CN116994396A - Intelligent AI alarm system and method based on Internet of things - Google Patents

Intelligent AI alarm system and method based on Internet of things Download PDF

Info

Publication number
CN116994396A
CN116994396A CN202311243457.0A CN202311243457A CN116994396A CN 116994396 A CN116994396 A CN 116994396A CN 202311243457 A CN202311243457 A CN 202311243457A CN 116994396 A CN116994396 A CN 116994396A
Authority
CN
China
Prior art keywords
fire
module
data
video
intelligent
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
Application number
CN202311243457.0A
Other languages
Chinese (zh)
Inventor
洪明权
刘培
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shenzhen Qianhai Aoruina Safety Technology Co ltd
Original Assignee
Shenzhen Qianhai Aoruina Safety Technology Co ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Shenzhen Qianhai Aoruina Safety Technology Co ltd filed Critical Shenzhen Qianhai Aoruina Safety Technology Co ltd
Priority to CN202311243457.0A priority Critical patent/CN116994396A/en
Publication of CN116994396A publication Critical patent/CN116994396A/en
Pending legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/40—Scenes; Scene-specific elements in video content
    • G06V20/41—Higher-level, semantic clustering, classification or understanding of video scenes, e.g. detection, labelling or Markovian modelling of sport events or news items
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/0464—Convolutional networks [CNN, ConvNet]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
    • G06V10/764—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video 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
    • G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
    • G06V10/82—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/40—Scenes; Scene-specific elements in video content
    • G06V20/44—Event detection
    • G—PHYSICS
    • G08—SIGNALLING
    • G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
    • G08B17/00—Fire alarms; Alarms responsive to explosion
    • G08B17/10—Actuation by presence of smoke or gases, e.g. automatic alarm devices for analysing flowing fluid materials by the use of optical means
    • G—PHYSICS
    • G08—SIGNALLING
    • G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
    • G08B17/00—Fire alarms; Alarms responsive to explosion
    • G08B17/12—Actuation by presence of radiation or particles, e.g. of infrared radiation or of ions
    • G08B17/125—Actuation by presence of radiation or particles, e.g. of infrared radiation or of ions by using a video camera to detect fire or smoke
    • G—PHYSICS
    • G08—SIGNALLING
    • G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
    • G08B25/00—Alarm systems in which the location of the alarm condition is signalled to a central station, e.g. fire or police telegraphic systems
    • G—PHYSICS
    • G08—SIGNALLING
    • G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
    • G08B29/00—Checking or monitoring of signalling or alarm systems; Prevention or correction of operating errors, e.g. preventing unauthorised operation
    • G08B29/18—Prevention or correction of operating errors
    • G—PHYSICS
    • G08—SIGNALLING
    • G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
    • G08B29/00—Checking or monitoring of signalling or alarm systems; Prevention or correction of operating errors, e.g. preventing unauthorised operation
    • G08B29/18—Prevention or correction of operating errors
    • G08B29/185—Signal analysis techniques for reducing or preventing false alarms or for enhancing the reliability of the system
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16Y—INFORMATION AND COMMUNICATION TECHNOLOGY SPECIALLY ADAPTED FOR THE INTERNET OF THINGS [IoT]
    • G16Y40/00—IoT characterised by the purpose of the information processing
    • G16Y40/10—Detection; Monitoring
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16Y—INFORMATION AND COMMUNICATION TECHNOLOGY SPECIALLY ADAPTED FOR THE INTERNET OF THINGS [IoT]
    • G16Y40/00—IoT characterised by the purpose of the information processing
    • G16Y40/20—Analytics; Diagnosis

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Computing Systems (AREA)
  • Software Systems (AREA)
  • Multimedia (AREA)
  • Health & Medical Sciences (AREA)
  • Evolutionary Computation (AREA)
  • General Health & Medical Sciences (AREA)
  • Artificial Intelligence (AREA)
  • Biomedical Technology (AREA)
  • Emergency Management (AREA)
  • Business, Economics & Management (AREA)
  • Computational Linguistics (AREA)
  • Medical Informatics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Biophysics (AREA)
  • Data Mining & Analysis (AREA)
  • Molecular Biology (AREA)
  • General Engineering & Computer Science (AREA)
  • Mathematical Physics (AREA)
  • Databases & Information Systems (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Computer Security & Cryptography (AREA)
  • Chemical & Material Sciences (AREA)
  • Analytical Chemistry (AREA)
  • Alarm Systems (AREA)

Abstract

The application discloses an intelligent AI alarm system and method based on the Internet of things, comprising a data acquisition unit, a data transmission unit, an identification judging unit, a data verification unit, a coping processing unit and a result pushing unit, wherein video smoke feeling data of a fire scene are acquired and uploaded to a fire video intelligent identification device; transmitting video smoke sensing data acquired at a fire scene to a fire video intelligent identification device; the video smoke sensing data is identified, judged and analyzed through an AI algorithm model so as to judge the severity of the fire disaster in the fire scene; verifying fire scene judgment to prevent false alarm of fire situations; according to the verified judgment result, selecting a corresponding coping strategy to carry out intelligent coping processing; the fire disaster judgment result is uploaded to the cloud and pushed to the attendees, and the fire disaster on-site condition can be rapidly identified and judged through the flow, so that the probability of fire disaster false alarm is reduced, and the response speed is high.

Description

Intelligent AI alarm system and method based on Internet of things
Technical Field
The application belongs to the technical field of fire alarm systems, and particularly relates to an intelligent AI alarm system and method based on the Internet of things.
Background
Fire alarm systems are one of the most important means to prevent fires. Fire alarms in the prior art are mostly manually controlled. For example, a button is arranged on the fire alarm, and the alarm can be realized by pressing the button when a dangerous situation occurs, wherein the fire control center is a central point of the fire alarm system.
The Chinese patent with the authority (bulletin) number of CN114792465B discloses a fire safety monitoring system based on a regional alarm model, which comprises a fire alarm device for monitoring fire conditions and sending fire information, wherein the fire safety monitoring device receives the fire information and makes delay judgment. According to the application, after the fire alarm equipment gives out an alarm, the fire information of the fire alarm equipment is received through the safety monitoring equipment and is continuously analyzed, if the alarm duration time in the fire information is within five seconds, the false alarm information is judged, and if the alarm duration time is out of five seconds, the fire information is judged, and the alarm is triggered to give out an alarm after the fire information is judged, so that the condition that the false alarm causes interference to fire can be reduced.
The fire alarm technology related to the prior art and the above-mentioned published patent can not realize effectively solving the problem of false alarm in the fire scene, and meanwhile, the unattended personnel is required to be on duty uninterruptedly, and the unattended personnel can not carry out alarm when the unattended personnel is on duty. Therefore, the intelligent AI alarm system and the intelligent AI alarm method based on the Internet of things solve the problems in the prior art, so that the intelligent AI alarm system and the intelligent AI alarm method can quickly identify and judge the condition of a fire scene, reduce the probability of false alarm of the fire and have higher response speed.
Disclosure of Invention
The application aims to provide an intelligent AI alarm system and method based on the Internet of things, which are used for solving the problems that the prior art cannot effectively solve the problem of false alarm in a fire scene, and meanwhile, a person on duty is required to be on duty uninterruptedly, and the person on duty cannot carry out alarm confirmation.
In order to achieve the above purpose, the application adopts the following technical scheme:
an intelligent AI alarm system based on the internet of things, comprising:
the data acquisition unit is used for acquiring video smoke sensing data of the fire scene and uploading the video smoke sensing data to the fire video intelligent recognition device;
the data transmission unit is used for transmitting the video smoke sensing data acquired at the fire scene to the fire video intelligent identification device;
the identification judging unit is used for carrying out identification judgment and analysis on the video smoke sensing data through the AI algorithm model so as to judge the severity of the fire disaster in the fire scene;
the data verification unit is used for verifying the fire scene judgment so as to prevent false alarm of the fire situation;
the coping processing unit is used for selecting a corresponding coping strategy to carry out intelligent coping processing according to the verified judging result;
a result pushing unit; and the fire judgment result is uploaded to the cloud and pushed to the attendees.
Preferably, the fire disaster video intelligent recognition device comprises a JETSON module, a PLC module, a 4G module and a WIFI module, wherein:
the JETSON module is connected with the PLC module through an Ethernet, and is respectively connected with the 4G module and the WIFI module through a USB interface;
the JETSON module is hardware for realizing an AI algorithm model;
the PLC module is used for converting the PLC signals into Ethernet data;
the 4G module is used for realizing the functions of 4G internet surfing, short messages and telephones;
the WIFI module is used for achieving a WIFI networking function.
Preferably, the data acquisition unit comprises a video acquisition module and a smoke acquisition module, wherein:
the video acquisition module is used for shooting videos of fire sites;
the smoke collection module is used for collecting and analyzing smoke of a fire scene so as to judge fire conditions of the fire through smoke concentration.
Preferably, the data transmission unit includes a data capture module, a data encapsulation module, a signal conversion module, and a data transmission module, wherein:
the data grabbing module is used for grabbing video smoke feeling data acquired by the data acquisition unit;
the data packaging module is used for packaging the captured video smoke sensing data into an Ethernet packet;
the signal conversion module is used for converting the Ethernet packet into a PLC signal;
the data transmission module is used for transmitting the PLC signals to the fire disaster video intelligent identification device through the fire control bus.
Preferably, the identification judging unit includes a data extracting module and a data analyzing module, wherein:
the data extraction module is used for extracting data received by the fire video intelligent recognition device;
the data analysis module is used for identifying and judging the extracted data through an algorithm model.
Preferably, the data verification unit includes a secondary data extraction module, a secondary data analysis module and a data comparison module, wherein:
the secondary data extraction module is used for extracting fire scene video smoke feeling data which is different from the last data extraction time;
the secondary data analysis module is used for carrying out recognition judgment on the extracted verification data again;
the data comparison module is used for comparing and analyzing the data judgment result of the second extraction with the data judgment result of the last extraction.
Preferably, the handling processing unit includes an execution module, a dialing module, a monitoring module and a recording module, wherein:
the execution module is used for executing the tasks of switching on emergency broadcasting, starting a fire alarm bell, cutting a non-fire power supply, starting a smoke prevention and discharge system, lowering a fire prevention rolling curtain, forced landing of an elevator and automatic water pump;
the dialing module is used for executing dialing tasks and comprises:
notifying a security supervisor;
starting a fire emergency plan;
dialing a fire alarm telephone 119; and
notifying personnel to connect the fire truck and the like;
the monitoring module is used for monitoring the running condition of each device under the condition of fire;
the recording module is used for recording the coping process.
Preferably, in the process of pushing the result pushing unit to the attendees, the pushing mode includes one or more of short messages, telephones or APP.
The application also discloses an intelligent AI alarm method of the Internet of things based on the intelligent AI alarm system of the Internet of things, which comprises the following steps:
collecting video smoke sensing data of a fire scene and uploading the video smoke sensing data to a fire video intelligent recognition device;
transmitting video smoke sensing data acquired at a fire scene to a fire video intelligent identification device;
the video smoke sensing data is identified, judged and analyzed through an AI algorithm model so as to judge the severity of the fire disaster in the fire scene;
verifying fire scene judgment to prevent false alarm of fire situations;
according to the verified judgment result, selecting a corresponding coping strategy to carry out intelligent coping processing;
and uploading the fire judgment result to the cloud and pushing the fire judgment result to the attendees.
Preferably, the verification of the fire scene judgment is performed to prevent false alarm of fire situations, and the verification method comprises the following steps:
and the manual verification mode informs patrol personnel to confirm the fire disaster condition on site, and timely uploads the fire disaster condition on site through portable communication equipment.
The application has the technical effects and advantages that: compared with the prior art, the intelligent AI alarm system and method based on the Internet of things have the following advantages:
the intelligent AI alarm system and method related in the application comprises a data acquisition unit, a data transmission unit, an identification judging unit, a data verification unit, a coping processing unit and a result pushing unit, wherein the video smoke feeling data of the fire scene is acquired and uploaded to a fire video intelligent identification device; transmitting video smoke sensing data acquired at a fire scene to a fire video intelligent identification device; the video smoke sensing data is identified, judged and analyzed through an AI algorithm model so as to judge the severity of the fire disaster in the fire scene; verifying fire scene judgment to prevent false alarm of fire situations; according to the verified judgment result, selecting a corresponding coping strategy to carry out intelligent coping processing; the fire disaster judgment result is uploaded to the cloud and pushed to the attendees, and the fire disaster on-site condition can be rapidly identified and judged through the flow, so that the probability of fire disaster false alarm is reduced, and the response speed is high.
Drawings
FIG. 1 is a system block diagram of an intelligent AI alarm system based on the Internet of things in an embodiment of the application;
FIG. 2 is a flowchart of an intelligent AI alarm method based on the Internet of things in an embodiment of the application;
FIG. 3 is a block diagram of a fire video intelligent recognition device according to an embodiment of the present application;
FIG. 4 is a system block diagram of a data acquisition unit in an embodiment of the application;
FIG. 5 is a system block diagram of a data transmission unit in an embodiment of the present application;
FIG. 6 is a system block diagram of an identification judgment unit according to an embodiment of the present application;
FIG. 7 is a system block diagram of a data verification unit in an embodiment of the application;
FIG. 8 is a system block diagram of a processing unit according to an embodiment of the present application;
FIG. 9 is a system diagram of a first networking mode according to an embodiment of the present application;
FIG. 10 is a schematic diagram of a second networking mode in an embodiment of the present application;
FIG. 11 is a software framework diagram of a fire video intelligent recognition device in an embodiment of the application;
FIG. 12 is a schematic diagram of a PLC module in an embodiment of the application;
FIG. 13 is a schematic diagram of a switch in an embodiment of the application;
FIG. 14 is a flow chart of determining the severity of a fire scene according to an embodiment of the application.
Detailed Description
The following description of the embodiments of the present application will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present application, but not all embodiments. The specific embodiments described herein are merely illustrative of the application and are not intended to limit the application. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application.
The application discloses an intelligent AI alarm system based on the Internet of things, as shown in fig. 1, comprising:
the data acquisition unit is used for acquiring video smoke sensing data of the fire scene and uploading the video smoke sensing data to the fire video intelligent recognition device;
specifically, as shown in fig. 4, the data acquisition unit includes a video acquisition module and a smoke acquisition module, where: the video acquisition module is used for shooting video of a fire scene; the smoke collection module is used for collecting and analyzing smoke of a fire scene so as to judge fire of the fire through the smoke concentration.
The video smoke sensing data acquisition to the scene of fire among this embodiment adopts a smoke sensing device, and it is including casing and fixed mounting in the dish seat on the ceiling, and the casing can dismantle the cooperation with the dish seat, is fixed with the PCB board in the casing, is equipped with smoke transducer on the PCB board, and the diameter of casing is greater than the diameter of dish seat, and the junction of casing and dish seat is formed with cone transition portion, and cone transition portion is the fretwork form, and smoke alarm's inlet orientation cone transition portion.
The cone-shaped transition part is provided with a plurality of strip-shaped slotted holes which are arranged in parallel, and the plurality of strip-shaped slotted holes are uniformly distributed along the circumferential direction of the cone-shaped transition part.
One end of the shell far away from the tray seat is fixed with a face cover.
The PCB is provided with a camera, the face cover is provided with a window, and the camera and the window are mutually aligned.
A vertical shaft and a gear motor are fixed in the shell, a shaft sleeve is sleeved on the vertical shaft and is in running fit with the vertical shaft, an outwards extending baffle and a sector gear are formed at the edge of the shaft sleeve, a driving gear is fixed on an output shaft of the gear motor, teeth at the edge of the driving gear and the sector gear are meshed with each other, the gear motor drives the shaft sleeve to rotate forwards or reversely through the transmission fit of the driving gear and the sector gear, and then the baffle is driven to be blocked between a camera and a window or removed from the space between the camera and the window.
A camera jacket is fixed in the shell, and the camera is arranged in the camera jacket.
The vertical height of the camera is smaller than or equal to that of the camera jacket.
When the baffle shields between the camera and the window, the baffle covers the opening of the camera jacket.
The camera outer sleeve and the baffle are round.
Be equipped with a plurality of LED lamp pearls on the PCB board, a plurality of LED lamp pearls are along the edge evenly distributed of PCB board.
Further, as shown in fig. 3, the fire video intelligent recognition device includes a JETSON module, a PLC module, a 4G module and a WIFI module, wherein:
the JETSON module is connected with the PLC module through an Ethernet, and is respectively connected with the 4G module and the WIFI module through a USB interface; as shown in fig. 12 and 13, fig. 12 is a schematic diagram of a PLC module in the embodiment of the present application, and fig. 13 is a schematic diagram of a switch in the embodiment of the present application:
the JETSON module is hardware for realizing an AI algorithm model;
the PLC module is used for converting the PLC signals into Ethernet data;
the 4G module is used for realizing the functions of 4G internet surfing, short message service and telephone;
the WIFI module is used for realizing the WIFI networking function.
Specifically, the intelligent fire video identification device is powered by DC12V, and is positive and negative from inside to outside.
JETSON module, model JETSONTX2 of inflight. Is hardware that implements the algorithm function.
The PLC module is composed of circuits of two main chips, namely MSE510CE and RTL8305 NB. And the conversion of the PLC signals into Ethernet data is realized.
The 4G module is remote QUECTEEC 200A-CN, and the functions of 4G Internet surfing, short message service and telephone are realized.
The WIFI module is RTL8188EU, and the WIFI networking function is achieved.
In this embodiment, the software of the intelligent fire video recognition device adopts a multi-task architecture, as shown in fig. 11, and specifically includes the following steps:
GUI tasks: a QT-based user interface is provided to the user for setting parameters, presenting operating status, etc.
Fire control host computer interaction task: interaction with the H5800 host feeds back the results of the video analysis.
Cloud platform interaction tasks: and interacting with the fire cloud platform, and feeding back the video analysis result.
Task interaction of the attendees: and feeding back the video analysis result by means of short messages, telephones and the like.
Video reception task: a real-time video stream is received.
Video conversion tasks: the video stream is converted into JPEG pictures with ffmpeg and sent to the AI algorithm.
AI algorithm task: and analyzing the JPEG picture to judge whether smoke or flame exists. If smoke or flame is judged, the smoke or flame is marked by a red square frame, and the fire is judged. And feeding back the judgment result and the calibrated picture to the corresponding receiver.
Further, the above-mentioned AI algorithm model adopts YOLOV6 as the target detection framework, and first uses the standard weight as the pre-training weight. Open source picture set: COCO, imageNet, and the like. Private photo album: collecting pictures: according to different sunshine and lamplight, different scenes are generated, and each scene shoots 3 groups, namely, smoke and flame are respectively generated. And 13000 pictures are collected in an accumulated mode.
Further, the AI algorithm model training process is as follows:
and carrying out Gaussian noise processing on the picture to increase the ambiguity so as to enhance the reasoning effect of the cross-scene.
And randomly rotating the picture by any angle, and randomly performing gray scale expansion so as to prevent the picture angle from generating overfitting.
Randomly enlarging and reducing the picture to prevent missed detection of small targets and over fitting of targets with the same scale.
And labeling the picture set by using labelImg, and marking a target by using a medium rectangular frame in the picture labeled as true. The target type is saved in XML format. 80% of the pictures are divided into training sets and 20% of the pictures are divided into test sets.
Training is performed on a host equipped with a GPU.
The environment is as follows: i9-10900, memory 32G, NVIDIARTX3090. Software environment: ubiuu-18.04, python-3.10, pytorch-2.0.0, opencv-4.7.0.72, numpy-1.24.2.
The super parameters for training are set as follows: img-size is set to 640, batch-size is set to 32, maximum number of iterations epoch is set to 240, works is set to 12, device is set to 0, eval-interval is set to 30. The remaining parameters are default values. When the number of training rounds reaches 150, the total loss value starts to drop, and when the number reaches about 240, the loss value is stable and converges to about 0.02.
In addition, a YOLOV6 model-related file generated by training was installed to JETSONTX2. In this environment, 120 pictures can be processed per second, the flame detection rate is 92%, the smoke detection rate is 83%, and the false detection rate is 5%. Meets the requirement of auxiliary duty.
As shown in fig. 14, the present embodiment also discloses a method for determining the severity of fire in a fire scene as shown in fig. 14, which is specifically as follows:
classifying the existing fire disaster pictures into training pictures, and then respectively storing the training pictures in corresponding picture databases;
carrying out data enhancement processing on the existing training pictures, and preprocessing all the training pictures;
loading a convolutional neural network, inputting training pictures in a picture database into the convolutional neural network to train the convolutional neural network continuously, and calculating weight loss through a loss function;
optimizing the convolutional neural network according to the weight loss calculated by the loss function, stopping training after obtaining the optimal convolutional neural network, and fixing the weight;
inputting the fire scene pictures to be classified into an optimal convolutional neural network model for classification, and obtaining the classification result of the pictures;
the user can divide the severity degree according to the classification according to the fire scene pictures with different severity degrees in advance, such as primary fire, secondary fire, tertiary fire and the like, and can input the fire scene pictures to be classified into an optimal convolutional neural network model for classification to obtain a classification result of the pictures, and judge the fire severity degree of the fire scene according to the classification result.
The data transmission unit is used for transmitting the video smoke sensing data acquired at the fire scene to the fire video intelligent identification device;
specifically, as shown in fig. 5, the data transmission unit includes a data capturing module, a data packaging module, a signal conversion module, and a data transmission module, where:
the data grabbing module is used for grabbing the video smoke sensing data acquired by the data acquisition unit;
the data packaging module is used for packaging the captured video smoke sensing data into an Ethernet packet;
the signal conversion module is used for converting the Ethernet packet into a PLC signal;
the data transmission module is used for transmitting the PLC signals to the fire disaster video intelligent recognition device through the fire control bus.
In this embodiment, two networking modes of the smoke sensing device and the fire video intelligent recognition device are provided, and the networking modes are specifically as follows:
as shown in fig. 9, the first networking mode is as follows:
the H5800 intelligent fire alarm control system is connected with the intelligent fire video identification device through a network cable. The video smoke sensor encapsulates the video data (H264 format) into Ethernet packets, converts the Ethernet packets into PLC signals, and transmits the PLC signals to H5800 through a fire bus. And H5800 converts the PLC signals into Ethernet packets and forwards the Ethernet packets to the fire video intelligent recognition device.
The fire disaster video intelligent recognition device extracts an H264 video in the Ethernet packet, performs recognition judgment through an algorithm model, feeds back a recognition result to H5800, and responds according to the result. Meanwhile, the fire video intelligent recognition device reports the recognition result to the cloud platform, and informs the attendees through short messages, telephones, APP and the like.
As shown in fig. 10, the second networking method is as follows:
the fire disaster video intelligent recognition device is connected in parallel to the fire fighting bus through the fire fighting bus interface. The video smoke sensor is connected in parallel on the fire bus. The video smoke sensor encapsulates video data (H264 format) into an Ethernet packet, converts the Ethernet packet into a PLC signal and transmits the PLC signal to the fire video intelligent recognition device through a fire bus.
The PLC module of the fire disaster video intelligent recognition device converts the PLC signal into an Ethernet packet, extracts an H264 video in the Ethernet packet, performs recognition judgment through an algorithm model, reports the recognition result to a cloud platform, and notifies the attendees through modes such as short messages, telephones, APP and the like.
In this embodiment, the foregoing H5800 intelligent fire alarm control system includes a controller host, a plurality of smoke sensors, a plurality of audible and visual alarms and a plurality of manual alarms, where the controller host includes a panel shell and a bottom shell, and the panel shell is hinged with the bottom shell, the panel shell is provided with a display and a key unit, the bottom shell is provided with a main control board and a circuit board, the display, the key unit and the circuit board are respectively electrically connected to the main control board, the smoke sensors, the audible and visual alarms and the manual alarms are respectively connected to the circuit board by bus, and the main control board is used for: setting a system through a key unit; acquiring and processing electric signals fed back by the smoke sensor and the manual alarm through the loop board; the audible and visual alarm is controlled by the loop board to give out audible and visual alarm prompts; and displaying information through a display.
The display is a display with an Android system installed.
The key unit comprises a plurality of conductive touch key parts positioned on the face shell, a plurality of accommodating barrels are formed on the back side of the face shell, the accommodating barrels are in one-to-one correspondence with the touch key parts, conductive springs are arranged in the accommodating barrels, a touch key circuit board is fixed on the back side of the face shell, the rear ends of the conductive springs are electrically connected with the touch key circuit board, the front ends of the conductive springs are mutually abutted to the touch key parts, and the touch key circuit board is electrically connected with the main control board.
The front end of the conductive spring is wound with a disc-shaped end part, and the disc-shaped end part is mutually abutted with the touch key part.
The back side of the face shell is fixedly provided with a display back shell, the display back shell is coated on the back side of the display, and the display back shell is fixedly connected with the face shell.
Two circuit boards are fixed in the bottom shell and are arranged side by side, and the two circuit boards are located above the main control board.
The both sides of drain pan have seted up the louvre respectively, and the inboard of drain pan is fixed with two fans, and the air outlet of two fans is towards the louvre of drain pan both sides respectively, and the income wind gap of two fans is towards two return circuit boards respectively.
The power box is fixed in the bottom shell and is used for supplying power to the display, the main control board and the circuit board.
The main control board comprises a processor U5 and a short-circuit protection circuit.
The identification judging unit is used for carrying out identification judgment and analysis on the video smoke sensing data through the AI algorithm model so as to judge the severity of the fire disaster in the fire scene;
specifically, as shown in fig. 6, the identification judging unit includes a data extracting module and a data analyzing module, where:
the data extraction module is used for extracting the data received by the fire video intelligent recognition device;
the data analysis module is used for identifying and judging the extracted data through the algorithm model.
The data verification unit is used for verifying the fire scene judgment so as to prevent false alarm of the fire situation;
specifically, as shown in fig. 7, the data verification unit includes a secondary data extraction module, a secondary data analysis module, and a data comparison module, where:
the secondary data extraction module is used for extracting the video smoke feeling data of the fire scene, which is different from the last data extraction time;
the secondary data analysis module is used for carrying out recognition judgment on the extracted verification data again;
the data comparison module is used for comparing and analyzing the data judgment result of the second extraction with the data judgment result of the last extraction.
The verification method also comprises a manual verification method, and the patrol personnel are informed of confirming the fire situation on site and upload the fire situation in time through portable communication equipment.
The coping processing unit is used for selecting a corresponding coping strategy to carry out intelligent coping processing according to the verified judging result;
specifically, as shown in fig. 8, the coping processing unit includes an executing module, a dialing module, a monitoring module and a recording module, where:
the execution module is used for executing the tasks of switching on emergency broadcasting, starting a fire alarm bell, cutting a non-fire power supply, starting a smoke prevention and discharge system, lowering a fire prevention rolling curtain, and forced landing of an elevator and an automatic water pump;
the dialing module is used for executing dialing tasks and comprises:
notifying a security supervisor;
starting a fire emergency plan;
dialing a fire alarm telephone 119; and
notifying personnel to connect the fire truck and the like;
the monitoring module is used for monitoring the running condition of each device under the condition of fire;
the recording module is used for recording the coping process.
A result pushing unit; and the fire judgment result is uploaded to the cloud and pushed to the attendees.
Based on the intelligent AI alarm system based on the Internet of things, as shown in fig. 2, the embodiment also discloses an intelligent AI alarm method based on the Internet of things, which comprises the following steps:
s1, acquiring video smoke sensing data of a fire scene, and uploading the video smoke sensing data to a fire video intelligent recognition device; fire video intelligent recognition device includes JETSON module, PLC module, 4G module and WIFI module, wherein: the JETSON module is connected with the PLC module through an Ethernet, and is respectively connected with the 4G module and the WIFI module through a USB interface; the JETSON module is hardware for realizing an AI algorithm model; the PLC module is used for converting the PLC signals into Ethernet data; the 4G module is used for realizing the functions of 4G internet surfing, short messages and telephones; the WIFI module is used for achieving a WIFI networking function.
S2, transmitting video smoke sensing data acquired at a fire scene to a fire video intelligent identification device; the data grabbing module is used for grabbing video smoke feeling data acquired by the data acquisition unit; the data packaging module is used for packaging the captured video smoke sensing data into an Ethernet packet; the signal conversion module is used for converting the Ethernet packet into a PLC signal; the data transmission module is used for transmitting the PLC signals to the fire disaster video intelligent identification device through the fire control bus.
S3, carrying out identification, judgment and analysis on the video smoke sensing data through an AI algorithm model so as to judge the severity of fire in a fire scene; the data extraction module is used for extracting the data received by the fire video intelligent recognition device; the data analysis module is used for identifying and judging the extracted data through an algorithm model.
S4, verifying fire scene judgment to prevent false alarm of fire situations; the secondary data extraction module is used for extracting the video smoke feeling data of the fire scene, which is different from the last data extraction time; the secondary data analysis module is used for carrying out recognition judgment on the extracted verification data again; the data comparison module is used for comparing and analyzing the data judgment result of the second extraction with the data judgment result of the last extraction.
Specifically, the secondary data extraction module is used for extracting fire scene video smoke feeling data different from the last data extraction time, comparing and analyzing the fire scene video smoke feeling data with the last data extraction time, and if the difference value between the analysis result and the fire scene video smoke feeling data of the last data extraction time exceeds a preset threshold value, indicating that a false alarm occurs in the fire situation.
S5, selecting a corresponding coping strategy to carry out intelligent coping according to the verified judgment result; the execution module is used for executing the tasks of switching on emergency broadcasting, starting a fire alarm bell, cutting a non-fire power supply strongly, starting a smoke prevention and discharge system, lowering a fire prevention rolling curtain, forced landing of an elevator and automatic water pump; the dialing module is used for executing dialing tasks and comprises: notifying a security supervisor; starting a fire emergency plan; dialing a fire alarm telephone 119; notifying personnel to receive a fire truck and the like; the monitoring module is used for monitoring the running condition of each device under the condition of fire; the recording module is used for recording the coping process.
And S6, uploading the fire judgment result to the cloud end and pushing the fire judgment result to the attendees. In the process of pushing the result to the attendees in the result pushing unit, the pushing mode comprises one or more of short messages, telephones or APP.
In addition, the fire scene judgment is verified to prevent false alarm of fire situations, and the verification method comprises the following steps:
and the manual verification mode informs patrol personnel to confirm the fire disaster condition on site, and timely uploads the fire disaster condition on site through portable communication equipment.
In one embodiment, a computer device is provided, which may be a server. The computer device includes a processor, a memory, and a network interface connected by a device bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The nonvolatile storage medium stores an operating device, a computer program, and a database. The internal memory provides an environment for the operation of the operating device and the computer program in the non-volatile storage medium. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program, when executed by a processor, implements the intelligent AI alarm method.
Those skilled in the art will appreciate that the application is not limited in its construction to the arrangements of parts illustrated in the drawings, but to the computer apparatus in which the application is applied, and that a particular computer apparatus may include more or less elements than those illustrated in the drawings, or may have a different arrangement of elements.
In one embodiment, the application also provides a computer device, which comprises a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to realize the fire alarm method.
In one embodiment, the present application also provides a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements the fire alarm method described above.
The fire alarm system, the computer equipment and the computer readable storage medium collect video smoke sensing data of a fire scene and upload the video smoke sensing data to the fire video intelligent recognition device; transmitting video smoke sensing data acquired at a fire scene to a fire video intelligent identification device; the video smoke sensing data is identified, judged and analyzed through an AI algorithm model so as to judge the severity of the fire disaster in the fire scene; verifying fire scene judgment to prevent false alarm of fire situations; according to the verified judgment result, selecting a corresponding coping strategy to carry out intelligent coping processing; and uploading the fire judgment result to the cloud and pushing the fire judgment result to the attendees.
Those skilled in the art will appreciate that implementing all or part of the above described methods may be accomplished by way of a computer program stored on a non-transitory computer readable storage medium, which when executed, may comprise the steps of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium used in embodiments provided herein may include non-volatile and/or volatile memory. The nonvolatile memory can include Read Only Memory (ROM), programmable ROM (PROM), electrically Programmable ROM (EPROM), electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double Data Rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous Link DRAM (SLDRAM), memory bus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), among others.
Finally, it should be noted that: the foregoing description is only illustrative of the preferred embodiments of the present application, and although the present application has been described in detail with reference to the foregoing embodiments, it will be apparent to those skilled in the art that modifications may be made to the embodiments described, or equivalents may be substituted for elements thereof, and any modifications, equivalents, improvements or changes may be made without departing from the spirit and principles of the present application.

Claims (10)

1. Intelligent AI alarm system based on thing networking, its characterized in that includes:
the data acquisition unit is used for acquiring video smoke sensing data of the fire scene and uploading the video smoke sensing data to the fire video intelligent recognition device;
the data transmission unit is used for transmitting the video smoke sensing data acquired at the fire scene to the fire video intelligent identification device;
the identification judging unit is used for carrying out identification judgment and analysis on the video smoke sensing data through the AI algorithm model so as to judge the severity of the fire disaster in the fire scene;
the data verification unit is used for verifying the fire scene judgment so as to prevent false alarm of the fire situation;
the coping processing unit is used for selecting a corresponding coping strategy to carry out intelligent coping processing according to the verified judging result;
a result pushing unit; and the fire judgment result is uploaded to the cloud and pushed to the attendees.
2. The intelligent AI alarm system based on the internet of things of claim 1, wherein: the fire disaster video intelligent recognition device comprises a JETSON module, a PLC module, a 4G module and a WIFI module, wherein:
the JETSON module is connected with the PLC module through an Ethernet, and is respectively connected with the 4G module and the WIFI module through a USB interface;
the JETSON module is hardware for realizing an AI algorithm model;
the PLC module is used for converting the PLC signals into Ethernet data;
the 4G module is used for realizing the functions of 4G internet surfing, short messages and telephones;
the WIFI module is used for achieving a WIFI networking function.
3. The intelligent AI alarm system based on the internet of things of claim 2, wherein: the data acquisition unit comprises a video acquisition module and a smoke acquisition module, wherein:
the video acquisition module is used for shooting videos of fire sites;
the smoke collection module is used for collecting and analyzing smoke of a fire scene so as to judge fire conditions of the fire through smoke concentration.
4. The intelligent AI alarm system based on the internet of things of claim 3, wherein: the data transmission unit comprises a data grabbing module, a data packaging module, a signal conversion module and a data transmission module, wherein:
the data grabbing module is used for grabbing video smoke feeling data acquired by the data acquisition unit;
the data packaging module is used for packaging the captured video smoke sensing data into an Ethernet packet;
the signal conversion module is used for converting the Ethernet packet into a PLC signal;
the data transmission module is used for transmitting the PLC signals to the fire disaster video intelligent identification device through the fire control bus.
5. The intelligent AI alarm system based on the internet of things of claim 4, wherein: the identification judging unit comprises a data extraction module and a data analysis module, wherein:
the data extraction module is used for extracting data received by the fire video intelligent recognition device;
the data analysis module is used for identifying and judging the extracted data through an algorithm model.
6. The intelligent AI alarm system based on the internet of things of claim 5, wherein: the data verification unit comprises a secondary data extraction module, a secondary data analysis module and a data comparison module, wherein:
the secondary data extraction module is used for extracting fire scene video smoke feeling data which is different from the last data extraction time;
the secondary data analysis module is used for carrying out recognition judgment on the extracted verification data again;
the data comparison module is used for comparing and analyzing the data judgment result of the second extraction with the data judgment result of the last extraction.
7. The intelligent AI alarm system based on the internet of things of claim 6, wherein: the handling processing unit comprises an execution module, a dialing module, a monitoring module and a recording module, wherein:
the execution module is used for executing the tasks of switching on emergency broadcasting, starting a fire alarm bell, cutting a non-fire power supply, starting a smoke prevention and discharge system, lowering a fire prevention rolling curtain, forced landing of an elevator and automatic water pump;
the dialing module is used for executing dialing tasks and comprises:
notifying a security supervisor;
starting a fire emergency plan;
dialing a fire alarm telephone 119; and
notifying personnel to connect the fire truck and the like;
the monitoring module is used for monitoring the running condition of each device under the condition of fire;
the recording module is used for recording the coping process.
8. The intelligent AI alarm system based on the internet of things of claim 7, wherein: in the process of pushing the result pushing unit to the attendees, the pushing mode comprises one or more of short messages, telephones or APP.
9. An intelligent AI alarm method based on the internet of things, based on the intelligent AI alarm system based on the internet of things as set forth in any one of claims 1 to 8, comprising the steps of:
collecting video smoke sensing data of a fire scene and uploading the video smoke sensing data to a fire video intelligent recognition device;
transmitting video smoke sensing data acquired at a fire scene to a fire video intelligent identification device;
the video smoke sensing data is identified, judged and analyzed through an AI algorithm model so as to judge the severity of the fire disaster in the fire scene;
verifying fire scene judgment to prevent false alarm of fire situations;
according to the verified judgment result, selecting a corresponding coping strategy to carry out intelligent coping processing;
and uploading the fire judgment result to the cloud and pushing the fire judgment result to the attendees.
10. The intelligent AI alarm method based on the internet of things according to claim 9, wherein the verification of the fire scene judgment to prevent false alarm of the fire situation comprises the following verification steps:
and the manual verification mode informs patrol personnel to confirm the fire disaster condition on site, and timely uploads the fire disaster condition on site through portable communication equipment.
CN202311243457.0A 2023-09-26 2023-09-26 Intelligent AI alarm system and method based on Internet of things Pending CN116994396A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202311243457.0A CN116994396A (en) 2023-09-26 2023-09-26 Intelligent AI alarm system and method based on Internet of things

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202311243457.0A CN116994396A (en) 2023-09-26 2023-09-26 Intelligent AI alarm system and method based on Internet of things

Publications (1)

Publication Number Publication Date
CN116994396A true CN116994396A (en) 2023-11-03

Family

ID=88528643

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202311243457.0A Pending CN116994396A (en) 2023-09-26 2023-09-26 Intelligent AI alarm system and method based on Internet of things

Country Status (1)

Country Link
CN (1) CN116994396A (en)

Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR101934700B1 (en) * 2018-07-12 2019-01-03 주식회사동우유니온 Early fire detection system, server and method using image processing and artificial intelligence based on deep learning
CN111145488A (en) * 2018-11-06 2020-05-12 珠海精讯知识产权管理有限公司 AI video smoke detector fire alarm device
US20200388128A1 (en) * 2019-06-07 2020-12-10 Chang Sung Ace Co., Ltd. Fire detection system based on artificial intelligence and fire detection method based on artificial intelligence
CN112489340A (en) * 2019-09-12 2021-03-12 华为技术有限公司 Fire-fighting terminal, fire-fighting control terminal and fire-fighting method
CN112489355A (en) * 2020-12-09 2021-03-12 深圳前海奥瑞那安全技术有限公司 Intelligent fire alarm control system and method
CN112562249A (en) * 2020-12-09 2021-03-26 深圳前海奥瑞那安全技术有限公司 Smoke induction device
CN217305948U (en) * 2021-11-19 2022-08-26 合肥云通物联科技有限公司 Firework detection monitoring system based on Internet of things + AI
CN115331384A (en) * 2022-08-22 2022-11-11 重庆科技学院 Operation platform fire accident early warning system based on edge calculation
CN116682223A (en) * 2022-12-08 2023-09-01 浪潮软件集团有限公司 Fire inspection system and method based on Internet of things and AI under domestic environment

Patent Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR101934700B1 (en) * 2018-07-12 2019-01-03 주식회사동우유니온 Early fire detection system, server and method using image processing and artificial intelligence based on deep learning
CN111145488A (en) * 2018-11-06 2020-05-12 珠海精讯知识产权管理有限公司 AI video smoke detector fire alarm device
US20200388128A1 (en) * 2019-06-07 2020-12-10 Chang Sung Ace Co., Ltd. Fire detection system based on artificial intelligence and fire detection method based on artificial intelligence
CN112489340A (en) * 2019-09-12 2021-03-12 华为技术有限公司 Fire-fighting terminal, fire-fighting control terminal and fire-fighting method
CN112489355A (en) * 2020-12-09 2021-03-12 深圳前海奥瑞那安全技术有限公司 Intelligent fire alarm control system and method
CN112562249A (en) * 2020-12-09 2021-03-26 深圳前海奥瑞那安全技术有限公司 Smoke induction device
CN217305948U (en) * 2021-11-19 2022-08-26 合肥云通物联科技有限公司 Firework detection monitoring system based on Internet of things + AI
CN115331384A (en) * 2022-08-22 2022-11-11 重庆科技学院 Operation platform fire accident early warning system based on edge calculation
CN116682223A (en) * 2022-12-08 2023-09-01 浪潮软件集团有限公司 Fire inspection system and method based on Internet of things and AI under domestic environment

Similar Documents

Publication Publication Date Title
CN107911653B (en) Intelligent video monitoring module, system, method and storage medium for residence
CN103150856B (en) Fire flame video monitoring and early warning system
CN113706817B (en) Transformer substation intelligent safety management method and system based on big data
CN108389359B (en) Deep learning-based urban fire alarm method
CN102176270A (en) Safety monitoring and fire alarming integrated system and method
CN112737121A (en) Intelligent video monitoring, analyzing, controlling and managing system for power grid
CN201897856U (en) Safety monitoring and fire hazard early-warning integrated system
CN110240036B (en) Device and method for detecting electric bicycle trying to enter elevator
CN109993949B (en) Fire safety detection method based on multi-sensor fusion
CN212782246U (en) Smoke and fire recognition system based on artificial intelligence
CN110759199A (en) Device and method for detecting electric bicycle trying to enter elevator
CN114900513B (en) Remote pet safety alarm system
CN210839070U (en) Laser external-damage-prevention device for power transmission line
CN207883080U (en) A kind of video fire hazard alarm system
CN111908288A (en) TensorFlow-based elevator safety system and method
CN116994396A (en) Intelligent AI alarm system and method based on Internet of things
CN107909766A (en) A kind of monitoring system for multi-storied garage
KR102081577B1 (en) Intelligence Fire Detecting System Using CCTV
CN202661864U (en) Centralized control station running environment remote comprehensive monitoring system
CN115083132A (en) Research and judgment method for reducing false alarm rate of fire alarm
KR101167605B1 (en) Intelligent controller using camera and method thereof
CN116740874A (en) Intrusion detection method and related device
CN203204762U (en) A fire flame video early warning system
CN111796550A (en) 5g intelligent control district transformer substation auxiliary system with early warning function
CN205490954U (en) Warehouse fire alarm system based on image detection

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