,
(3) image in this region is partitioned into dynamic region of variation by background computer, and is recorded as A,
(4) the coordinate position A of each barycenter in the connected domain of region A is calculated1(x1, y1), A2(x2, y2) depend on
Secondary record also preserves,
(5) relatively each center-of-mass coordinate size, and remember that the maximum center-of-mass coordinate of coordinate figure is B0(x0, y0), sit
The center-of-mass coordinate of scale value minimum is B1(x1, y1), coordinate figure be second largest center-of-mass coordinate be B2(x2, y2), take
B0With B1Between barycenter, line is θ, B with the angle of horizontal direction0With B2Line and the angle of horizontal direction between barycenter
For α,
(6) value of θ and α is calculated,
(7) judge the size of θ and α, and determine scope,
(8) if θ and α is between 0 °~90 °, then background computer window display result: be, otherwise,
Show no.
2. forest fires smog image detecting method based on path analysis as claimed in claim 1, its feature exists
In, different according to different objects motion path under atmospheric effect, and smog is solid at diffusion starting resource
Some features, through the image procossing of background computer, draw the scope at motion path angle, thus judgement are
No have fire to produce.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610206104.7A CN105869184B (en) | 2016-04-06 | 2016-04-06 | Forest fires smog image detecting method based on path analysis |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610206104.7A CN105869184B (en) | 2016-04-06 | 2016-04-06 | Forest fires smog image detecting method based on path analysis |
Publications (2)
Publication Number | Publication Date |
---|---|
CN105869184A true CN105869184A (en) | 2016-08-17 |
CN105869184B CN105869184B (en) | 2019-06-28 |
Family
ID=56627989
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610206104.7A Expired - Fee Related CN105869184B (en) | 2016-04-06 | 2016-04-06 | Forest fires smog image detecting method based on path analysis |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN105869184B (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108279287A (en) * | 2018-02-01 | 2018-07-13 | 李绍辉 | Smog Quick diffusing system based on the communication technology |
CN109410512A (en) * | 2018-11-07 | 2019-03-01 | 北京林业大学 | A kind of smog root node detection method based on least square method |
CN109903490A (en) * | 2019-03-25 | 2019-06-18 | 哈尔滨工程大学 | A kind of calculation method of fire evacuation path and instruction system |
CN114549442A (en) * | 2022-02-14 | 2022-05-27 | 常州市新创智能科技有限公司 | Real-time monitoring method, device and equipment for moving object and storage medium |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103456123A (en) * | 2013-09-03 | 2013-12-18 | 中国科学技术大学 | Video smoke detection method based on flowing and diffusion characters |
CN104715559A (en) * | 2015-03-06 | 2015-06-17 | 温州大学 | Smoke detection and fire alarm method based on trajectory identification |
CN104978733A (en) * | 2014-04-11 | 2015-10-14 | 富士通株式会社 | Smoke detection method and smoke detection device |
-
2016
- 2016-04-06 CN CN201610206104.7A patent/CN105869184B/en not_active Expired - Fee Related
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103456123A (en) * | 2013-09-03 | 2013-12-18 | 中国科学技术大学 | Video smoke detection method based on flowing and diffusion characters |
CN104978733A (en) * | 2014-04-11 | 2015-10-14 | 富士通株式会社 | Smoke detection method and smoke detection device |
CN104715559A (en) * | 2015-03-06 | 2015-06-17 | 温州大学 | Smoke detection and fire alarm method based on trajectory identification |
Non-Patent Citations (3)
Title |
---|
PING-HE HUANG 等: "A Fire-Alarming Method Based on Video Processing", 《COMPUTER SOCIETY》 * |
卢胜男等: "视频火灾识别方法研究综述", 《电视技术》 * |
毕振波等: "图像型火灾探测预处理方法综述", 《消防科学与技术》 * |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108279287A (en) * | 2018-02-01 | 2018-07-13 | 李绍辉 | Smog Quick diffusing system based on the communication technology |
CN108279287B (en) * | 2018-02-01 | 2020-12-18 | 嘉兴市丰成五金材料股份有限公司 | Smoke rapid emission system based on communication technology |
CN109410512A (en) * | 2018-11-07 | 2019-03-01 | 北京林业大学 | A kind of smog root node detection method based on least square method |
CN109903490A (en) * | 2019-03-25 | 2019-06-18 | 哈尔滨工程大学 | A kind of calculation method of fire evacuation path and instruction system |
CN109903490B (en) * | 2019-03-25 | 2020-12-01 | 哈尔滨工程大学 | Fire evacuation path calculation method and indication system |
CN114549442A (en) * | 2022-02-14 | 2022-05-27 | 常州市新创智能科技有限公司 | Real-time monitoring method, device and equipment for moving object and storage medium |
Also Published As
Publication number | Publication date |
---|---|
CN105869184B (en) | 2019-06-28 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106600888B (en) | Automatic forest fire detection method and system | |
CN111462451B (en) | Straw burning detection alarm system based on video information | |
CN107437318B (en) | Visible light intelligent recognition algorithm | |
CN111985365A (en) | Straw burning monitoring method and system based on target detection technology | |
CN105869184A (en) | Forest fire smoke image detection method based on path analysis | |
KR102149832B1 (en) | Automated Violence Detecting System based on Deep Learning | |
Li et al. | Toward efficient safety helmet detection based on YoloV5 with hierarchical positive sample selection and box density filtering | |
CN110096945B (en) | Indoor monitoring video key frame real-time extraction method based on machine learning | |
CN111899452A (en) | Forest fire prevention early warning system based on edge calculation | |
CN111046827A (en) | Video smoke detection method based on convolutional neural network | |
CN105575034A (en) | Image processing and analysis method of double-waveband forest fireproof intelligent monitoring software | |
CN106778488B (en) | Low-light (level) smog video detecting method based on image correlation | |
KR101196678B1 (en) | Real-time fire detection device and method | |
CN111476964A (en) | Remote forest fire prevention monitoring system and method | |
CN111401310B (en) | Kitchen sanitation safety supervision and management method based on artificial intelligence | |
CN110738218A (en) | Method and device for identifying hidden danger of smoke and fire of power transmission line channels | |
CN110852303A (en) | Eating behavior identification method based on OpenPose | |
CN112784835A (en) | Method and device for identifying authenticity of circular seal, electronic equipment and storage medium | |
CN101930540A (en) | Video-based multi-feature fusion flame detecting device and method | |
CN114067396A (en) | Vision learning-based digital management system and method for live-in project field test | |
CN101916380B (en) | Video-based device and method for detecting smog | |
CN106303461A (en) | Movable-type intelligent safety device based on video analysis | |
KR102602439B1 (en) | Method for detecting rip current using CCTV image based on artificial intelligence and apparatus thereof | |
CN101930541A (en) | Video-based flame detecting device and method | |
CN111797761B (en) | Three-stage smoke detection system, method and readable medium |
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 | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20190628 |