CN106485281A - A kind of detection method of the Air haze class of pollution - Google Patents
A kind of detection method of the Air haze class of pollution Download PDFInfo
- Publication number
- CN106485281A CN106485281A CN201610910115.3A CN201610910115A CN106485281A CN 106485281 A CN106485281 A CN 106485281A CN 201610910115 A CN201610910115 A CN 201610910115A CN 106485281 A CN106485281 A CN 106485281A
- Authority
- CN
- China
- Prior art keywords
- image
- pollution
- information
- class
- edge
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
- G06F18/253—Fusion techniques of extracted features
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N33/00—Investigating or analysing materials by specific methods not covered by groups G01N1/00 - G01N31/00
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/30—Noise filtering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Data Mining & Analysis (AREA)
- Chemical & Material Sciences (AREA)
- Health & Medical Sciences (AREA)
- Multimedia (AREA)
- General Engineering & Computer Science (AREA)
- Medicinal Chemistry (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Computation (AREA)
- Evolutionary Biology (AREA)
- Food Science & Technology (AREA)
- Artificial Intelligence (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- General Health & Medical Sciences (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Image Processing (AREA)
Abstract
The invention discloses a kind of detection method of the Air haze class of pollution, the image in different pollutions is carried out Fourier transformation, the picture quality under quantitative analyses difference pollution level, extracts edge and quantified, find out its Characteristic Distribution；High frequency filter experiment is carried out to the image in different pollutions, to filter the low frequency component in image, draws the contrast in image and entropy information；Solbel Image Edge-Detection is carried out to image, carries out convolution algorithm by building the window operator mating with original image, obtain image edge information；In conjunction with the result of image feature value and the pollution of humidity information comprehensive descision, judged；By the corresponding class of pollution of fusion image information characteristics, judge the humidity of humidity sensor, then export the class of pollution.The present invention is standardized using information technology, and standardized being implemented with is beneficial to the unification of impression information and metrical information, and reliability is high, real-time is high and low cost.
Description
Technical field
The present invention relates to haze detection field, specifically a kind of detection method of the Air haze class of pollution.
Background technology
The meaning of haze is that the aerosol systems of the particulate matters such as the dust of gray haze in the air, sulphuric acid, nitric acid composition cause vision
Obstacle be haze.When condensation vapor aggravation, air humidity increase, haze will be converted into mist.Haze is haze with the difference of mist
When relative humidity little.Haze is common in city.Much mist is incorporated to haze and enters together as diastrous weather phenomenon by area for China
Row early-warning and predicting, is referred to as " haze weather ".
Haze is the result that specific weather condition is interacted with mankind's activity.The economy of high density population and social activity
A large amount of fine particles will necessarily be discharged（PM 2.5）, once discharge exceedes atmospheric air circulation ability and carrying degree, fine particle concentration
By continued accumulation, if affected by quiet steady weather etc., large-scale haze easily occurs.Haze not only has seriously to environment
Harm, also have serious harm to our human bodies.But I found that based on present by entering to air particles material composition
The method of row analysis often occurs in that to the detection of haze it is dirty that local in short-term in such as instrument location with the phenomenon of a capping
Dye, instrument will show that representative urban area occurs in that different degrees of pollution, or when instrument location not by
When grain covers, and in fact occur in that fairly large air pollution, then instrument still will not show objective result.
In addition, the display of PM2.5 data has little significance from civilian angle, people more concerned be Current air pollution grade
What is, such as air quality is good, slight pollution, intermediate pollution, serious pollution etc..
In view of when haze sky occurs, people can feel the physiology such as obvious vision low visibility, air drying
Impression is although different people is different to the sensitivity of this impression, but the impression of people has height when haze occurs
Concordance.But, different people be experienced as subjectivity, not through quantifying, so in order to be able to by this physiological impression
Quantify, and corresponding with the air pollution grade of standard at that time, how to obtain one and there is unified quantization index and real-time display
Air pollution level detection method be particularly important.When haze weather occurs, human eye visual perception low visibility,
And air is relatively dry.Existing sol gel process can not summarize the air quality within certain limit.In order to vision
The result of perception carries out quantification, to exclude the diversity between individual perception, makes the equipment that can reflect the class of pollution, full
The demand that sufficient people forecast to the class of pollution.
Content of the invention
It is an object of the invention to provide a kind of detection method of the Air haze class of pollution, to solve above-mentioned background technology
The problem of middle proposition.
For achieving the above object, the present invention provides following technical scheme：
A kind of detection method of the Air haze class of pollution, step is as follows：
（1）Image in different pollutions is carried out Fourier transformation, the picture quality under quantitative analyses difference pollution level, extracts
Edge is simultaneously quantified, and finds out its Characteristic Distribution；
（2）If the size of image f (x, y) is M*N, then its two-dimensional Fourier transform formula is：
By its discretization, calculate the Fourier spectrum under different pollution levels；
（3）High frequency filter experiment is carried out to the image in different pollutions, to filter the low frequency component in image, draws in image
Contrast and entropy information；
（4）Edge detection algorithm using difference approximation differential carries out Solbel Image Edge-Detection to image, by structure
The window operator joined and original image carry out convolution algorithm, obtain image edge information；
Using solbel operator be：
Intermediate point pixel value after its convolution algorithm is：
；
Wherein, f is the pixel value of a sub-picture, and x, y are respectively the coordinate of pixel in image；
Convolution operator B gives central point pixel assignment in calculating process again, is exported with central point for result of calculation, result convergence
In 0, show as " black " in gradation of image；Result levels off to 255, shows as " white " in gradation of image；
（5）On the basis of image recognition, select external humidity sensor, comprehensively sentence in conjunction with image feature value and humidity information
The result of disconnected pollution, is judged；
（6）By the corresponding class of pollution of fusion image information characteristics, judge whether the humidity of humidity sensor is less than simultaneously
80%, if two conditions meet simultaneously, export the class of pollution, otherwise input as there being greasy weather gas.
As the further scheme of the present invention：Described external humidity sensor is the humidity of Minitype digital signaling mode
Sensor.
Compared with prior art, the invention has the beneficial effects as follows：
The present invention be directed to the limited local detection run in the detection of existing haze, with put general face, poor real, use cost high,
The problems such as be difficult to large-scale promotion civilian use it is proposed that merge the Air haze pollution of humidity information detection based on image recognition
Level detection method, by extract haze image marginal information, and carried out quantization operations obtain reflect haze pollution level
Characteristic quantity, simultaneously merge humidity information detection result carry out comprehensive descision.Present invention physiology in haze sky by people
Experience result to be standardized using information technology, standardized being implemented with is beneficial to the unification of impression information and metrical information,
Thus provide the detection method of a kind of reliability height, real-time height and low cost for people.
Brief description
Fig. 1 is that the image under different degrees of pollution is illustrated such as.
Image spectrum figure under the difference pollutional condition of Fig. 2 position.
Fig. 3 is the image low frequency filtering result figure under intermediate pollution and non-contaminated state.
Fig. 4 is the convolution process schematic diagram of Sobel operator and image.
The haze image edge result schematic diagram that Fig. 5 obtains for Sobel operator.
Fig. 6 pollutes figure for partial air haze.
Fig. 7 is that the result of calculation of characteristics of image represents schematic diagram.
Fig. 8 is that characteristics of image merges decision flow chart with humidity information.
Specific embodiment
Below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is carried out clear, complete
Site preparation description is it is clear that described embodiment is only a part of embodiment of the present invention, rather than whole embodiments.It is based on
Embodiment in the present invention, it is every other that those of ordinary skill in the art are obtained under the premise of not making creative work
Embodiment, broadly falls into the scope of protection of the invention.
The intuitive visual of haze image is experienced as have impact on the contrast of image, as can be seen from Figure 1 dirty with haze
The increase of dye degree, vision visibility is reducing, therefore, if to inform the angle of user current air pollution level,
User only needs to obtain the qualitative conclusions of pollution level.Meanwhile, the appearance in order to avoid the visibility that mist causes reduces
Erroneous judgement, the present invention is judged using the result that Humidity Detection fusion image identifies.
For the picture quality under quantitative analyses difference pollution level, first the image in different pollutions is carried out Fourier
Conversion, finds out its Characteristic Distribution.From figure 2 it can be seen that with the increase of pollution level, the high fdrequency components of image are subtracting
Weak.And in graphical analyses, high fdrequency components mean the edge hopping part in image, so the essence that haze image is processed is
Extract edge and the process quantifying.
If the size of image f (x, y) is M*N, then its two-dimensional Fourier transform formula is：
,
Will be as shown in Figure 2 for the Fourier spectrum calculating after its discretization under different pollution levels.
In order to verify effectiveness assumed above, do High frequency filter experiment, sent out by filtering the low frequency component in image
Contrast in existing image and entropy information, as shown in Figure 3.By Fig. 3 it is recognised that free of contamination image is by high-pass filtering
After device, more clearly marginal information can be shown, otherwise edge is fuzzyyer.Illustrate that the detection to haze grade is final etc.
Imitate in the extraction to image edge information, the essence of therefore haze image feature extraction is number of edges statistics of variables, then adopt side
The method of edge detection fusion quantitative description.
By above analysis, the vision visibility description conversion in the measurement of the haze class of pollution in order to image border
The extraction of information, therefore, thinking below has been converted to two aspects：First, how image is detected from original image
Edge；Second, the image that how be contains only with marginal information carries out quantitative expression and description, that is, characteristic quantity
Change.
The method that image edge information extracts has many kinds, for example：Wavelet analysis method, grey scale difference statistics, Mathematical Morphology
Etc. multiple methods, the extraction of haze image marginal information mainly can reflect the perceived effect of vision, therefore special in image
The precision aspect levying extraction need not pursue too high precision, on the premise of can embodying image vision profile, simultaneously need to and
Get everything ready the requirement having calculating speed fast.
The maximum advantage of spatial domain image processing method is that have higher calculating speed, also needs on this algorithm direction
To weaken computation complexity further, such as grey scale difference statistical method needs to carry out secondary calculating and the form of statistic
Opening operation in method and closed operation.
In view of the extraction that it is critical only that margin signal of haze image feature, and the essence at edge is because gray value
Drastically change causes, and is equivalent to calculating extreme point in the picture, and asking for of extreme point is to pass through from the point of view of mathematics
Derivation is completing.And the discontinuity due to picture signal, derivative operation can be with the convolution between Space Operators and image
Computing is completing.Concrete calculating process is described as follows section.
Solbel Image Edge-Detection is the edge detection algorithm using difference approximation differential, by building suitable window
Operator and original image carry out convolution algorithm, obtain image edge information.
Using solbel operator be：
Intermediate point pixel value after its convolution algorithm is：
；
Wherein, f is the pixel value of a sub-picture, and x, y are respectively the coordinate of pixel in image；
Convolution operator B gives central point pixel assignment in calculating process again, is exported with central point for result of calculation, to original graph
In the 3*3 contiguous range of picture, pixel value is multiplied by the weights on B matrix element.If the region that B matrix is covered has gray value phase
As feature, then the essence after weights computing be by difference operation, result, close to 0, shows as in gradation of image
" black ".If the position that B matrix area is located is image border, the essence after convolution calculus of differences has been by ash
The enhancing of angle value, result levels off to 255, shows as " white " in gradation of image.
Collecting 100 width has the picture of class of pollution record, wherein comprises pollution-free, slight pollution, intermediate pollution, moderate dirty
Dye respectively has 25 width, and parts of images is as shown in Figure 6.As can be seen from the figure with the increase of pollution level, eigenvalue reduces therewith,
Mean that the quantity of information of image reduces, reflect that the visibility in image reduces.
Because with the difference of haze, mist is that humidity is different, in order to prevent image visibility because mist causes, contrast fall
Low problem, therefore separately through image vision visibility can not the fully defining class of pollution, need the knot with reference to Humidity Detection
Really.The present invention proposes, on the basis of images above information characteristics extract, to incorporate humidity sensor detection.According to haze and mist
Difference understands, differentiation therebetween is mainly whether dependence humidity is distinguished more than 80%.Therefore, in the base of image recognition
On plinth, need the result of calculation of characteristics of image to merge the result of Humidity Detection, therefore on the selection issue of humidity sensor
Following standard should be followed：
（1）External humidity sensor.In view of the particularity of air humidity detection, sensor is needed fully to connect with composition of air
Touch, except image detection and processing unit, humidity sensor is structurally independent from image section, and need to do is to the two
Operation result merged.
（2）Miniature and digital signal standard.With portable equipment as goal in research, then should consider embedding from lectotype selection angle
Enter formula image recognition hardware system, and the humidity sensor of miniaturization, additionally, in order that characteristics of image signal and moisture signal
Between efficiently merge, then be used uniformly across digital signal outut device.Entered by computer program in follow-up fusion judges
The comprehensive descision of the row haze class of pollution.
According to above principle, select DWTCP-D（T）The external humidity sensor of model, in conjunction with image feature value and wet
Degree informix judges the result of pollution, and the flow chart of algorithm is as shown in Figure 8.By having merged the corresponding dirt of image information feature
Dye grade, judges whether the humidity of humidity sensor is less than 80% simultaneously, if two conditions meet simultaneously, exports this pollution etc.
Level, otherwise inputs as there being greasy weather gas.
Result and analysis
Result above shows effectiveness in processing haze image for the inventive method and real-time, and particular content is analyzed as follows：
（1）Data Source employed in experiment, in China Meteorological net weather forecast result and typical image features, therefore gathers
Image have reliable data to support for the quantitative analyses of the haze class of pollution.
（2）Data test result indicate that being incremented by with the haze class of pollution, image feature value occurs in that under significantly
Stage display is dropped and is presented.Wherein there is parts of images saltus step more obvious, reason is in the collection of image to contain part
Close shot region, and the image acquisition of the present invention can obtain more preferable test effect under having the openr visual field.
（3）Algorithm due to adopting is the analysis theories of spatial domain, and the method for character representation and description is that image is entered
The direct computing of row, therefore fundamentally ensure that the calculating real-time of algorithm, and this point is critically important during commercialization
Consider index.During commercialization, the cost budgeting of this inventive method sent for 3 generations embedded for 50 yuan+Fructus Rubi of RGB photographic head
210 yuan of+DWTCP-D of processor（T）420 yuan=680 yuan of humidity sensor.
（4）The method of image information fusion humidity sensor signal of the present invention has the work distinguishing haze and mist
With the key link not referring in conventional document and solving.The thinking that the inventive method is realized can be by the inspection of the two
Survey result to make a distinction, detecting for civilian haze provides accurate information.
（5）The present invention from the feature description fineness improving algorithm and overcomes image two-shot object, proposes not to be subject to
The feature recognition algorithms of image capturing angle constraint.
It is obvious to a person skilled in the art that the invention is not restricted to the details of above-mentioned one exemplary embodiment, and do not carrying on the back
In the case of the spirit or essential attributes of the present invention, the present invention can be realized in other specific forms.Therefore, no matter from which
From the point of view of a bit, embodiment all should be regarded as exemplary, and be nonrestrictive, the scope of the present invention will by appended right
Ask rather than described above limits, it is intended that all changes that will fall in the implication and scope of the equivalency of claim
Include in the present invention.Any reference in claim should not be considered as limiting involved claim.
Claims (2)
1. a kind of detection method of the Air haze class of pollution is it is characterised in that step is as follows：
（1）Image in different pollutions is carried out Fourier transformation, the picture quality under quantitative analyses difference pollution level, extracts
Edge is simultaneously quantified, and finds out its Characteristic Distribution；
（2）If the size of image f (x, y) is M*N, then its two-dimensional Fourier transform formula is：
,
By its discretization, calculate the Fourier spectrum under different pollution levels；
（3）High frequency filter experiment is carried out to the image in different pollutions, to filter the low frequency component in image, draws in image
Contrast and entropy information；
（4）Edge detection algorithm using difference approximation differential carries out Solbel Image Edge-Detection to image, by structure
The window operator joined and original image carry out convolution algorithm, obtain image edge information；
Using solbel operator be：
Intermediate point pixel value after its convolution algorithm is：
；
Wherein, f is the pixel value of a sub-picture, and x, y are respectively the coordinate of pixel in image；
Convolution operator B gives central point pixel assignment in calculating process again, is exported with central point for result of calculation, result convergence
In 0, show as " black " in gradation of image；Result levels off to 255, shows as " white " in gradation of image；
（5）On the basis of image recognition, select external humidity sensor, comprehensively sentence in conjunction with image feature value and humidity information
The result of disconnected pollution, is judged；
（6）By the corresponding class of pollution of fusion image information characteristics, judge whether the humidity of humidity sensor is less than simultaneously
80%, if two conditions meet simultaneously, export the class of pollution, otherwise input as there being greasy weather gas.
2. the detection method of the Air haze class of pollution according to claim 1 is it is characterised in that described step（5）In,
External humidity sensor is the humidity sensor of Minitype digital signaling mode.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610910115.3A CN106485281A (en) | 2016-10-19 | 2016-10-19 | A kind of detection method of the Air haze class of pollution |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610910115.3A CN106485281A (en) | 2016-10-19 | 2016-10-19 | A kind of detection method of the Air haze class of pollution |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106485281A true CN106485281A (en) | 2017-03-08 |
Family
ID=58271048
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610910115.3A Pending CN106485281A (en) | 2016-10-19 | 2016-10-19 | A kind of detection method of the Air haze class of pollution |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106485281A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108052980A (en) * | 2017-12-21 | 2018-05-18 | 南京理工大学 | Air quality grade detection method based on image |
| CN110879276A (en) * | 2019-10-31 | 2020-03-13 | 青岛崂应海纳光电环保集团有限公司 | Filtering method for gas analysis instrument |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103218622A (en) * | 2013-04-22 | 2013-07-24 | 武汉大学 | Haze monitoring method based on computer vision |
| CN103558133A (en) * | 2013-11-11 | 2014-02-05 | 河南牧业经济学院 | Method and system for monitoring haze concentration based on image analyzing technology |
| CN104517126A (en) * | 2014-12-26 | 2015-04-15 | 北京邮电大学 | Air quality assessment method based on image analysis |
| CN104749078A (en) * | 2015-04-13 | 2015-07-01 | 武汉大学 | PM2.5 concentration detector |
| CN104809735A (en) * | 2015-05-12 | 2015-07-29 | 公安部第三研究所 | System and method for realizing image fog-haze evaluation based on Fourier transformation |
| CN105403490A (en) * | 2015-05-22 | 2016-03-16 | 杨晓富 | Fog and haze detection method based on teller machine |
-
2016
- 2016-10-19 CN CN201610910115.3A patent/CN106485281A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103218622A (en) * | 2013-04-22 | 2013-07-24 | 武汉大学 | Haze monitoring method based on computer vision |
| CN103558133A (en) * | 2013-11-11 | 2014-02-05 | 河南牧业经济学院 | Method and system for monitoring haze concentration based on image analyzing technology |
| CN104517126A (en) * | 2014-12-26 | 2015-04-15 | 北京邮电大学 | Air quality assessment method based on image analysis |
| CN104749078A (en) * | 2015-04-13 | 2015-07-01 | 武汉大学 | PM2.5 concentration detector |
| CN104809735A (en) * | 2015-05-12 | 2015-07-29 | 公安部第三研究所 | System and method for realizing image fog-haze evaluation based on Fourier transformation |
| CN105403490A (en) * | 2015-05-22 | 2016-03-16 | 杨晓富 | Fog and haze detection method based on teller machine |
Non-Patent Citations (5)
| Title |
|---|
| H.J.LIU 等: "Aerosol hygroscopicity derived from size-segregated chemical composition and its parameterization in the North China Plain", 《AGU FALL MEETING》 * |
| 周理 等: "基于人眼视觉机理的雾霾图像质量提升算法", 《中南大学学报（自然科学版）》 * |
| 范相阁 等: "图像识别在环境空气质量监测中的应用", 《河南科学》 * |
| 马楠 等: "基于实测PM2.5、能见度和相对湿度分辨雾霾的新方法", 《中国科学：地球科学》 * |
| 鹿丽鹏 等: "基于图像灰度差分统计的雾霾污染等级检测方法", 《计算机工程》 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108052980A (en) * | 2017-12-21 | 2018-05-18 | 南京理工大学 | Air quality grade detection method based on image |
| CN108052980B (en) * | 2017-12-21 | 2021-05-14 | 南京理工大学 | Image-based air quality grade detection method |
| CN110879276A (en) * | 2019-10-31 | 2020-03-13 | 青岛崂应海纳光电环保集团有限公司 | Filtering method for gas analysis instrument |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103218816B (en) | A kind of crowd density estimation method and people flow rate statistical method based on video analysis | |
| Palmer et al. | A performance measure for boundary detection algorithms | |
| CN103632158B (en) | Forest fire prevention monitor method and forest fire prevention monitor system | |
| CN109190473A (en) | The application of a kind of " machine vision understanding " in remote monitoriong of electric power | |
| CN101431664A (en) | Automatic detection method and system for intensity of passenger flow based on video image | |
| CN105976383A (en) | Power transmission equipment fault diagnosis method based on limit learning machine image recognition | |
| CN106530281A (en) | Edge feature-based unmanned aerial vehicle image blur judgment method and system | |
| CN101877127A (en) | Image reference-free quality evaluation method and system based on gradient profile | |
| CN104853151A (en) | Large-space fire monitoring system based on video image | |
| CN109087363B (en) | HSV color space-based sewage discharge detection method | |
| CN106483143A (en) | A kind of solar energy photovoltaic panel dust stratification on-Line Monitor Device and its detection method | |
| Tao et al. | Automatic smoky vehicle detection from traffic surveillance video based on vehicle rear detection and multi‐feature fusion | |
| CN106485281A (en) | A kind of detection method of the Air haze class of pollution | |
| CN105787912A (en) | Classification-based step type edge sub pixel localization method | |
| CN106529455B (en) | A kind of quick human posture recognition method based on SoC FPGA | |
| CN116152697A (en) | Three-dimensional model measuring method and related device for concrete structure cracks | |
| CN117576632B (en) | Multi-mode AI large model-based power grid monitoring fire early warning system and method | |
| CN112289022B (en) | Black smoke vehicle detection and judgment method and system based on space-time background comparison | |
| CN109815784A (en) | A kind of intelligent method for classifying based on thermal infrared imager, system and storage medium | |
| CN103996266A (en) | Tailing reservoir safety monitoring device based on video analysis | |
| CN104123569B (en) | Video person number information statistics method based on supervised learning | |
| CN111914743A (en) | Method and device for detecting safety helmet of transformer substation worker | |
| CN102682291B (en) | A kind of scene demographic method, device and system | |
| Cheng et al. | Automated real-time pavement distress analysis | |
| CN103278505B (en) | Blast furnace fly ash constituent analysis method based on multi-feature analysis |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20170308 |