CN108133178A - A kind of intelligent environment monitoring system and method based on image identification - Google Patents
A kind of intelligent environment monitoring system and method based on image identification Download PDFInfo
- Publication number
- CN108133178A CN108133178A CN201711296952.2A CN201711296952A CN108133178A CN 108133178 A CN108133178 A CN 108133178A CN 201711296952 A CN201711296952 A CN 201711296952A CN 108133178 A CN108133178 A CN 108133178A
- Authority
- CN
- China
- Prior art keywords
- image
- task
- unit
- environment monitoring
- intelligent environment
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
-
- 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
- G06V10/443—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 by matching or filtering
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Image Analysis (AREA)
Abstract
The present invention provides a kind of intelligent environment monitoring system based on image identification, including：Image acquisition units, for acquiring the image in environmental monitoring region；Task dispatch unit distributes for carrying out task to image acquisition units；Image pre-processing unit pre-processes for the image to acquisition, and to treated, image screens according to task；Image analyzing unit, for the image after screening to be identified according to deep neural network；Disposition and emergency unit, are matched for the recognition result according to image analyzing unit with preset emergency plan, are generated matching result and are exported；The present invention can utilize existing image capturing system, image is pre-processed and is screened, special sensor is laid with without large-scale, cost can not only be reduced, it can also realize intelligentized efficient environmental monitoring system, more crypto set and accurately environmental correclation problem can be fed back, strong tool and solution are provided to efforts at environmental protection.
Description
Technical field
The present invention relates to environmental testing more particularly to a kind of intelligent environment monitoring systems and side based on image identification
Method.
Background technology
With economic growth, the resource of nature storage is widely developed and utilizes.Due to densely populated big city
City and the foundation of industrial and mining area, make a large amount of chemical substances enter environment, have been more than the self-purification capacity of the Nature, constantly long-pending in the environment
It is tired, produce the public hazards for jeopardizing human survival.The reason of in order to seek environmental quality variation, it is necessary to first from the property of pollutant,
The analysis in source, content and its distribution starts.But judge the quality of environmental quality, only to the single pollutant short time
Sample analysis be inadequate, it is necessary to have the data for the various marks for representing environmental quality, i.e., various pollutants are in a certain range
Prolonged contamination data, environmental quality could be made and definitely evaluated.
At present, environment is monitored, typically using special sensor (such as particulate matter detector) or manually
Collecting sample is judged (such as artificial selective examination dust from construction sites pollution), these monitoring methods are generally required for disposing a large amount of spy
Different sensor, one side cost is higher, on the other hand often constructional difficulties, and efficiency is very low by way of manually monitoring,
And the experience of operator is limited to, monitored results can be caused not accurate enough due to human factor.Therefore, it is necessary to a kind of new skills
Art means more crypto set and accurately can feed back environmental correclation problem in the efficiency for improving environmental monitoring.
Invention content
In view of the foregoing deficiencies of prior art, the present invention provides a kind of intelligent environment monitoring system based on image identification
System and method, to solve above-mentioned technical problem.
The present invention provides the intelligent environment monitoring system identified based on image, including：
Image acquisition units, for acquiring the image in environmental monitoring region；
Task dispatch unit distributes for carrying out task to image acquisition units；
Image pre-processing unit is connect respectively with image processing unit and task dispatch unit, for the image to acquisition
It is pre-processed, and to treated, image screens according to task；
Image analyzing unit is connect with image pre-processing unit, for according to deep neural network to the image after screening
It is identified；
Disposition and emergency unit, connect with image analyzing unit, for the recognition result according to image analyzing unit and in advance
If emergency plan matched, generation matching result simultaneously export.
Further, the different task that described image analytic unit is distributed according to task dispatch unit passes through different depth
Neural network carries out image identification, and the different deep neural network includes convolutional neural networks disaggregated model and returns mould
Type.
Further, the task includes qualitative detection task and quantitative Detection task, and the qualitative detection task passes through volume
Product neural network classification model is identified, and the quantitative Detection task passes through convolutional neural networks disaggregated model and regression model
Common identification.
Further, server is further included, described image analytic unit is connect with server, and the server is cloud service
Device, described image collecting unit include multiple image acquisition devices being set in environmental monitoring region, and described image collector is
Camera, the task dispatch unit acquire different images, described image pretreatment unit root according to task control camera
The image after screening according to task is classified according to the load of server, and sorted image is sent into cloud server
Different nodes.
Correspondingly, the present invention also provides it is a kind of based on image identification intelligent environment monitoring method, including：
Acquire the image in environmental monitoring region；
Task is carried out to image acquisition units to distribute；
Figure pre-processes the image of acquisition, and to treated, image screens according to task；
The image after screening is identified according to deep neural network；
It is matched according to the recognition result of image analyzing unit with preset emergency plan, generation matching result is simultaneously defeated
Go out.
Further, the different task distributed according to task dispatch unit carries out image by different deep neural networks
Identification, the different deep neural network include convolutional neural networks disaggregated model and regression model, and the task includes fixed
Property Detection task and quantitative Detection task, the qualitative detection task are identified by convolutional neural networks disaggregated model, institute
Quantitative Detection task is stated to identify jointly by convolutional neural networks disaggregated model and regression model.
Further, Screening Treatment is carried out to the output result of deep neural network, the Screening Treatment is included according to task
Characteristics of image difference in result will be exported to come out more than the optical sieving of preset discrepancy threshold, by the image of image classification mistake
Training is iterated by deep neural network, until error sample quantity is less than preset sample threshold.
Further, it for quantitative Detection task, is extracted in image according to task type and meets the regional frame of task, then pass through
Convolutional neural networks carry out quantitative judge to the regional frame, obtain recognition result, the regional frame is by selecting recognition result
Middle image reliability is obtained more than the mode of preset believability threshold.
Further, the qualitative detection task includes Vehicular exhaust detection, emits dress detection and rubbish detection, the quantitative inspection
Survey includes Detection of Air Quality, and the pretreatment includes screening the image of acquisition according to task, removes invalid image.
The present invention also provides a kind of computer readable storage mediums, are stored thereon with computer program, which is handled
Device realizes any one of above-mentioned the method when performing.
Beneficial effects of the present invention：The intelligent environment monitoring system based on image identification in the present invention, can be existing
Image capturing system pre-processes image and is screened, and is identified by deep neural network, without large-scale
Special sensor is laid with, cost can be not only reduced, can also realize intelligentized efficient environmental monitoring system, it can be more
It is intensive and accurately environmental correclation problem is fed back, provide strong tool and solution to efforts at environmental protection.
Description of the drawings
Fig. 1 is the structure diagram of the intelligent environment monitoring system identified in the embodiment of the present invention based on image.
Fig. 2 is the flow diagram of intelligent environment monitoring method identified in the embodiment of the present invention based on image.
Fig. 3 is that the flow of image preprocessing in the intelligent environment monitoring method identified in the embodiment of the present invention based on image is shown
It is intended to.
Specific embodiment
Illustrate embodiments of the present invention below by way of specific specific example, those skilled in the art can be by this specification
Disclosed content understands other advantages and effect of the present invention easily.The present invention can also pass through in addition different specific realities
The mode of applying is embodied or practiced, the various details in this specification can also be based on different viewpoints with application, without departing from
Various modifications or alterations are carried out under the spirit of the present invention.It should be noted that in the absence of conflict, following embodiment and implementation
Feature in example can be combined with each other.
It should be noted that the diagram provided in following embodiment only illustrates the basic structure of the present invention in a schematic way
Think, component count, shape and size when only display is with related component in the present invention rather than according to actual implementation in schema then
It draws, kenel, quantity and the ratio of each component can be a kind of random change during actual implementation, and its assembly layout kenel
It is likely more complexity.
As shown in Figure 1, the intelligent environment monitoring system based on image identification in the present embodiment, including：
Image acquisition units, for acquiring the image in environmental monitoring region；
Task dispatch unit distributes for carrying out task to image acquisition units；
Image pre-processing unit is connect respectively with image processing unit and task dispatch unit, for the image to acquisition
It is pre-processed, and to treated, image screens according to task；
Image analyzing unit is connect with image pre-processing unit, for according to deep neural network to the image after screening
It is identified；
Disposition and emergency unit, connect with image analyzing unit, for the recognition result according to image analyzing unit and in advance
If emergency plan matched, generation matching result simultaneously export.
In the present embodiment, image acquisition units include multiple image acquisition devices being set in environmental monitoring region, institute
Image acquisition device is stated as camera, existing monitoring system may be used in camera, can by being captured to real-time video frame
To be captured according to setup parameter such as time interval, ineligible image is filtered out by image pre-processing unit,
As shown in figure 3, to reduce the load of system, task dispatch unit acquires different images, image according to task control camera
Pretreatment unit classifies the image after screening according to task, and sorted image is sent according to the load of server
The different nodes of cloud server, the filter method in the present embodiment can be filtered according to the monitor task of different scenes,
Such as task dispatch unit distributes when being vehicle exhaust monitor task, is directed to vehicle exhaust monitor task, filters out and do not wrap
Image containing automobile；When it is airborne dust task to distribute, the monitoring for dust from construction sites filters out the image of rainy day, to non-
Face region carries out detection filtering in advance etc. (mainly including trees, building)；When the task that distributes is rubbish Detection task, to non-
Face region carries out detection filtering in advance (mainly including trees, building)；When distributing task to emit dress detection, dilly is carried out
Filtering etc.；Disposition is mainly matched with emergency unit according to the recognition result of image analyzing unit with preset emergency plan,
Generation matching result simultaneously exports, such as can generate relevant report, can be certainly when generation major pollution incident is detected
It is dynamic to alarm, it notifies relevant person liable, testing result can also be directed to, to some foreseeable environmental problems, accomplish to carry
Preceding early warning.
In the present embodiment, the function of task dispatch unit is mainly that a large amount of camera shooting head node of deployment is collected
Image is distributed according to monitor task, and correspondingly, image analyzing unit can be divided into multiple special places according to task type
The subelement of the corresponding personage of reason, such as airborne dust monitor task, system will be sent to server in all image sets when being executed
In corresponding airborne dust monitor task image analysis subelement in, certainly, the camera in the present embodiment can take into account multiple
Business, in this case, task distributes system and image is separately sent to multiple subelements.Server in the present embodiment is high in the clouds
Server, operation and resource allocation for a variety of identification missions.According to server load condition, by different task or phase
Different disposal object with task is assigned to different high in the clouds processing nodes progress calculation process.
In the present embodiment, the different task that image analyzing unit is distributed according to task dispatch unit passes through different depths
It spends neural network and carries out image identification, different deep neural networks includes convolutional neural networks disaggregated model and regression model,
Task include qualitative detection task and quantitative Detection task, the qualitative detection task by convolutional neural networks disaggregated model into
Row identification, the quantitative Detection task is identified jointly by convolutional neural networks disaggregated model and regression model, in the present embodiment
Deep neural network recognition methods be mainly based upon different task using classify or return two kinds of models, for qualitative detection
Task such as white garbage detects, Vehicular exhaust detection, white garbage detection, the method classified using CNN, and defeated for needing
The task such as airborne dust for going out quantitative result detects, then two kinds of models of combining classification and recurrence, are first detected, then return and divide
Analysis.The method of detection is to meet the possible regional frame of particular task by being chosen in image, then carry out grader identification, is identified
The region that confidence level is obtained more than certain threshold value then represents to detect respective classes object.
As shown in Fig. 2, correspondingly the present invention also provides it is a kind of based on image identification intelligent environment monitoring method, including：
Acquire the image in environmental monitoring region；
Task is carried out to image acquisition units to distribute；
Figure pre-processes the image of acquisition, and to treated, image screens according to task；
The image after screening is identified according to deep neural network；
It is matched according to the recognition result of image analyzing unit with preset emergency plan, generation matching result is simultaneously defeated
Go out.
In the present embodiment, the different task distributed according to task dispatch unit, by different deep neural networks into
Row image identifies that the different deep neural network includes convolutional neural networks disaggregated model and regression model, and task includes
Qualitative detection task and quantitative Detection task, qualitative detection task are identified by convolutional neural networks disaggregated model, quantitative
Detection task is identified jointly by convolutional neural networks disaggregated model and regression model, for times of qualitative detection in the present embodiment
Business such as white garbage detects, Vehicular exhaust detection, white garbage detection, the method classified using CNN, and for needing to export
The task of quantitative result such as airborne dust detects, then two kinds of models of combining classification and recurrence, are first detected, then return and divide
Analysis.The method of detection is to meet the possible regional frame of particular task by being chosen in image, then carry out grader identification, is identified
The region that confidence level is obtained more than preset believability threshold then represents to detect respective classes object.
As shown in figure 3, in the present embodiment, Screening Treatment, Screening Treatment are carried out to the output result of deep neural network
It is come out, and figure is carried out to it more than the optical sieving of preset discrepancy threshold including image difference in result will be exported according to task
As signature analysis, image classification is obtained as a result, the image of image classification mistake is iterated instruction by deep neural network again
Practice, until error sample quantity is less than preset sample threshold,
By selecting the sample that image difference is larger in Given task in the present embodiment, this step can pass through characteristics of image
Analysis is automatically performed, and then picks out the result of these image classification mistakes, finally by error result image to depth god
Training is finely adjusted through network, this process iteration always is less than until for given quantity as a result, selecting error sample quantity
Until preset sample threshold.
Computer readable storage medium in the present embodiment, one of ordinary skill in the art will appreciate that：It realizes above-mentioned each
The all or part of step of embodiment of the method can be completed by the relevant hardware of computer program.Aforementioned computer program
It can be stored in a computer readable storage medium.The program when being executed, performs the step for including above-mentioned each method embodiment
Suddenly；And aforementioned storage medium includes：The various media that can store program code such as ROM, RAM, magnetic disc or CD.
In the present embodiment, memory may include random access memory (RandomAccessMemory, abbreviation
RAM), it is also possible to further include nonvolatile memory (non-volatilememory), for example, at least a magnetic disk storage.
Above-mentioned processor can be general processor, including central processing unit (CentralProcessingUnit, letter
Claim CPU), network processing unit (NetworkProcessor, abbreviation NP) etc.；It can also be digital signal processor
(DigitalSignalProcessing, abbreviation DSP), application-specific integrated circuit
(ApplicationSpecificIntegratedCircuit, abbreviation ASIC), field programmable gate array (Field-
ProgrammableGateArray, abbreviation FPGA) either other programmable logic device, discrete gate or transistor logic device
Part, discrete hardware components.
The above-described embodiments merely illustrate the principles and effects of the present invention, and is not intended to limit the present invention.It is any ripe
The personage for knowing this technology all can carry out modifications and changes under the spirit and scope without prejudice to the present invention to above-described embodiment.Cause
This, those of ordinary skill in the art is complete without departing from disclosed spirit and institute under technological thought such as
Into all equivalent modifications or change, should by the present invention claim be covered.
Claims (10)
1. a kind of intelligent environment monitoring system based on image identification, which is characterized in that including：
Image acquisition units, for acquiring the image in environmental monitoring region；
Task dispatch unit distributes for carrying out task to image acquisition units；
Image pre-processing unit is connect respectively with image processing unit and task dispatch unit, for being carried out to the image of acquisition
Pretreatment, and to treated, image screens according to task；
Image analyzing unit is connect with image pre-processing unit, for being carried out according to deep neural network to the image after screening
Identification；
Disposition and emergency unit, connect with image analyzing unit, for the recognition result according to image analyzing unit with it is preset
Emergency plan is matched, and is generated matching result and is exported.
2. the intelligent environment monitoring system according to claim 1 based on image identification, which is characterized in that described image point
The different task that analysis unit is distributed according to task dispatch unit carries out image identification by different deep neural networks, described
Different deep neural networks includes convolutional neural networks disaggregated model and regression model.
3. the intelligent environment monitoring system according to claim 2 based on image identification, which is characterized in that the task packet
Qualitative detection task and quantitative Detection task are included, the qualitative detection task is known by convolutional neural networks disaggregated model
Not, the quantitative Detection task is identified jointly by convolutional neural networks disaggregated model and regression model.
4. the intelligent environment monitoring system according to claim 1 based on image identification, which is characterized in that further include service
Device, described image analytic unit are connect with server, and the server is cloud server, and described image collecting unit includes more
A image acquisition device being set in environmental monitoring region, described image collector be camera, the task dispatch unit root
Different image is acquired according to task control camera, and described image pretreatment unit is according to the load of server by the figure after screening
Picture is classified according to task, and sorted image is sent to the different nodes of cloud server.
A kind of 5. intelligent environment monitoring method based on image identification, which is characterized in that including：
Acquire the image in environmental monitoring region；
Task is carried out to image acquisition units to distribute；
Figure pre-processes the image of acquisition, and to treated, image screens according to task；
The image after screening is identified according to deep neural network；
It is matched, generate matching result and is exported with preset emergency plan according to the recognition result of image analyzing unit.
6. the intelligent environment monitoring method according to claim 5 based on image identification, which is characterized in that sent according to task
The different task that bill member distributes carries out image identification, the different depth nerve net by different deep neural networks
Network includes convolutional neural networks disaggregated model and regression model, and the task includes qualitative detection task and quantitative Detection task,
The qualitative detection task is identified by convolutional neural networks disaggregated model, and the quantitative Detection task passes through convolutional Neural
Network class model and regression model identify jointly.
7. the intelligent environment monitoring method according to claim 6 based on image identification, which is characterized in that depth nerve
The output result of network carries out Screening Treatment, and the Screening Treatment includes according to task that characteristics of image difference in output result is big
It is come out in the optical sieving of preset discrepancy threshold, the image of image classification mistake is iterated instruction by deep neural network
Practice, until error sample quantity is less than preset sample threshold.
8. the intelligent environment monitoring method according to claim 6 based on image identification, which is characterized in that for quantitative inspection
Survey task extracts the regional frame for meeting task in image according to task type, then by convolutional neural networks to the regional frame
Carry out quantitative judge, obtain recognition result, the regional frame by select in recognition result image reliability be more than it is preset can
The mode of confidence threshold obtains.
9. the intelligent environment monitoring method according to claim 5 based on image identification, which is characterized in that the qualitative inspection
Survey task includes Vehicular exhaust detection, emits dress detection and rubbish detection, and the quantitative detection includes Detection of Air Quality, described pre-
Processing includes screening the image of acquisition according to task, removes invalid image.
10. a kind of computer readable storage medium, is stored thereon with computer program, it is characterised in that：The program is by processor
Any one of claim 5 to 9 the method is realized during execution.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711296952.2A CN108133178B (en) | 2017-12-08 | 2017-12-08 | Intelligent environment monitoring system and method based on image recognition |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711296952.2A CN108133178B (en) | 2017-12-08 | 2017-12-08 | Intelligent environment monitoring system and method based on image recognition |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108133178A true CN108133178A (en) | 2018-06-08 |
| CN108133178B CN108133178B (en) | 2021-11-30 |
Family
ID=62389339
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711296952.2A Active CN108133178B (en) | 2017-12-08 | 2017-12-08 | Intelligent environment monitoring system and method based on image recognition |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108133178B (en) |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108994855A (en) * | 2018-08-15 | 2018-12-14 | 深圳市烽焌信息科技有限公司 | Rubbish periodic cleaning method and robot |
| CN109523013A (en) * | 2018-10-15 | 2019-03-26 | 西北大学 | A kind of air particle pollution level estimation method based on shallow-layer convolutional neural networks |
| CN109753890A (en) * | 2018-12-18 | 2019-05-14 | 吉林大学 | A kind of pavement garbage object intelligent recognition and cognitive method and its realization device |
| CN110210568A (en) * | 2019-06-06 | 2019-09-06 | 中国民用航空飞行学院 | The recognition methods of aircraft trailing vortex and system based on convolutional neural networks |
| CN110287939A (en) * | 2018-12-29 | 2019-09-27 | 中国科学院软件研究所 | Space-based intelligent image processing method |
| CN110298347A (en) * | 2019-05-30 | 2019-10-01 | 长安大学 | A kind of recognition methods of the automobile exhaust analyzer screen based on GrayWorld and PCA-CNN |
| CN110321932A (en) * | 2019-06-10 | 2019-10-11 | 浙江大学 | A kind of whole city city air quality index estimation method based on depth multisource data fusion |
| CN110688945A (en) * | 2019-09-26 | 2020-01-14 | 成都睿云物联科技有限公司 | Cleanliness detection method and device, computer equipment and storage medium |
| CN110782653A (en) * | 2018-11-07 | 2020-02-11 | 北京嘀嘀无限科技发展有限公司 | Road information acquisition method and system |
| CN111723772A (en) * | 2020-06-30 | 2020-09-29 | 平安国际智慧城市科技股份有限公司 | Perishable garbage identification method and device based on image identification and computer equipment |
| CN113420631A (en) * | 2021-06-17 | 2021-09-21 | 广联达科技股份有限公司 | Safety alarm method and device based on image recognition |
| CN116883755A (en) * | 2023-07-20 | 2023-10-13 | 广州新城建筑设计院有限公司 | Rural construction environment monitoring method, system, equipment and storage medium |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103345656A (en) * | 2013-07-17 | 2013-10-09 | 中国科学院自动化研究所 | Method and device for data identification based on multitask deep neural network |
| CN103984310A (en) * | 2014-05-12 | 2014-08-13 | 华迪计算机集团有限公司 | Chemical industry park environment pollution detection method and device based on multi-source remote sensing data |
| CN105630882A (en) * | 2015-12-18 | 2016-06-01 | 哈尔滨工业大学深圳研究生院 | Remote sensing data deep learning based offshore pollutant identifying and tracking method |
| CN106327738A (en) * | 2016-08-26 | 2017-01-11 | 特斯联（北京）科技有限公司 | Intelligent grading monitoring system |
| CN106919065A (en) * | 2015-12-24 | 2017-07-04 | 天津市军联科技有限公司 | Construction site remote real-time monitoring and airborne dust noise monitoring system based on Internet of Things |
| CN106951889A (en) * | 2017-05-23 | 2017-07-14 | 煤炭科学技术研究院有限公司 | Underground high risk zone moving target monitoring and management system |
| US20170337467A1 (en) * | 2016-05-18 | 2017-11-23 | Nec Laboratories America, Inc. | Security system using a convolutional neural network with pruned filters |
-
2017
- 2017-12-08 CN CN201711296952.2A patent/CN108133178B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103345656A (en) * | 2013-07-17 | 2013-10-09 | 中国科学院自动化研究所 | Method and device for data identification based on multitask deep neural network |
| CN103984310A (en) * | 2014-05-12 | 2014-08-13 | 华迪计算机集团有限公司 | Chemical industry park environment pollution detection method and device based on multi-source remote sensing data |
| CN105630882A (en) * | 2015-12-18 | 2016-06-01 | 哈尔滨工业大学深圳研究生院 | Remote sensing data deep learning based offshore pollutant identifying and tracking method |
| CN106919065A (en) * | 2015-12-24 | 2017-07-04 | 天津市军联科技有限公司 | Construction site remote real-time monitoring and airborne dust noise monitoring system based on Internet of Things |
| US20170337467A1 (en) * | 2016-05-18 | 2017-11-23 | Nec Laboratories America, Inc. | Security system using a convolutional neural network with pruned filters |
| CN106327738A (en) * | 2016-08-26 | 2017-01-11 | 特斯联（北京）科技有限公司 | Intelligent grading monitoring system |
| CN106951889A (en) * | 2017-05-23 | 2017-07-14 | 煤炭科学技术研究院有限公司 | Underground high risk zone moving target monitoring and management system |
Cited By (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108994855A (en) * | 2018-08-15 | 2018-12-14 | 深圳市烽焌信息科技有限公司 | Rubbish periodic cleaning method and robot |
| CN109523013A (en) * | 2018-10-15 | 2019-03-26 | 西北大学 | A kind of air particle pollution level estimation method based on shallow-layer convolutional neural networks |
| CN110782653A (en) * | 2018-11-07 | 2020-02-11 | 北京嘀嘀无限科技发展有限公司 | Road information acquisition method and system |
| CN109753890B (en) * | 2018-12-18 | 2022-09-23 | 吉林大学 | Intelligent recognition and sensing method for road surface garbage and implementation device thereof |
| CN109753890A (en) * | 2018-12-18 | 2019-05-14 | 吉林大学 | A kind of pavement garbage object intelligent recognition and cognitive method and its realization device |
| CN110287939A (en) * | 2018-12-29 | 2019-09-27 | 中国科学院软件研究所 | Space-based intelligent image processing method |
| CN110287939B (en) * | 2018-12-29 | 2024-01-05 | 中国科学院软件研究所 | Space-based intelligent image processing method |
| CN110298347A (en) * | 2019-05-30 | 2019-10-01 | 长安大学 | A kind of recognition methods of the automobile exhaust analyzer screen based on GrayWorld and PCA-CNN |
| CN110298347B (en) * | 2019-05-30 | 2022-11-01 | 长安大学 | Method for identifying automobile exhaust analyzer screen based on GrayWorld and PCA-CNN |
| CN110210568A (en) * | 2019-06-06 | 2019-09-06 | 中国民用航空飞行学院 | The recognition methods of aircraft trailing vortex and system based on convolutional neural networks |
| CN110321932B (en) * | 2019-06-10 | 2021-06-25 | 浙江大学 | Full-city air quality index estimation method based on deep multi-source data fusion |
| CN110321932A (en) * | 2019-06-10 | 2019-10-11 | 浙江大学 | A kind of whole city city air quality index estimation method based on depth multisource data fusion |
| CN110688945A (en) * | 2019-09-26 | 2020-01-14 | 成都睿云物联科技有限公司 | Cleanliness detection method and device, computer equipment and storage medium |
| CN111723772A (en) * | 2020-06-30 | 2020-09-29 | 平安国际智慧城市科技股份有限公司 | Perishable garbage identification method and device based on image identification and computer equipment |
| CN111723772B (en) * | 2020-06-30 | 2024-03-19 | 平安国际智慧城市科技股份有限公司 | Perishable garbage identification method and device based on image identification and computer equipment |
| CN113420631A (en) * | 2021-06-17 | 2021-09-21 | 广联达科技股份有限公司 | Safety alarm method and device based on image recognition |
| CN116883755A (en) * | 2023-07-20 | 2023-10-13 | 广州新城建筑设计院有限公司 | Rural construction environment monitoring method, system, equipment and storage medium |
| CN116883755B (en) * | 2023-07-20 | 2024-03-26 | 广州新城建筑设计院有限公司 | Rural construction environment monitoring method, system, equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108133178B (en) | 2021-11-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108133178A (en) | A kind of intelligent environment monitoring system and method based on image identification | |
| CN111429726B (en) | Monitoring video illegal parking vehicle detection and management method and corresponding system | |
| CN102239402B (en) | Non-linear histogram segmentation for particle analysis | |
| CN109639481A (en) | A kind of net flow assorted method, system and electronic equipment based on deep learning | |
| Bu et al. | Crack detection using a texture analysis-based technique for visual bridge inspection | |
| CN102915432B (en) | A kind of vehicle-mounted microcomputer image/video data extraction method and device | |
| KR20150011329A (en) | Closed-loop automatic defect inspection and classification | |
| CN112871703A (en) | Intelligent management coal preparation platform and method thereof | |
| CN111223129A (en) | Detection method, detection device, monitoring equipment and computer readable storage medium | |
| CN113642474A (en) | Hazardous area personnel monitoring method based on YOLOV5 | |
| CN109815945A (en) | A kind of respiratory tract inspection result interpreting system and method based on image recognition | |
| EP4044066A1 (en) | Network of intelligent machines | |
| CN108230669B (en) | Road vehicle violation detection method and system based on big data and cloud analysis | |
| KR102403170B1 (en) | Apparatus for monitoring image employing to detect of vehicle number and cotrolling device | |
| CN106781167B (en) | Method and device for monitoring motion state of object | |
| CN115269438A (en) | Automatic testing method and device for image processing algorithm | |
| CN111801783B (en) | Jammer mining for novel defect discovery | |
| CN202815869U (en) | Vehicle microcomputer image and video data extraction apparatus | |
| CN110263622A (en) | Train fire monitoring method, apparatus, terminal and storage medium | |
| CN117274245A (en) | AOI optical detection method and system based on image processing technology | |
| Shahrabadi et al. | Benchmarking deep learning models and hyperparameters for bridge defects classification | |
| CN112418362A (en) | Target detection training sample screening method | |
| CN109977783B (en) | Method for detecting whether baby carriage takes advantage of ladder independently based on vertical ladder scene | |
| CN109544855B (en) | Computer vision-based closed circuit television system for comprehensively monitoring fire disaster in rail transit and implementation method | |
| CN112108263A (en) | Fluid suspended matter classification system and method based on vortex filter and 3D U-Net network |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |