CN114320469A - Cloud-edge intelligence-based underground hazard source detection method - Google Patents
Cloud-edge intelligence-based underground hazard source detection method Download PDFInfo
- Publication number
- CN114320469A CN114320469A CN202111613806.4A CN202111613806A CN114320469A CN 114320469 A CN114320469 A CN 114320469A CN 202111613806 A CN202111613806 A CN 202111613806A CN 114320469 A CN114320469 A CN 114320469A
- Authority
- CN
- China
- Prior art keywords
- sensor
- network
- edge
- camera
- cloud
- 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
Images
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Landscapes
- Alarm Systems (AREA)
Abstract
The invention discloses an underground hazard source detection method based on cloud-edge intelligence, which comprises the following steps: step 1, cloud-side parameter optimization: selecting an SSD network model as a detection model, replacing a VGG-16 network for extracting the primary characteristics of a target with a MobileNet network, and reducing the network depth of the characteristic part of the model extraction depth by using a MobilNet-SSD target detection model consisting of the SSD network model and the MobilNet network, turning and cutting a data set picture, removing redundant objects and partially amplifying the target; step 2, establishing a cloud-edge detection system of the coal mine: the camera and the sensor are deployed on the edge equipment, real-time detection is realized on the edge side, and the sensor comprises a temperature sensor, a smoke sensor and a CO sensor; step 3, sensing the danger source by fusing edge side multiple sensors: performing dynamic weighting fusion processing on the information detected by the camera and the sensor in the step 2, and judging whether flame is generated; step 4, multi-network convergence network: and (3) the information detected by the multiple devices in the steps 2 and 3 is wirelessly transmitted to an upper computer in a converged manner.
Description
Technical Field
The invention relates to the technical field of mine safety monitoring, in particular to an underground hazard source detection method based on cloud-edge intelligence.
Background
Mine fire is one of the major prevention and control disasters in the field of coal mine safety, in recent years, underground fire accidents are high and constant, property loss is brought, life safety of workers is threatened, coal mine safety accidents are divided into 8 categories of gas, roofs, transportation and the like, and from accident statistical analysis in 2017, the average number of deaths of fire/gas accidents at each time is up to 21/5.
Therefore, effective monitoring measures are taken, the fire prevention is very important, and at the present stage, the method for monitoring the external cause fire in the underground coal mine mainly adopts sensors such as gas, temperature, smoke and the like, and a large number of sensors are deployed above a mine tunnel to monitor the operation condition of equipment. However, this method has many disadvantages, firstly, the deployment of a large number of sensors is not only time-consuming, labor-consuming and expensive, but also needs to be regularly checked and maintained at a later stage, when the deployed sensors are few, the fire source far away from the sensors cannot be detected in time, which is the most important problem facing the monitoring of the fire under the mine by using sensing currently.
The development of Artificial Intelligence (AI) brings a new idea for the design of a safety monitoring system, and the AI-based safety monitoring system mainly has two deployment methods: the cloud-based method is characterized in that a large amount of collected data are transmitted to the cloud through the Internet, a series of data processing is carried out in the cloud, if the data are attacked, privacy is leaked, and in addition, as the data volume collected by terminal equipment is increased continuously, the cloud-based method brings huge data transmission pressure to a network and cannot meet the real-time requirement of a safety monitoring system; the security monitoring system based on the edge architecture can process data on the edge device without uploading the data to the cloud by using edge computing, so that network attacks can be effectively resisted, privacy leakage of the data can be prevented, and the real-time requirement of the system can be responded, however, the computing resource of the edge device is limited, and when the computing intensive monitoring tasks (such as video processing and image recognition) are processed, the performance bottleneck is usually met.
Disclosure of Invention
Aiming at the problems in the prior art, the invention provides a cloud-edge intelligent underground hazard source detection method.
In order to achieve the purpose, the invention adopts the following technical scheme:
a cloud-edge intelligence-based underground hazard source detection method comprises the following steps:
step 1, cloud-edge parameter optimization
Selecting an SSD network model as a detection model, replacing a VGG-16 network for extracting the primary characteristics of a target with a MobileNet network, and reducing the network depth of the characteristic part of the model extraction depth by using a MobilNet-SSD target detection model consisting of the SSD network model and the MobilNet network, turning and cutting a data set picture, removing redundant objects and partially amplifying the target;
step 2, establishing a cloud-edge detection system of the coal mine
Arranging a camera and a multi-sensor on edge equipment, and realizing real-time detection on an edge side, wherein the multi-sensor comprises a temperature sensor, a smoke sensor and a CO sensor;
step 3, sensing the danger source by fusing edge side multiple sensors:
performing dynamic weighting fusion processing on the information detected by the cameras and the multiple sensors in the step 2, and judging whether flame is generated;
step 4, multi-network convergence network
And (3) the information detected by the multiple devices in the steps 2 and 3 is wirelessly transmitted to an upper computer in a converged manner.
Further, the step 2 is specifically implemented according to the following steps:
s21, training the MobileNet-SSD target detection model in the step 1, and transplanting the trained model on an edge camera;
s22, the video processing module calls a camera to acquire a video frame, and then transmits the acquired picture to the image analysis and detection module;
s23, transmitting the picture acquired in the S22 to a dynamic random access memory of a Cambricon 1H8 edge smart device, calling a neural network model embedded in the Cambricon 1H8 edge smart device by a program to process the picture and identify fireworks in the picture, and then transmitting the positions of the fireworks to an OSD detection target superposition module;
s24, recognizing the image every 0.5 second by the program, and only recognizing the fireworks on three continuous pictures to judge that flame or smoke is generated currently;
and S25, detecting by the aid of the temperature sensor, the smoke sensor and the CO sensor, and detecting abnormal conditions in real time.
Further, the step 3 is specifically implemented according to the following steps:
s31, defining a dynamic weighted decision algorithm, as follows:
F＝[a1，a2，a3，a4][b1，b2，b3，b4]T
wherein a is1、a2、a3、a4Respectively detected by a camera, a temperature sensor, a smoke sensor and a CO sensor, a1、a2、a3、a4Is 1 or 0, wherein when a1When the value is 0, the camera does not detect flame, and when a1When the flame is 1, the camera detects the flame; when a is2When the value is 0, the downhole temperature is in the normal working range, and when a21 indicates that the downhole temperature has exceeded normal; when a is3When the value is 0, the underground smoke is not generated or the smoke concentration is ignored, and when the value is a3When the smoke concentration is 1, the smoke concentration in the well is abnormal; when a is4When 0 indicates that the concentration of CO downhole is within the normal range, when a4When the value is 1, the concentration of the CO in the well is abnormal; b1、b2、b3、b4For the experimentally derived weight parameters, b1、b2、b3、b4The sum of the values of (a) and (b) is 1;
s32, setting parameter b when the camera detects that the flame exists1、b2、b3、b4Respectively 0.75, 0.12, 0.05, 0.08, a threshold value F is setθIs 0.80;
s33, multiplying the results detected in the step 2 by the weight respectively to obtain a value F and a threshold value FeComparing the sizes, when F is more than or equal to FθWhen F < F, indicating that a hazard source fire is detectedθWhen, it indicates that no source of danger is detected;
s34, setting parameter b when the camera does not detect flame but the sensor detects abnormal condition2、b3、b4Respectively 0.4, 0.2, 0.4, a threshold value F is seteIs 0.6;
s35, multiplying the results detected in the step 2 by the weight respectively to obtain a value F and a threshold value FeComparing the sizes, when F is more than or equal to FeWhen F < F, indicating that a hazard source fire is detectedeWhen, it indicates that no source of danger is detected;
and S36, judging whether to send out a danger alarm according to the judgment result obtained in the step S33 or the step S35.
Further, the step 4 is specifically implemented according to the following steps:
s41, a ZigBee network is used, and a plurality of sensor terminal nodes, camera terminal nodes and communication nodes of the inspection robot are wirelessly connected to a ZigBee coordinator;
s42, wirelessly connecting the ZigBee coordinator in the S41 to an embedded intelligent gateway, wherein the embedded intelligent gateway is connected with a router;
and S43, the upper computer receives the information detected by the sensor and the camera and the patrol inspection fed back by the patrol inspection robot through WiFi generated by the router.
The invention has the beneficial effects that:
(1) according to the cloud-edge intelligent-based underground hazard source detection method, the detection method is deployed at the edge side, and the target is directly detected at the edge side, so that the step of uploading data to the cloud is omitted, a hazard source can be quickly detected compared with the traditional detection method, and the accident rate is greatly reduced.
(2) The invention combines the edge intelligent technology, compared with the cloud computing, the edge computing has the excellent performances of reducing communication delay, lightening transmission load and preventing privacy leakage of users.
(3) The invention also combines the technologies of robots and mine internet of things, and really realizes underground intelligent detection of dangerous sources.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings needed to be used in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings without inventive exercise.
FIG. 1 is a flow chart of a cloud-edge intelligence based downhole hazard detection method of the present invention;
FIG. 2 is a flow chart of a dynamic weighting decision module in the cloud-edge intelligence-based downhole hazard detection method of the present invention;
FIG. 3 is a flow chart of a weighting judgment module for camera missing detection conditions in the cloud-edge intelligence based underground hazard source detection method of the present invention;
FIG. 4 is a flow chart of a multi-network fusion network in the cloud-edge intelligence-based downhole hazard detection method of the present invention.
Detailed Description
The invention is illustrated by the following specific examples, which are not intended to be limiting.
Aiming at the problems that the computing resources of edge equipment are limited, and the performance bottleneck is usually met when the computing-intensive monitoring tasks such as video processing and image recognition are processed, the method optimizes the parameters of the detection model, aims at the problem of shooting angle transformation in the running process of the camera, and simultaneously enhances the trained data set, thereby improving the generalization capability of the model.
Referring to fig. 1-4, the invention relates to a cloud-edge intelligence-based downhole hazard source detection method, which comprises the following steps:
step 1, cloud-edge parameter optimization
Selecting an SSD network model as a detection model, replacing a VGG-16 network for extracting the primary characteristics of a target with a MobileNet network, and reducing the network depth of the characteristic part of the model extraction depth by using a MobilNet-SSD target detection model consisting of the SSD network model and the MobilNet network, turning and cutting a data set picture, removing redundant objects and partially amplifying the target;
step 2, establishing a cloud-edge detection system of the coal mine
The coal mine cloud-edge system comprises an inspection track system, an inspection robot, a camera and a sensor;
for the design of the inspection track system, according to the actual condition in the pit, an annular inspection track is arranged above a roadway, the track adopts a multi-freedom-degree track system consisting of a flexible guide rail and a rigid guide rail, and the inspection robot is hung on the track to travel.
The robot is mainly designed to comprise a power module, a wireless communication module, a control module, a driving module and a position acquisition module; the power module comprises a lithium ion power battery pack and a power supply protection board, different power supplies are provided for the whole inspection robot according to the explosion-proof design requirement of the inspection robot, the power module comprises an intrinsic safety power supply and an explosion-proof power supply, the control module is used as an information traffic junction of the robot and is responsible for allocation, use and control of other functional modules of the system, the control module uses an STM32 single chip microcomputer as a robot control core device, and functions of walking and data acquisition, data transmission and the like of the inspection robot are achieved.
According to the intelligent inspection robot, the inspection robot is installed on an inspection track system, the camera and the sensor are deployed on the intelligent inspection robot, underground safety conditions are inspected back and forth through the inspection robot, collected data are wirelessly transmitted to the terminal, and when an onboard detection system detects abnormal conditions, the robot can feed back the inspection position to the terminal.
Step 2 is carried out in the following specific steps
Arranging a camera and a multi-sensor on edge equipment, and realizing real-time detection on an edge side, wherein the multi-sensor comprises a temperature sensor, a smoke sensor and a CO sensor;
step 3, dangerous source perception of edge side multi-sensor fusion
Performing dynamic weighting fusion processing on the information detected by the cameras and the multiple sensors in the step 2, and judging whether flame is generated;
step 4, multi-network convergence network
And (4) the information detected by the multiple devices in the step (2) and the step (3) is fused and wirelessly transmitted to an upper computer (host computer).
Further, the step 2 is specifically implemented according to the following steps:
s21, training the MobileNet-SSD target detection model in the step 1, and transplanting the trained model on an edge camera;
s22, the video processing module calls a camera to acquire a video frame, and then transmits the acquired picture to the image analysis and detection module;
s23, transmitting the picture acquired in the S22 to a dynamic random access memory of a Cambricon 1H8 edge smart device, calling a neural network model embedded in the Cambricon 1H8 edge smart device by a program to process the picture and identify fireworks in the picture, and then transmitting the positions of the fireworks to an OSD detection target superposition module;
s24, recognizing the image every 0.5 second by the program, and only recognizing the fireworks on three continuous pictures to judge that flame or smoke is generated currently;
and S25, detecting by the aid of the temperature sensor, the smoke sensor and the CO sensor, and detecting abnormal conditions in real time.
Further, the step 3 is specifically implemented according to the following steps:
s31, defining a dynamic weighted decision algorithm, as follows:
F＝[a1，a2，a3，a4][b1，b2，b3，b4]T
wherein a is1、a2、a3、a4Respectively detected by a camera, a temperature sensor, a smoke sensor and a CO sensor, a1、a2、a3、a4Is 1 or 0, wherein when a1When the value is 0, the camera does not detect flame, and when a1When the flame is 1, the camera detects the flame; when a is2When the value is 0, the downhole temperature is in the normal working range, and when a21 indicates that the downhole temperature has exceeded normal; when a is3When the value is 0, the underground smoke is not generated or the smoke concentration is ignored, and when the value is a3When the smoke concentration is 1, the smoke concentration in the well is abnormal; when a is4When 0 indicates that the concentration of CO downhole is within the normal range, when a4When the value is 1, the concentration of the CO in the well is abnormal; b1、b2、b3、b4For the experimentally derived weight parameters, b1、b2、b3、b4The sum of the values of (a) and (b) is 1;
s32, setting parameter b when the camera detects that the flame exists1、b2、b3、b4Respectively 0.75, 0.12, 0.05, 0.08, a threshold value F is seteIs 0.80;
s33, multiplying the results detected in the step 2 by the weight respectively to obtain a value F and a threshold value FeComparing the sizes, when F is more than or equal to FeWhen F < F, indicating that a hazard source fire is detectedsWhen, it indicates that no source of danger is detected;
s34, setting parameter b when the camera does not detect flame but the sensor detects abnormal condition2、b3、b4Respectively 0.4, 0.2, 0.4, a threshold value F is seteIs 0.6;
s35, multiplying the results detected in the step 2 by the weight respectively to obtain a value F and a threshold value FsComparing the sizes, when F is more than or equal to FsWhen F < F, indicating that a hazard source fire is detectedsWhen, it indicates that no source of danger is detected;
and S36, judging whether to send out a danger alarm according to the judgment result obtained in the step S33 or the step S35.
Further, the step 4 is specifically implemented according to the following steps:
s41, a ZigBee network is used, and a plurality of sensor terminal nodes, camera terminal nodes and communication nodes of the inspection robot are wirelessly connected to a ZigBee coordinator;
s42, wirelessly connecting the ZigBee coordinator in the S41 to an embedded intelligent gateway, wherein the embedded intelligent gateway is connected with a router;
and S43, the upper computer receives the information detected by the sensor and the camera and the patrol inspection fed back by the patrol inspection robot through WiFi generated by the router.
The cloud-edge intelligent-based underground hazard source detection method is deployed at the edge side, and the target is directly detected at the edge side, so that the step of uploading data to the cloud is omitted, and compared with the traditional detection method, the underground hazard source detection method can quickly detect a hazard source, so that the accident rate is greatly reduced; in combination with the edge intelligent technology, compared with cloud computing, edge computing shows excellent performance of reducing communication delay, reducing transmission load and preventing privacy leakage of users; meanwhile, the technology of the Internet of things of a robot and a mine is combined, and underground intelligent detection of dangerous sources is really achieved.
Finally, it should be noted that the above embodiments are only used for illustrating and not limiting the technical solutions of the present invention, and although the present invention has been described in detail with reference to the above embodiments, it should be understood by those skilled in the art that modifications or equivalent substitutions can be made to the present invention without departing from the spirit and scope of the present invention, and all modifications or partial substitutions should be covered by the scope of the claims of the present invention.
Claims (4)
1. A cloud-edge intelligence-based underground hazard source detection method is characterized by comprising the following steps:
step 1, cloud-edge parameter optimization
Selecting an SSD network model as a detection model, replacing a VGG-16 network for extracting the primary characteristics of a target with a MobileNet network, and reducing the network depth of the characteristic part of the model extraction depth by using a MobilNet-SSD target detection model consisting of the SSD network model and the MobilNet network, turning and cutting a data set picture, removing redundant objects and partially amplifying the target;
step 2, cloud-edge detection system of the coal mine:
arranging a camera and a multi-sensor on edge equipment, and realizing real-time detection on an edge side, wherein the multi-sensor comprises a temperature sensor, a smoke sensor and a CO sensor;
step 3, dangerous source perception of edge side multi-sensor fusion
Performing dynamic weighting fusion processing on the information detected by the cameras and the multiple sensors in the step 2, and judging whether flame is generated;
step 4, multi-network convergence network
And (3) the information detected by the multiple devices in the steps 2 and 3 is wirelessly transmitted to an upper computer in a converged manner.
2. The cloud-edge intelligence based downhole hazard source detection method according to claim 1, wherein the step 2 is specifically implemented according to the following steps:
s21, training the MobileNet-SSD target detection model in the step 1, and transplanting the trained model on an edge camera;
s22, the video processing module calls a camera to acquire a video frame, and then transmits the acquired picture to the image analysis and detection module;
s23, transmitting the picture acquired in the S22 to a dynamic random access memory of a Cambricon 1H8 edge smart device, calling a neural network model embedded in the Cambricon 1H8 edge smart device by a program to process the picture and identify fireworks in the picture, and then transmitting the positions of the fireworks to an OSD detection target superposition module;
s24, recognizing the image every 0.5 second by the program, and only recognizing the fireworks on three continuous pictures to judge that flame or smoke is generated currently;
and S25, detecting by the aid of the temperature sensor, the smoke sensor and the CO sensor, and detecting abnormal conditions in real time.
3. The cloud-edge intelligence based downhole hazard source detection method according to claim 1, wherein the step 3 is specifically implemented according to the following steps:
s31, defining a dynamic weighted decision algorithm, as follows:
F＝[a1,a2,a3,a4][b1,b2,b3,b4]T
wherein a is1、a2、a3、a4Respectively detected by a camera, a temperature sensor, a smoke sensor and a CO sensor, a1、a2、a3、a4Is 1 or 0, wherein when a1When the value is 0, the camera does not detect flame, and when a1When the flame is 1, the camera detects the flame; when a is2When the value is 0, the downhole temperature is in the normal working range, and when a21 indicates that the downhole temperature has exceeded normal; when a is30 hour tableIndicating no smoke or negligible smoke concentration under the well when a3When the smoke concentration is 1, the smoke concentration in the well is abnormal; when a is4When 0 indicates that the concentration of CO downhole is within the normal range, when a4When the value is 1, the concentration of the CO in the well is abnormal; b1、b2、b3、b4For the experimentally derived weight parameters, b1、b2、b3、b4The sum of the values of (a) and (b) is 1;
s32, setting parameter b when the camera detects that the flame exists1、b2、b3、b4Respectively 0.75, 0.12, 0.05, 0.08, a threshold value F is seteIs 0.80;
s33, multiplying the results detected in the step 2 by the weight respectively to obtain a value F and a threshold value FeComparing the sizes when F>FeWhen a hazard source fire is detected, when F<FeWhen, it indicates that no source of danger is detected;
s34, setting parameter b when the camera does not detect flame but the sensor detects abnormal condition2、b3、b4Respectively 0.4, 0.2, 0.4, a threshold value F is seteIs 0.6;
s35, multiplying the results detected in the step 2 by the weight respectively to obtain a value F and a threshold value FeComparing the sizes, when F is more than or equal to FeWhen a hazard source fire is detected, when F<FeWhen, it indicates that no source of danger is detected;
and S36, judging whether to send out a danger alarm according to the judgment result obtained in the step S33 or the step S35.
4. The cloud-edge intelligence based downhole hazard source detection method according to claim 1, wherein the step 4 is specifically implemented according to the following steps:
s41, a ZigBee network is used, and a plurality of sensor terminal nodes, camera terminal nodes and communication nodes of the inspection robot are wirelessly connected to a ZigBee coordinator;
s42, wirelessly connecting the ZigBee coordinator in the S41 to an embedded intelligent gateway, wherein the embedded intelligent gateway is connected with a router;
and S43, the upper computer receives the information detected by the sensor and the camera and the patrol inspection fed back by the patrol inspection robot through WiFi generated by the router.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111613806.4A CN114320469B (en) | 2021-12-27 | 2021-12-27 | Cloud-edge intelligence-based underground hazard source detection method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111613806.4A CN114320469B (en) | 2021-12-27 | 2021-12-27 | Cloud-edge intelligence-based underground hazard source detection method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114320469A true CN114320469A (en) | 2022-04-12 |
| CN114320469B CN114320469B (en) | 2023-04-07 |
Family
ID=81013422
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111613806.4A Active CN114320469B (en) | 2021-12-27 | 2021-12-27 | Cloud-edge intelligence-based underground hazard source detection method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114320469B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115049988A (en) * | 2022-08-17 | 2022-09-13 | 南方电网数字电网研究院有限公司 | Edge calculation method and device for power distribution network monitoring and prejudging |
| CN117579625A (en) * | 2024-01-17 | 2024-02-20 | 中国矿业大学 | Inspection task pre-distribution method for double prevention mechanism |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090315722A1 (en) * | 2008-06-20 | 2009-12-24 | Billy Hou | Multi-wavelength video image fire detecting system |
| CN102890754A (en) * | 2012-10-31 | 2013-01-23 | 中国科学院自动化研究所 | Danger source monitoring system for mine |
| CN110135564A (en) * | 2019-05-15 | 2019-08-16 | 深圳朴生智能科技有限公司 | Coal mine gas sensing data method for detecting abnormality based on edge calculations |
| CN111024158A (en) * | 2019-12-23 | 2020-04-17 | 广东工业大学 | Kitchen electrical hazard intelligent monitoring method combined with edge calculation |
| CN111739254A (en) * | 2020-07-01 | 2020-10-02 | 安徽理工大学 | Coal mine underground network edge video real-time processing joint control system |
| CN111899452A (en) * | 2020-08-04 | 2020-11-06 | 成都云图睿视科技有限公司 | Forest fire prevention early warning system based on edge calculation |
| CN112801148A (en) * | 2021-01-14 | 2021-05-14 | 西安电子科技大学 | Fire recognition and positioning system and method based on deep learning |
| KR20210063751A (en) * | 2019-11-25 | 2021-06-02 | 주식회사엘디티 | Fire detection system |
-
2021
- 2021-12-27 CN CN202111613806.4A patent/CN114320469B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090315722A1 (en) * | 2008-06-20 | 2009-12-24 | Billy Hou | Multi-wavelength video image fire detecting system |
| CN102890754A (en) * | 2012-10-31 | 2013-01-23 | 中国科学院自动化研究所 | Danger source monitoring system for mine |
| CN110135564A (en) * | 2019-05-15 | 2019-08-16 | 深圳朴生智能科技有限公司 | Coal mine gas sensing data method for detecting abnormality based on edge calculations |
| KR20210063751A (en) * | 2019-11-25 | 2021-06-02 | 주식회사엘디티 | Fire detection system |
| CN111024158A (en) * | 2019-12-23 | 2020-04-17 | 广东工业大学 | Kitchen electrical hazard intelligent monitoring method combined with edge calculation |
| CN111739254A (en) * | 2020-07-01 | 2020-10-02 | 安徽理工大学 | Coal mine underground network edge video real-time processing joint control system |
| CN111899452A (en) * | 2020-08-04 | 2020-11-06 | 成都云图睿视科技有限公司 | Forest fire prevention early warning system based on edge calculation |
| CN112801148A (en) * | 2021-01-14 | 2021-05-14 | 西安电子科技大学 | Fire recognition and positioning system and method based on deep learning |
Non-Patent Citations (1)
| Title |
|---|
| 潘三明;袁明强;: "基于边缘计算的视频监控***及应用" * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115049988A (en) * | 2022-08-17 | 2022-09-13 | 南方电网数字电网研究院有限公司 | Edge calculation method and device for power distribution network monitoring and prejudging |
| CN117579625A (en) * | 2024-01-17 | 2024-02-20 | 中国矿业大学 | Inspection task pre-distribution method for double prevention mechanism |
| CN117579625B (en) * | 2024-01-17 | 2024-04-09 | 中国矿业大学 | Inspection task pre-distribution method for double prevention mechanism |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114320469B (en) | 2023-04-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114320469B (en) | Cloud-edge intelligence-based underground hazard source detection method | |
| US20220355147A1 (en) | Smart fire-extinguishing device and smart fire-extinguishing system including same | |
| CN106050306A (en) | Mine monitoring system | |
| CN210691345U (en) | Cloud intelligence safety inspection auxiliary system based on degree of depth learning | |
| AU2020102181A4 (en) | An intelligent recognition system and method of tunnel structure health based on robot vision recognition | |
| CN109667621A (en) | A kind of Coal Mine Safety Monitoring System | |
| CN112112629A (en) | Safety business management system and method in drilling operation process | |
| CN110136391A (en) | A kind of energy-accumulating power station fire early-warning system and method | |
| CN110135564A (en) | Coal mine gas sensing data method for detecting abnormality based on edge calculations | |
| CN112863100B (en) | Intelligent construction safety monitoring system and method | |
| CN102610050A (en) | Online monitoring system for protecting overhead power transmission lines against frost fire and implementing method thereof | |
| CN105898898A (en) | Information monitoring system based on wireless sensor network | |
| CN105704244A (en) | Mobile communication based forest safety monitoring system | |
| CN103761827A (en) | Fire monitor system | |
| CN110700887A (en) | Coal mine safety production monitoring and early warning system and method | |
| CN214409649U (en) | Coal spontaneous combustion monitoring and predicting system based on infrared radiation characteristics | |
| CN202889652U (en) | Fire hazard grading early warning network system router based on wireless sensor network | |
| CN105701557A (en) | Forest fire prediction and forecasting system | |
| CN209746757U (en) | intelligent visual fire-fighting monitoring system | |
| CN111047812A (en) | Forest fire prevention monitoring device and monitoring method based on wireless sensor network | |
| CN111105582B (en) | Forest fire prevention monitoring method and system, computer equipment and readable storage medium | |
| CN212854397U (en) | Intelligent early warning and fire extinguishing system for high-voltage cable joint | |
| CN116342999A (en) | Open space fire smoke detection and identification method for ultrahigh voltage transformer substation | |
| CN206058456U (en) | It is a kind of for fire identification can monitor in real time image processing system | |
| CN115620480A (en) | Early warning system based on high-altitude intelligent automatic monitoring |
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 |