CN115190277B - Safety monitoring method, device and equipment for construction area and storage medium - Google Patents

Safety monitoring method, device and equipment for construction area and storage medium Download PDF

Info

Publication number
CN115190277B
CN115190277B CN202211095238.8A CN202211095238A CN115190277B CN 115190277 B CN115190277 B CN 115190277B CN 202211095238 A CN202211095238 A CN 202211095238A CN 115190277 B CN115190277 B CN 115190277B
Authority
CN
China
Prior art keywords
abnormal
monitoring
area
image
coordinate
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.)
Active
Application number
CN202211095238.8A
Other languages
Chinese (zh)
Other versions
CN115190277A (en
Inventor
丁骏韬
邵琼
张沛
罗元飞
黄亮
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.)
Sinodaan Co ltd
Original Assignee
Sinodaan 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 Sinodaan Co ltd filed Critical Sinodaan Co ltd
Priority to CN202211095238.8A priority Critical patent/CN115190277B/en
Publication of CN115190277A publication Critical patent/CN115190277A/en
Application granted granted Critical
Publication of CN115190277B publication Critical patent/CN115190277B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N7/00—Television systems
    • H04N7/18—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/10—Office automation; Time management
    • G06Q10/103—Workflow collaboration or project management
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/08—Construction
    • 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
    • G06V10/77—Processing image or video features in feature spaces; using data integration or data reduction, e.g. principal component analysis [PCA] or independent component analysis [ICA] or self-organising maps [SOM]; Blind source separation
    • G06V10/774—Generating sets of training patterns; Bootstrap methods, e.g. bagging or boosting
    • 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
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/50—Context or environment of the image
    • G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
    • G—PHYSICS
    • 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
    • G08B25/01—Alarm systems in which the location of the alarm condition is signalled to a central station, e.g. fire or police telegraphic systems characterised by the transmission medium
    • G08B25/08—Alarm systems in which the location of the alarm condition is signalled to a central station, e.g. fire or police telegraphic systems characterised by the transmission medium using communication transmission lines
    • 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
    • G16Y40/50—Safety; Security of things, users, data or systems
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N5/00—Details of television systems
    • H04N5/14—Picture signal circuitry for video frequency region

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Business, Economics & Management (AREA)
  • General Physics & Mathematics (AREA)
  • Multimedia (AREA)
  • Human Resources & Organizations (AREA)
  • Strategic Management (AREA)
  • General Health & Medical Sciences (AREA)
  • Computing Systems (AREA)
  • Health & Medical Sciences (AREA)
  • Evolutionary Computation (AREA)
  • General Business, Economics & Management (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Signal Processing (AREA)
  • Software Systems (AREA)
  • Marketing (AREA)
  • Economics (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Medical Informatics (AREA)
  • Artificial Intelligence (AREA)
  • Tourism & Hospitality (AREA)
  • Databases & Information Systems (AREA)
  • Computer Security & Cryptography (AREA)
  • Data Mining & Analysis (AREA)
  • Operations Research (AREA)
  • Primary Health Care (AREA)
  • Quality & Reliability (AREA)
  • Emergency Management (AREA)
  • Alarm Systems (AREA)

Abstract

The invention relates to the technical field of building monitoring, and discloses a safety monitoring method, a safety monitoring device, safety monitoring equipment and a storage medium for a construction area. The method comprises the following steps: acquiring a monitoring area coordinate set of a directional area corresponding to each acquisition device, and acquiring a monitoring video corresponding to a construction area through the acquisition devices; performing image framing processing on the monitoring video through the IOT platform, detecting abnormal areas in the monitoring images through the safety monitoring module, and extracting corresponding abnormal coordinates of the abnormal areas in the monitoring images; detecting whether the abnormal coordinates are coordinate areas corresponding to the coordinate set of the monitored area through a safety monitoring module, and performing grid division and abnormal category probability prediction on the abnormal areas; and matching corresponding alarm events to the abnormal objects in the abnormal area through the IOT platform, generating alarm information according to the alarm events through the ispM platform, and pushing the alarm information to the corresponding monitoring terminal. The invention improves the early warning efficiency of safety monitoring of the enclosure inlet and the enclosure outlet of the construction site.

Description

Safety monitoring method, device and equipment for construction area and storage medium
Technical Field
The invention relates to the technical field of building monitoring, in particular to a safety monitoring method, a safety monitoring device, safety monitoring equipment and a storage medium for a construction area.
Background
Along with the rapid advance of the urbanization process in China, the construction industry also enters in and goes out the development stage. So that the quantity of related construction projects in the city is continuously increased, and the construction site can be seen everywhere in the city. The construction site is used as a main site of building construction, a plurality of large dangerous construction equipment can be seen on the construction site, and the mobility of construction personnel is very high, so that the whole construction site has a plurality of dangerous sources and is easy to generate safety accidents. Therefore, in order to supervise and protect the potential safety hazard of the construction site, the construction site needs to be monitored and managed safely.
Nowadays, often adopt to enclose the formula management to the job site, through enclosing the construction area who corresponds to set up corresponding public security pavilion to the exit that encloses and monitor, prevent that the outsider or other living beings from carrying out the construction area, but because this kind of safety monitoring lacks automatic realization, when not special periods such as construction period, the safety problem that the living beings accident was gone into can not discover in time, early warning in advance to the job site. Namely, the prior early warning efficiency for the safety monitoring of the closure inlet and outlet of the construction site is lower.
Disclosure of Invention
The invention mainly aims to solve the problem of low early warning efficiency of the existing safety monitoring of the entrance and exit of the enclosure of the construction site.
The invention provides a safety monitoring method of a construction area, which is applied to a safety monitoring system of the construction area, wherein the safety monitoring system comprises a safety monitoring module, an IOT platform and an ispM platform, the IOT platform comprises a plurality of acquisition devices, and the safety monitoring method of the construction area comprises the following steps: acquiring a monitoring area coordinate set of the directional area corresponding to each acquisition device, and acquiring a monitoring video corresponding to the construction area through the acquisition device; performing image framing processing on the monitoring video through the IOT platform to obtain a plurality of initial monitoring images, detecting abnormal areas in the monitoring images through the safety monitoring module, and extracting corresponding abnormal coordinates of the abnormal areas in the monitoring images; detecting whether the abnormal coordinates are coordinate areas corresponding to the monitored area coordinate set or not through the safety monitoring module, and performing grid division and abnormal category probability prediction on the abnormal areas according to the coordinate detection result to obtain an abnormal identification result; and matching a corresponding alarm event for the abnormal object in the abnormal area through the IOT platform based on the abnormal identification result, generating alarm information according to the alarm event through the ispM platform, and pushing the alarm information to a corresponding monitoring terminal.
Optionally, in a first implementation manner of the first aspect of the present invention, the performing, by the IOT platform, image framing processing on the surveillance video to obtain a plurality of initial surveillance images includes: establishing a single-thread framing channel for each acquisition device through the IOT platform; and performing image framing on each monitoring video by using the single-thread framing channel, and performing data stream packaging on the framed images to obtain a plurality of initial monitoring images.
Optionally, in a second implementation manner of the first aspect of the present invention, the detecting, by the security monitoring module, an abnormal region in the monitored image includes: selecting a preset number of monitoring images in the same acquisition device through the safety monitoring module to obtain a plurality of groups of primary monitoring image sets, and performing image data enhancement processing on each primary monitoring image set to obtain a plurality of groups of enhanced monitoring image sets; performing element slicing on each monitoring image set, and performing weighted average processing on the result of the element slicing by using a preset pixel size to obtain a plurality of monitoring feature maps; carrying out image grid division on each monitoring feature map, and carrying out abnormal feature recognition on the monitoring feature maps of each grid after division to obtain an abnormal recognition result; and respectively calculating the abnormal probability of the corresponding image grids in the abnormal recognition result, and selecting the image grids exceeding the preset abnormal probability to generate abnormal regions.
Optionally, in a third implementation manner of the first aspect of the present invention, the extracting the abnormal coordinates corresponding to the abnormal region in the monitored image includes: determining a plurality of abnormal image grids corresponding to the abnormal regions; identifying a plurality of grid coordinate points of each abnormal image grid based on a preset characteristic image coordinate axis; and selecting a coordinate point of the area boundary in the grid coordinate points to obtain abnormal coordinates.
Optionally, in a fourth implementation manner of the first aspect of the present invention, the performing mesh division and abnormal category probability prediction on the abnormal region according to a result of the coordinate detection to obtain an abnormal recognition result includes: calculating the predicted value of the boundary frame of each grid in the abnormal area according to the coordinate detection result; performing coordinate calculation on the object in each grid in the abnormal area by using a full-connection layer in a preset construction monitoring model, and detecting whether the result of the coordinate calculation belongs to a predicted value of a bounding box; and if the two types of abnormal recognition belong to the same category, performing confidence calculation on the result of the coordinate calculation and the predicted value of each bounding box to obtain an abnormal recognition result corresponding to the accuracy of the abnormal category.
Optionally, in a fifth implementation manner of the first aspect of the present invention, the matching, by the IOT platform, a corresponding alarm event to the abnormal object in the abnormal region based on the abnormal recognition result includes: if the abnormal recognition result contains an abnormal object, marking the abnormal object in the monitoring image according to the abnormal recognition result; and generating an alarm image set according to the labeled monitoring image, calling back the alarm image set to the IOT platform, and generating an alarm event corresponding to the abnormal type according to the called-back alarm image set through the IOT platform.
Optionally, in a sixth implementation manner of the first aspect of the present invention, the generating, by the isPM platform, alarm information according to the alarm event, and pushing the alarm information to a corresponding monitoring terminal includes: the ispM platform selects a corresponding data format type according to a monitoring scene corresponding to the alarm event and executes preprocessing on the alarm event; generating corresponding alarm content according to the preprocessing result, and generating a plurality of alarm information by adopting the alarm content by utilizing a preset monitoring object template; and pushing the alarm information to the corresponding monitoring terminal.
The second aspect of the present invention provides a safety monitoring device for a construction area, which is applied to a safety monitoring system for a construction area, wherein the safety monitoring system comprises a safety monitoring module, an IOT platform and an isPM platform, the IOT platform comprises a plurality of acquisition devices, and the safety monitoring device for a construction area comprises: the video acquisition module is used for acquiring a monitoring area coordinate set of the directional area corresponding to each acquisition device and acquiring a monitoring video corresponding to the construction area through the acquisition device; the coordinate extraction module is used for performing image framing processing on the monitoring video through the IOT platform to obtain a plurality of initial monitoring images, detecting abnormal areas in the monitoring images through the safety monitoring module and extracting corresponding abnormal coordinates of the abnormal areas in the monitoring images; the abnormal recognition module is used for detecting whether the abnormal coordinates are coordinate regions corresponding to the monitored region coordinate set through the safety monitoring module, and performing grid division and abnormal category probability prediction on the abnormal regions according to the coordinate detection result to obtain an abnormal recognition result; and the alarm pushing module is used for matching a corresponding alarm event to the abnormal object in the abnormal area through the IOT platform based on the abnormal identification result, generating alarm information according to the alarm event through the ispM platform, and pushing the alarm information to a corresponding monitoring terminal.
Optionally, in a first implementation manner of the second aspect of the present invention, the coordinate extraction module includes: a channel establishing unit, configured to establish a single-threaded framing channel for each acquisition device through the IOT platform; and the image framing unit is used for performing image framing on each monitoring video by using the single-thread framing channel and performing data stream packaging on the framed images to obtain a plurality of initial monitoring images.
Optionally, in a second implementation manner of the second aspect of the present invention, the coordinate extraction module further includes: the image enhancement unit is used for selecting a preset number of monitoring images in the same acquisition device through the safety monitoring module to obtain a plurality of groups of primary monitoring image sets, and performing image data enhancement processing on each primary monitoring image set to obtain a plurality of groups of enhanced monitoring image sets; the weighted average unit is used for carrying out element slicing on each monitoring image set and carrying out weighted average processing on the result of the element slicing by a preset pixel size to obtain a plurality of monitoring characteristic graphs; the characteristic identification unit is used for carrying out image grid division on each monitoring characteristic graph and carrying out abnormal characteristic identification on the monitoring characteristic graphs of each grid after division to obtain an abnormal identification result; and the probability calculation unit is used for respectively calculating the abnormal probability of the corresponding image grids in the abnormal recognition result, selecting the image grids exceeding the preset abnormal probability and generating an abnormal region.
Optionally, in a third implementation manner of the second aspect of the present invention, the coordinate extraction module further includes: a grid determining unit, configured to determine a plurality of abnormal image grids corresponding to the abnormal region; the coordinate identification unit is used for identifying a plurality of grid coordinate points of each abnormal image grid based on a preset characteristic image coordinate axis; and the coordinate selection unit is used for selecting the coordinate points of the area boundary in the grid coordinate points to obtain abnormal coordinates.
Optionally, in a fourth implementation manner of the second aspect of the present invention, the abnormality recognition module includes: the boundary prediction unit is used for calculating the predicted value of the boundary frame of each grid in the abnormal area according to the coordinate detection result; the coordinate calculation unit is used for performing coordinate calculation on the objects in each grid in the abnormal area by using a full connection layer in a preset construction monitoring model and detecting whether the result of the coordinate calculation belongs to a predicted value of the boundary frame; and the abnormality identification unit is used for carrying out confidence calculation on the result of the coordinate calculation and the predicted value of each boundary frame if the abnormal recognition result belongs to the boundary frame, and obtaining an abnormality identification result corresponding to the accuracy of the abnormal category.
Optionally, in a fifth implementation manner of the second aspect of the present invention, the alert pushing module includes: the abnormal marking unit is used for marking the abnormal construction area in the monitoring image according to the identification result if the identification result contains the abnormal construction area; and the image callback unit is used for generating an alarm image set according to the labeled monitoring image, calling back the alarm image set to the IOT platform, and generating an alarm event corresponding to the abnormal type according to the called-back alarm image set through the IOT platform.
Optionally, in a sixth implementation manner of the second aspect of the present invention, the alert pushing module further includes: the type selection unit is used for selecting a corresponding data format type by the ispM platform according to the monitoring scene corresponding to the alarm event and executing preprocessing on the alarm event; the information generating unit is used for generating corresponding alarm content according to the preprocessing result and generating a plurality of alarm information by adopting the alarm content by utilizing a preset monitoring object template; and the alarm pushing unit is used for pushing the alarm information to the corresponding monitoring terminal.
A third aspect of the present invention provides a safety monitoring apparatus for a construction area, comprising: a memory and at least one processor, the memory having instructions stored therein; the at least one processor calls the instructions in the memory to cause the safety monitoring equipment of the construction area to execute the steps of the safety monitoring method of the construction area.
A fourth aspect of the present invention provides a computer-readable storage medium having stored therein instructions, which, when run on a computer, cause the computer to perform the steps of the above-described method for safety monitoring of a construction area.
In the technical scheme provided by the invention, the method is applied to a safety monitoring system of a construction area, a monitoring area coordinate set of a directional area corresponding to each acquisition device is obtained, and a monitoring video corresponding to the construction area is acquired through the acquisition devices; performing image framing processing on the monitoring video through an IOT platform to obtain a plurality of initial monitoring images, detecting abnormal areas in the monitoring images through a safety monitoring module, and extracting corresponding abnormal coordinates of the abnormal areas in the monitoring images; detecting whether the abnormal coordinates are coordinate areas corresponding to the coordinate set of the monitored area through a safety monitoring module, and performing grid division and abnormal category probability prediction on the abnormal areas according to the coordinate detection result to obtain an abnormal identification result; and matching corresponding alarm events to the abnormal objects in the abnormal area through the IOT platform based on the abnormal identification result, generating alarm information according to the alarm events through the ispM platform, and pushing the alarm information to the corresponding monitoring terminal. Compared with the prior art, the method and the device have the advantages that the video of the corresponding import and export area is collected through the collection device, the collected video is subjected to framing and abnormal coordinate matching, and then the abnormal area is identified in abnormal categories, so that corresponding early warning information is generated according to an abnormal identification result and pushed. The real-time detection of the fixed-point acquisition area in the construction entrance and exit area is realized, the corresponding early warning information is generated according to the abnormal type with abnormal intrusion and is pushed, and the early warning efficiency of safety monitoring of the entrance and exit enclosed in the construction field is improved.
Drawings
FIG. 1 is a schematic diagram of a first embodiment of a safety monitoring method for a construction area according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of a second embodiment of a safety monitoring method for a construction area according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of a safety monitoring method for a construction area according to a third embodiment of the present invention;
FIG. 4 is a schematic diagram of an embodiment of a safety monitoring device for a construction area according to an embodiment of the present invention;
FIG. 5 is a schematic diagram of another embodiment of a safety monitoring device for a construction area according to an embodiment of the present invention;
fig. 6 is a schematic diagram of an embodiment of a safety monitoring device for a construction area according to an embodiment of the present invention.
Detailed Description
The embodiment of the invention provides a safety monitoring method, a safety monitoring device, safety monitoring equipment and a storage medium for a construction area, wherein the method comprises the following steps: acquiring a monitoring video of a construction area through the IOT platform, and performing framing processing on the monitoring video to obtain a plurality of monitoring images; performing data enhancement and abstract feature point extraction on each monitoring image by using a preset scene monitoring model through the safety monitoring module, and performing abnormal construction area identification on each monitoring image according to the extracted abstract feature points to obtain an identification result; if the identification result contains an abnormal construction area, triggering an alarm event of a corresponding abnormal type to the abnormal construction area in the IOT platform according to the identification result; and the ispM platform generates alarm information according to the alarm event and pushes the alarm information to a corresponding monitoring terminal. The invention improves the early warning efficiency of the safety monitoring of the construction site.
The terms "first," "second," "third," "fourth," and the like in the description and in the claims, as well as in the drawings, if any, are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It will be appreciated that the data so used may be interchanged under appropriate circumstances such that the embodiments described herein may be practiced otherwise than as specifically illustrated or described herein. Furthermore, the terms "comprises," "comprising," or "having," and any variations thereof, are intended to cover non-exclusive inclusions, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed, but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
For convenience of understanding, a specific flow of the embodiment of the present invention is described below, and referring to fig. 1, a first embodiment of a method for monitoring safety of a construction area in the embodiment of the present invention includes:
101. acquiring a monitoring area coordinate set of a directional area corresponding to each acquisition device, and acquiring a monitoring video corresponding to a construction area through the acquisition devices;
it is to be understood that the execution subject of the present invention may be a safety monitoring device of a construction area, and may also be a terminal or a server, which is not limited herein. The embodiment of the present invention is described by taking a server as an execution subject.
The embodiment of the application can acquire and process related data based on an artificial intelligence technology. Among them, artificial Intelligence (AI) is a theory, method, technique and application system that simulates, extends and expands human Intelligence using a digital computer or a machine controlled by a digital computer, senses the environment, acquires knowledge and uses the knowledge to obtain the best result.
The artificial intelligence base technologies generally include technologies such as sensors, dedicated artificial intelligence chips, cloud computing, distributed storage, big data processing technologies, operation/interaction systems, mechatronics, and the like. The artificial intelligence software technology mainly comprises a computer vision technology, a robot technology, a biological recognition technology, a voice processing technology, a natural language processing technology, machine learning/deep learning and the like.
In this embodiment, the capturing device herein refers to a device that can be used to capture a corresponding video, and includes cameras of various functional types; the monitored area coordinate set refers to areas needing special monitoring in corresponding entrances and exits or other construction areas, and corresponding coordinate positions of the areas in the acquisition equipment are marked, so that the monitored area coordinate set is obtained.
In practical application, the parameter setting data of the monitoring area in the background is called through the plurality of acquisition devices connected in the IOT platform, so that the coordinate set of the monitoring area corresponding to the directional area of each acquisition device is obtained, the acquisition devices are connected through the Internet of things technology, and each acquisition device is controlled to acquire the monitoring video data of the construction work area where the acquisition device is located.
102. Performing image framing processing on the monitoring video through the IOT platform to obtain a plurality of initial monitoring images, detecting abnormal areas in the monitoring images through the safety monitoring module, and extracting corresponding abnormal coordinates of the abnormal areas in the monitoring images;
in this embodiment, the IOT platform refers to an internet of things platform, and refers to an internet of things platform that collects and monitors object state information through various cameras and other sensor devices, and communicates the object state information with the internet through various network connection means, so as to realize intelligent sensing, monitoring and management of articles and processes; the method comprises the steps that a monitoring video shot by the networked monitoring camera equipment is obtained through the IOT platform, and then image framing processing is carried out on the monitoring video, so that a corresponding construction area is monitored in real time; the abnormal area refers to a corresponding area where an abnormal object (such as a person or other living things) enters and exits abnormally in the construction import and export area, wherein the construction import and export area is subjected to abnormal monitoring, namely the import and export of the construction import and export area can be monitored abnormally in a construction stage or a non-construction stage; the abnormal coordinates here refer to the relative coordinate positions of the abnormal creatures in the monitored area in the whole picture.
In practical application, the IOT platform establishes corresponding single-thread framing channels for data processing for each acquisition device; performing image framing on each monitoring video by using a single-thread framing channel, and performing data stream encapsulation on the framed images by using a data transmission mode (such as json data stream) to obtain a plurality of initial monitoring images; selecting a preset number of monitoring images in the same acquisition equipment to combine into a corresponding image set through a safety monitoring module to obtain a plurality of groups of primary monitoring image sets, and performing image data enhancement processing on each primary monitoring image set, for example, splicing images in each primary monitoring image set in a mode of random scaling, random cutting and random arrangement, so as to obtain a plurality of groups of enhanced monitoring image sets; further, element slicing is conducted on each monitoring image set, for example, various non-blank areas in the images are segmented, weighted average processing of preset pixel sizes is conducted on results of the element slicing, convolution calculation is conducted on pixel features in the results, and a plurality of monitoring feature graphs are obtained; then, carrying out image grid division on each monitoring feature map, and carrying out abnormal feature recognition on the monitoring feature maps of each grid after division to obtain an abnormal recognition result; respectively calculating the abnormal probability of the corresponding image grids in the abnormal recognition result, selecting the image grids exceeding the preset abnormal probability, and combining the image grids to generate an abnormal region; further determining a plurality of abnormal image grids corresponding to the abnormal regions; then, based on a preset characteristic image coordinate axis, identifying a plurality of grid coordinate points of each abnormal image grid; therefore, the coordinate points of the area boundary in the grid coordinate points are selected to obtain abnormal coordinates.
103. Detecting whether the abnormal coordinates are coordinate areas corresponding to the coordinate set of the monitored area through a safety monitoring module, and performing grid division and abnormal category probability prediction on the abnormal areas according to the coordinate detection result to obtain an abnormal identification result;
in this embodiment, the anomaly category probability refers to an accuracy rate of corresponding determination of a very frequent in-and-out life type or other types of abnormal situations in the presence of monitoring anomalies.
In practical application, a safety monitoring module is used for detecting whether the current abnormal coordinate is a part of coordinates in a coordinate area corresponding to a monitored area coordinate set or not, namely whether the abnormal coordinate and an area formed by the abnormal coordinate belong to a part of monitored area coordinate set subset or intersection or not, and then according to the result of the coordinate detection, if the abnormal coordinate meets the requirement, the predicted value of a boundary frame of each grid in the abnormal area is calculated according to the result of the coordinate detection; then, performing coordinate calculation on the object in each grid in the abnormal area by using a full-connection layer in a preset construction monitoring model, and detecting whether the result of the coordinate calculation belongs to a predicted value of the bounding box; and if the abnormal type identification result belongs to the abnormal type identification result, performing confidence calculation on the result of the coordinate calculation and the predicted value of each bounding box, and identifying the condition that the abnormal organism enters and exits and the corresponding organism type to obtain the abnormal identification result corresponding to the abnormal type accuracy. And if the abnormal coordinates do not belong to the abnormal monitoring area, analyzing and processing the abnormal conditions of the abnormal monitoring area corresponding to the abnormal coordinates by using a construction monitoring model, analyzing whether the current abnormal conditions belong to other extra abnormal construction states (such as abnormal conditions that construction building materials fall off, workers do not perform construction according to regulations, fire and explosion happen in the construction rest stage) and generating corresponding abnormal type identification results.
104. And matching corresponding alarm events to the abnormal objects in the abnormal area through the IOT platform based on the abnormal identification result, generating alarm information according to the alarm events through the ispM platform, and pushing the alarm information to the corresponding monitoring terminal.
In this embodiment, the alarm event refers to analyzing the corresponding construction violation condition in the abnormal area after analyzing the monitored image, so as to obtain a construction alarm event corresponding to the construction violation condition; through the analysis of the construction abnormal area, the construction alarm event corresponding to the abnormal condition in the current monitoring construction area can be known, so that the corresponding alarm information can be generated according to the corresponding alarm event, and corresponding workers are reminded to patrol and improve. The ispM platform (intelligent engineering project management platform) refers to a platform for realizing business functions for a single project, and has the advantages of the functions of a process control desk account (the whole process control desk account of an engineering is clear and visible), an engineering review workflow form (engineering on-line review, flexible and convenient), a plurality of collaborative work forms of participating units (on-line multi-party collaboration, high efficiency and convenience), an ispM archive center (on-line file management, easy storage and easy checking) and a whole set of construction project form (which can be directly applied and improves the efficiency). The corresponding alarm information is generated for the corresponding alarm event, and the alarm information is archived and sent to the corresponding staff for processing by utilizing the isPM platform.
In practical application, based on an abnormal recognition result, carrying out abnormal in-and-out labeling on illegal organisms in an abnormal construction area in a monitored image; generating an alarm image set according to the marked monitoring image, calling back the alarm image set to the IOT platform, and generating an alarm event corresponding to the abnormal access type according to the called-back alarm image set through the IOT platform; selecting a corresponding data format type according to a monitoring scene corresponding to the abnormal in-out alarm event through the ispM platform, and performing preprocessing on the alarm event; according to the preprocessing result, generating corresponding alarm content according to the abnormal in-and-out personnel or organisms and the corresponding prediction condition of the abnormal in-and-out personnel or organisms, and generating a plurality of alarm information by using the alarm content by using a preset monitoring object template; and pushing each alarm message to the corresponding monitoring terminal.
In the embodiment of the invention, the method is applied to a safety monitoring system of a construction area, a monitoring area coordinate set of a directional area corresponding to each acquisition device is obtained, and a monitoring video corresponding to the construction area is acquired through the acquisition devices; performing image framing processing on the monitoring video through the IOT platform to obtain a plurality of initial monitoring images, detecting abnormal areas in the monitoring images through the safety monitoring module, and extracting corresponding abnormal coordinates of the abnormal areas in the monitoring images; detecting whether the abnormal coordinates are coordinate areas corresponding to the coordinate set of the monitored area through a safety monitoring module, and performing grid division and abnormal category probability prediction on the abnormal areas according to the coordinate detection result to obtain an abnormal identification result; and matching corresponding alarm events to the abnormal objects in the abnormal area through the IOT platform based on the abnormal identification result, generating alarm information according to the alarm events through the ispM platform, and pushing the alarm information to the corresponding monitoring terminal. Compared with the prior art, the method and the device have the advantages that the video of the corresponding import and export area is collected through the collection device, the collected video is subjected to framing and abnormal coordinate matching, and then the abnormal area is identified in abnormal categories, so that corresponding early warning information is generated according to an abnormal identification result and pushed. The real-time detection of the fixed-point acquisition area in the construction entrance and exit area is realized, the corresponding early warning information is generated according to the abnormal type with abnormal intrusion and is pushed, and the early warning efficiency of safety monitoring of the entrance and exit enclosed in the construction field is improved.
Referring to fig. 2, a second embodiment of the method for monitoring the safety of the construction area according to the embodiment of the present invention includes:
201. acquiring a monitoring area coordinate set of a directional area corresponding to each acquisition device, and acquiring a monitoring video corresponding to a construction area through the acquisition devices;
202. establishing a single-thread framing channel for each acquisition device through an IOT platform;
in this embodiment, the single-thread framing channel refers to establishing a corresponding single-thread processing program channel for each acquisition device, so as to implement individual preliminary processing of the acquired data of each acquisition device, and accelerate the processing efficiency of the whole data, so that the back-end control system can only perform exception identification processing on the monitoring area with exceptions.
In practical application, each acquisition device is single-threaded bound through the IOT platform, so that a single-threaded framing channel is established for each acquisition device.
203. Performing image framing on each monitoring video by using a single-thread framing channel, and performing data stream packaging on the framed images to obtain a plurality of initial monitoring images;
in this embodiment, the image framing refers to image capturing of a video according to a preset time interval; here, the data stream encapsulation means that the captured picture is transmitted by using a corresponding data transmission method to generate a corresponding transmission data stream.
In practical application, according to a preset image capturing time interval, performing image framing processing on a corresponding monitoring video by using a single-thread framing channel, and packaging a data stream of a picture obtained by screenshot and obtained by the same acquisition device by using a json technology to obtain a plurality of initial monitoring images;
204. selecting a preset number of monitoring images in the same acquisition equipment through a safety monitoring module to obtain a plurality of groups of primary monitoring image sets, and performing image data enhancement processing on each primary monitoring image set to obtain a plurality of groups of enhanced monitoring image sets;
in this embodiment, the data enhancement means to make an original unclear image clear or emphasize some interesting features or increase the number of pictures, enlarge the difference between different object features in the image, suppress uninteresting features, improve the image quality and enrich the information content, enhance the image interpretation and recognition effects, and meet the needs of some special analyses.
In practical application, the monitoring images in the monitoring image sets are selected through the safety monitoring module according to a preset selection number (such as 4 pictures) to form a plurality of groups of primary monitoring image sets, then image data enhancement processing is carried out on each primary monitoring image set, and the primary monitoring image sets are spliced in a mode of random scaling, random cutting and random arrangement, so that the data sets are enriched, and a plurality of groups of enhanced monitoring image sets are obtained.
205. Performing element slicing on each monitoring image set, and performing weighted average processing on the result of the element slicing by using a preset pixel size to obtain a plurality of monitoring feature maps;
in this embodiment, the term "element slice" refers to a slice of a picture from a non-blank area, and the term "weighted average" refers to a process of performing multiple convolution calculations using a convolution kernel.
In practical application, according to the processed monitoring image set, identifying the non-blank corresponding area of each monitoring image set, further carrying out segmentation processing on different non-blank areas, further carrying out convolution calculation on the result of element slicing by using 32 convolution kernels, for example, the original monitoring image 608 × 3 is input into the convolution layer structure, and is first changed into a feature map of 304 × 12, and then is subjected to a convolution operation of 32 convolution kernels, and finally is changed into a feature map of 304 × 32, so as to obtain a plurality of monitoring feature maps.
206. Carrying out image grid division on each monitoring feature map, and carrying out abnormal feature recognition on the monitoring feature maps of each grid after division to obtain an abnormal recognition result;
in this embodiment, the image mesh division is performed on each monitoring feature map obtained through the processing, the monitoring feature maps are divided into a preset number of meshes through the full connection layer, then the abnormal feature recognition is performed on the monitoring feature maps of the divided meshes, and the abnormal recognition result is obtained by recognizing elements which do not belong to the normal monitoring image. Where the grid is a corresponding convolution grid.
207. Respectively calculating the abnormal probability of the corresponding image grids in the abnormal recognition result, selecting the image grids exceeding the preset abnormal probability, and generating abnormal areas;
in this embodiment, the abnormal probabilities of the corresponding image grids in the abnormal recognition result are respectively calculated, that is, whether the abnormal recognition result is the abnormal probability corresponding to the abnormal element is respectively calculated, and then the image grids exceeding the preset abnormal probability are selected, and the grid images are combined and merged into the abnormal region.
208. Determining a plurality of abnormal image grids corresponding to the abnormal regions;
in the present embodiment, a plurality of abnormal image grids corresponding to the abnormal region in the entire monitored image are determined, where the abnormal image grids are grids of corresponding coordinate axes.
209. Identifying a plurality of grid coordinate points of each abnormal image grid based on a preset characteristic image coordinate axis;
in this embodiment, based on a preset feature image coordinate axis, the relative value positions of the abnormal image grids in the overall feature image coordinate axis are identified, and a plurality of grid coordinate points are obtained.
210. Selecting a coordinate point of a region boundary in grid coordinate points to obtain an abnormal coordinate;
in this embodiment, the abnormal coordinates are obtained by selecting the coordinate points corresponding to the area boundary from the grid coordinate points and using the boundary coordinate points as the coordinate points of the entire abnormal area.
211. Detecting whether the abnormal coordinates are coordinate areas corresponding to the coordinate set of the monitored area through a safety monitoring module, and performing grid division and abnormal category probability prediction on the abnormal areas according to the coordinate detection result to obtain an abnormal identification result;
212. and matching corresponding alarm events to the abnormal objects in the abnormal area through the IOT platform based on the abnormal identification result, generating alarm information according to the alarm events through the ispM platform, and pushing the alarm information to the corresponding monitoring terminal.
In the embodiment of the invention, the method is applied to a safety monitoring system of a construction area, a monitoring area coordinate set of a directional area corresponding to each acquisition device is obtained, and a monitoring video corresponding to the construction area is acquired through the acquisition devices; the method comprises the steps of carrying out image framing processing on a monitoring video through an IOT platform to obtain a plurality of initial monitoring images, detecting abnormal areas in the monitoring images through a safety monitoring module, and extracting corresponding abnormal coordinates of the abnormal areas in the monitoring images. Compared with the prior art, the method and the device have the advantages that the video of the corresponding inlet and outlet areas is collected through the collection device, the collected video is subjected to framing and abnormal coordinate matching, and the abnormal areas are identified in abnormal categories, so that corresponding early warning information is generated according to abnormal identification results and pushed. The real-time detection of the fixed-point acquisition area in the construction entrance and exit area is realized, the corresponding early warning information is generated according to the abnormal type with abnormal intrusion and is pushed, and the early warning efficiency of safety monitoring of the entrance and exit enclosed in the construction field is improved.
Referring to fig. 3, a third embodiment of the method for monitoring the safety of the construction area according to the embodiment of the present invention includes:
301. acquiring a monitoring area coordinate set of a directional area corresponding to each acquisition device, and acquiring a monitoring video corresponding to a construction area through the acquisition devices;
302. performing image framing processing on the monitoring video through the IOT platform to obtain a plurality of initial monitoring images, detecting abnormal areas in the monitoring images through the safety monitoring module, and extracting corresponding abnormal coordinates of the abnormal areas in the monitoring images;
303. calculating the predicted value of the boundary frame of each grid in the abnormal area according to the coordinate detection result;
in this embodiment, the bounding box prediction value here refers to a bounding box value in the Yolov5s algorithm, where the bounding box includes 5 prediction values.
In practical application, whether the abnormal coordinates are coordinate areas corresponding to the coordinate set of the monitoring area is detected through the safety monitoring module, and then the predicted values of the corresponding number of the boundary frames of each grid in the abnormal area are calculated respectively according to the coordinate detection result.
304. Performing coordinate calculation on the object in each grid in the abnormal area by using a full connection layer in a preset construction monitoring model, and detecting whether the result of the coordinate calculation belongs to a predicted value of a bounding box;
in this embodiment, the construction monitoring model refers to a model of feature convolution calculation formed by using a correlation algorithm such as Yolov5s, and the model collects images shot by videos on the internet and the existing construction site through a correlation database or a webpage, so as to obtain historical monitoring images of a construction area, further remove the same images in the obtained historical monitoring images, and clean pictures irrelevant to construction area early warning, so as to obtain cleaned historical monitoring images; then, marking the washed historical monitoring pictures by using a marking tool to automatically (or through manual assistance) mark the pictures to form monitoring scene types such as 'construction operation specification', 'personnel dressing behavior specification', 'control area', 'environment', 'abnormal organism entering' and the like for classification and scene marking, so that an image training set is obtained; and then training and updating the preset pre-training model by using the monitoring scene training set, and improving the recognition precision of the scene type of the monitoring scene by repeatedly training the pre-training model algorithm, thereby obtaining the required scene monitoring model.
In practical application, the full-connection layer in the preset construction monitoring model is used for carrying out coordinate calculation on the objects in each grid in the abnormal area, namely the full-connection layer is used for calculating the coordinate values of the objects in each grid, and whether the coordinate calculation result belongs to the predicted value of the boundary box or not is detected.
305. If the abnormal recognition result belongs to the preset abnormal recognition result, performing confidence calculation on the result of the coordinate calculation and the predicted value of each boundary box to obtain an abnormal recognition result corresponding to the accuracy of the abnormal category;
in this embodiment, if the abnormal recognition result belongs to the above-mentioned category, the full-connection calculation formula of the YOLOV5 algorithm is used to perform convolution calculation on the coordinate calculation result and each boundary frame prediction value, so as to obtain the abnormal recognition result corresponding to the abnormal category accuracy. The method comprises the steps of predicting a scene type corresponding to a target by using a full connection layer in a construction monitoring model according to the abstract feature points, analyzing a problem area and calculating the probability of the biological type of the scene type to obtain a monitoring biological type and a monitoring area of the type corresponding to each monitoring image, performing abnormal biological identification on each monitoring area according to the monitoring scene type, identifying whether the monitoring area has abnormal in-out features or not by using the construction monitoring model, using the features with abnormal in-out as abnormal in-out areas, and using the features without abnormal construction as normal in-out areas to obtain abnormal identification results corresponding to the accuracy of abnormal categories, wherein the identification results comprise the abnormal in-out areas and the normal in-out areas.
306. If the abnormal recognition result contains an abnormal object, marking the abnormal object in the monitored image according to the abnormal recognition result;
in this embodiment, if the identification result is an area including an abnormal creature entering and exiting area or other construction abnormal condition areas, the abnormal area in the monitoring image is labeled according to the identification result.
307. Generating an alarm image set according to the marked monitoring image, calling back the alarm image set to the IOT platform, and generating an alarm event corresponding to the abnormal type according to the called-back alarm image set through the IOT platform;
in this embodiment, the monitoring images marked by the abnormal construction are extracted and classified, and the classified monitoring images are subjected to picture set generation to obtain a corresponding alarm image set, and then the alarm image set is called back to the IOT platform, the alarm image set is returned to the alarm ledger display of the IOT platform through HTTP web interfaces and the like, and an alarm event corresponding to an abnormal type is generated according to the called-back alarm image set through the IOT platform. The intelligent safety monitoring function of the construction area is realized.
308. The ispM platform selects a corresponding data format type according to a monitoring scene corresponding to the alarm event, and performs preprocessing on the alarm event;
in this embodiment, the format type of the data office refers to a format type of information correspondingly transmitted by different monitoring terminals.
In practical application, the ispM platform sends a monitoring terminal corresponding to monitoring information according to a monitoring scene corresponding to an alarm event, and further selects a corresponding data format type according to the monitoring terminal, so that the information is preprocessed on the alarm event. By preprocessing the alarm event, the corresponding monitoring information can be generated for different monitoring terminals according to the alarm event.
309. Generating corresponding alarm content according to the preprocessing result, and generating a plurality of alarm information by adopting the alarm content by utilizing a preset monitoring object template;
in this embodiment, the monitoring object template refers to an information generation template corresponding to information sent by different monitoring terminals or platforms.
In practical application, corresponding alarm content is generated according to the preprocessing result, and a plurality of alarm information is generated by adopting the alarm content by utilizing a preset monitoring object template.
310. And pushing each alarm message to the corresponding monitoring terminal.
In this embodiment, the generated alarm information is pushed to the corresponding monitoring terminal.
The preferred implementation manner of the isPM platform processing process is that after receiving the alarm data of the IOT platform, the isPM preprocesses the alarm event to generate corresponding alarm information, and automatically notifies the corresponding preset responsible person through short messages, public number messages and system internal messages according to different alarm types. The responsible person can log in the system to view on the alarm ledger page, or view related alarm data on APP and WeChat small program. And then the responsible person can send voice warning relevant constructors in real time aiming at unprocessed warning events and through a building site on-site broadcasting system which is in butt joint with the small programs according to actual conditions, or quickly initiate a rectification list through an APP and a micro-communication small program and attach warning pictures to instruct the relevant constructors to process, or inform criticism on a smart large screen of the ispM of the building site and instruct the corresponding constructors to pay penalties.
In the embodiment of the invention, the scene monitoring model is used for identifying the abnormal construction area of the monitoring image, the abnormal construction area is marked according to the identification result to generate the corresponding alarm event, the ispM platform is used for analyzing the alarm event, and the corresponding alarm information is generated according to different monitoring terminals to push the information. Realized to the analysis of monitoring image and report an emergency and ask for help or increased vigilance the propelling movement, through the different terminals of PC, APP, the applet that combines the ispM system, let the security personnel handle the safety alarm incident fast to realize that the file leaves the trace, the data is integrated fast, the function of flow closed loop, in addition through carrying out special mark to the image area that needs special monitoring, thereby strengthen the control to special scene. The identification efficiency of construction monitoring and the monitoring strength of special scenes are improved.
With reference to fig. 4, the method for monitoring the safety of the construction area in the embodiment of the present invention is described above, and a safety monitoring device for the construction area in the embodiment of the present invention is described below, where an embodiment of the safety monitoring device for the construction area in the embodiment of the present invention includes:
the video acquisition module 401 is configured to acquire a monitoring area coordinate set of an orientation area corresponding to each acquisition device, and acquire a monitoring video corresponding to a construction area through the acquisition device;
a coordinate extraction module 402, configured to perform image framing processing on the monitoring video through the IOT platform to obtain a plurality of initial monitoring images, detect an abnormal region in the monitoring image through the security monitoring module, and extract an abnormal coordinate corresponding to the abnormal region in the monitoring image;
an anomaly identification module 403, configured to detect, by the security monitoring module, whether the abnormal coordinate is a coordinate region corresponding to the monitored region coordinate set, and perform mesh division and abnormal category probability prediction on the abnormal region according to a result of the coordinate detection to obtain an anomaly identification result;
and an alarm pushing module 404, configured to match, by the IOT platform, a corresponding alarm event with the abnormal object in the abnormal region based on the abnormal identification result, generate alarm information according to the alarm event by the isPM platform, and push the alarm information to a corresponding monitoring terminal.
In the embodiment of the invention, the method is applied to a safety monitoring system of a construction area, a monitoring area coordinate set of a directional area corresponding to each acquisition device is obtained, and a monitoring video corresponding to the construction area is acquired through the acquisition devices; performing image framing processing on the monitoring video through the IOT platform to obtain a plurality of initial monitoring images, detecting abnormal areas in the monitoring images through the safety monitoring module, and extracting corresponding abnormal coordinates of the abnormal areas in the monitoring images; detecting whether the abnormal coordinates are coordinate areas corresponding to the coordinate set of the monitored area through a safety monitoring module, and performing grid division and abnormal category probability prediction on the abnormal areas according to the coordinate detection result to obtain an abnormal identification result; and matching corresponding alarm events to the abnormal objects in the abnormal area through the IOT platform based on the abnormal identification result, generating alarm information according to the alarm events through the ispM platform, and pushing the alarm information to the corresponding monitoring terminal. Compared with the prior art, the method and the device have the advantages that the video of the corresponding import and export area is collected through the collection device, the collected video is subjected to framing and abnormal coordinate matching, and then the abnormal area is identified in abnormal categories, so that corresponding early warning information is generated according to an abnormal identification result and pushed. The real-time detection of the fixed-point acquisition area in the construction inlet and outlet area is realized, the corresponding early warning information is generated according to the abnormal type with abnormal intrusion and is pushed, and the early warning efficiency of safety monitoring of the construction site enclosed inlet and outlet is improved.
Referring to fig. 5, another embodiment of the safety monitoring device for a construction area according to an embodiment of the present invention includes:
the video acquisition module 401 is configured to acquire a monitoring area coordinate set of an orientation area corresponding to each acquisition device, and acquire a monitoring video corresponding to a construction area through the acquisition device;
a coordinate extraction module 402, configured to perform image framing processing on the monitoring video through the IOT platform to obtain a plurality of initial monitoring images, detect an abnormal region in the monitoring image through the security monitoring module, and extract an abnormal coordinate corresponding to the abnormal region in the monitoring image;
an anomaly identification module 403, configured to detect, by the security monitoring module, whether the abnormal coordinate is a coordinate area corresponding to the monitored area coordinate set, and perform mesh division and abnormal category probability prediction on the abnormal area according to a result of the coordinate detection, to obtain an anomaly identification result;
and an alarm pushing module 404, configured to match, by the IOT platform, a corresponding alarm event with the abnormal object in the abnormal region based on the abnormal identification result, generate alarm information according to the alarm event by the isPM platform, and push the alarm information to a corresponding monitoring terminal.
Further, the coordinate extraction module 402 includes:
a channel establishing unit 4021, configured to establish a single-thread framing channel for each of the acquisition devices through the IOT platform;
the image framing unit 4022 is configured to perform image framing on each monitoring video through the single-thread framing channel, and perform data stream encapsulation on the framed images to obtain a plurality of initial monitoring images.
Further, the coordinate extraction module 402 further includes:
the image enhancement unit 4023 is configured to select, through the security monitoring module, a preset number of monitoring images in the same acquisition device to obtain multiple sets of primary monitoring image sets, and perform image data enhancement processing on each of the primary monitoring image sets to obtain multiple enhanced sets of monitoring image sets;
a weighted average unit 4024, configured to perform element slicing on each monitoring image set, and perform weighted average processing of a preset pixel size on an element slicing result to obtain a plurality of monitoring feature maps;
the feature identification unit 4025 is configured to perform image grid division on each monitoring feature map, and perform abnormal feature identification on the monitoring feature maps of each grid after division to obtain an abnormal identification result;
the probability calculation unit 4026 is configured to calculate the abnormal probabilities of the corresponding image grids in the abnormal recognition result, and select the image grids exceeding a preset abnormal probability to generate an abnormal region.
Further, the coordinate extraction module 402 further includes:
a mesh determination unit 4027 configured to determine a plurality of abnormal image meshes corresponding to the abnormal regions;
a coordinate identification unit 4028, configured to identify a plurality of grid coordinate points of each abnormal image grid based on a preset feature image coordinate axis;
the coordinate selecting unit 4029 is configured to select a coordinate point of an area boundary in the grid coordinate points to obtain an abnormal coordinate.
Further, the anomaly identification module 403 includes:
a boundary prediction unit 4031, configured to calculate a boundary frame prediction value of each grid in the abnormal region according to a result of the coordinate detection;
the coordinate calculation unit 4032 is used for performing coordinate calculation on the objects in each grid in the abnormal area by using a full-connection layer in a preset construction monitoring model, and detecting whether the result of the coordinate calculation belongs to a predicted value of a bounding box;
and an anomaly identification unit 4033, configured to perform confidence calculation on the coordinate calculation result and each of the bounding box prediction values if the coordinate calculation result belongs to the one or more bounding box prediction values, to obtain an anomaly identification result corresponding to the anomaly category accuracy.
Further, the alert pushing module 404 includes:
an anomaly labeling unit 4041, configured to label, according to the identification result, an abnormal construction area in the monitored image if the identification result includes the abnormal construction area;
the image callback unit 4042 is configured to generate an alarm image set according to the labeled monitoring image, call back the alarm image set to the IOT platform, and generate an alarm event corresponding to the abnormal type according to the called-back alarm image set through the IOT platform.
Further, the alarm pushing module 404 further includes:
a type selection unit 4043, configured to select, by the isPM platform, a corresponding data format type according to the monitoring scenario corresponding to the alarm event, and perform preprocessing on the alarm event;
an information generating unit 4044, configured to generate corresponding alarm content according to the preprocessing result, and generate multiple alarm information by using the alarm content using a preset monitoring object template;
and the alarm pushing unit 4045 is configured to push each alarm message to a corresponding monitoring terminal.
In the embodiment of the invention, the method is applied to a safety monitoring system of a construction area, a monitoring area coordinate set of a directional area corresponding to each acquisition device is obtained, and a monitoring video corresponding to the construction area is acquired through the acquisition devices; performing image framing processing on the monitoring video through the IOT platform to obtain a plurality of initial monitoring images, detecting abnormal areas in the monitoring images through the safety monitoring module, and extracting corresponding abnormal coordinates of the abnormal areas in the monitoring images; detecting whether the abnormal coordinates are coordinate areas corresponding to the coordinate set of the monitored area through a safety monitoring module, and performing grid division and abnormal category probability prediction on the abnormal areas according to the coordinate detection result to obtain an abnormal identification result; and matching corresponding alarm events to the abnormal objects in the abnormal area through the IOT platform based on the abnormal identification result, generating alarm information according to the alarm events through the ispM platform, and pushing the alarm information to the corresponding monitoring terminal. Compared with the prior art, the method and the device have the advantages that the video of the corresponding inlet and outlet areas is collected through the collection device, the collected video is subjected to framing and abnormal coordinate matching, and the abnormal areas are identified in abnormal categories, so that corresponding early warning information is generated according to abnormal identification results and pushed. The real-time detection of the fixed-point acquisition area in the construction inlet and outlet area is realized, the corresponding early warning information is generated according to the abnormal type with abnormal intrusion and is pushed, and the early warning efficiency of safety monitoring of the construction site enclosed inlet and outlet is improved.
Fig. 4 and 5 describe the safety monitoring device of the construction area in the embodiment of the present invention in detail from the perspective of the modular functional entity, and the safety monitoring device of the construction area in the embodiment of the present invention is described in detail from the perspective of hardware processing.
Fig. 6 is a schematic structural diagram of a safety monitoring device for a construction area according to an embodiment of the present invention, where the safety monitoring device 600 for a construction area may have a relatively large difference due to different configurations or performances, and may include one or more processors (CPUs) 610 (e.g., one or more processors) and a memory 620, and one or more storage media 630 (e.g., one or more mass storage devices) storing applications 633 or data 632. Memory 620 and storage medium 630 may be, among other things, transient or persistent storage. The program stored in the storage medium 630 may include one or more modules (not shown), each of which may include a series of instructional operations on the safety monitoring device 600 for a construction area. Still further, the processor 610 may be configured to communicate with the storage medium 630 to execute a series of instruction operations in the storage medium 630 on the safety monitoring device 600 at the construction area.
The safety monitoring apparatus 600 for a construction area may also include one or more power supplies 640, one or more wired or wireless network interfaces 650, one or more input-output interfaces 660, and/or one or more operating systems 631, such as Windows Server, mac OS X, unix, linux, freeBSD, and the like. Those skilled in the art will appreciate that the construction area safety monitoring device configuration shown in FIG. 6 does not constitute a limitation of the construction area safety monitoring device and may include more or fewer components than shown, or some components in combination, or a different arrangement of components.
The invention also provides a safety monitoring device for a construction area, which comprises a memory and a processor, wherein computer readable instructions are stored in the memory, and when the computer readable instructions are executed by the processor, the processor executes the steps of the safety monitoring method for the construction area in the embodiments.
The present invention also provides a computer-readable storage medium, which may be a non-volatile computer-readable storage medium, and which may also be a volatile computer-readable storage medium, having stored therein instructions, which, when run on a computer, cause the computer to perform the steps of the method for safety monitoring of a construction area.
It is clear to those skilled in the art that, for convenience and brevity of description, the specific working processes of the above-described systems, apparatuses and units may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.
The integrated unit, if implemented in the form of a software functional unit and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: various media capable of storing program codes, such as a usb disk, a removable hard disk, a read-only memory (ROM), a Random Access Memory (RAM), a magnetic disk, or an optical disk.
The application is operational with numerous general purpose or special purpose computing system environments or configurations. For example: personal computers, server computers, hand-held or portable devices, tablet-type devices, multiprocessor systems, microprocessor-based systems, set top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, distributed computing environments that include any of the above systems or devices, and the like. The application may be described in the general context of computer-executable instructions, such as program modules, being executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform particular tasks or implement particular abstract data types. The application may also be practiced in distributed computing environments where tasks are performed by remote processing devices that are linked through a communications network. In a distributed computing environment, program modules may be located in both local and remote computer storage media including memory storage devices.
The above-mentioned embodiments are only used for illustrating the technical solutions of the present invention, and not for limiting the same; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.

Claims (10)

1. A safety monitoring method of a construction area is applied to a safety monitoring system of the construction area, and is characterized in that the safety monitoring system comprises a safety monitoring module, an IOT platform and an ispM platform, wherein the IOT platform comprises a plurality of acquisition devices, and the safety monitoring method of the construction area comprises the following steps:
acquiring a monitoring area coordinate set of the directional area corresponding to each acquisition device, and acquiring a monitoring video corresponding to the construction area through the acquisition device;
performing image framing processing on the monitoring video through the IOT platform to obtain a plurality of initial monitoring images, detecting abnormal areas in the monitoring images through the safety monitoring module, and extracting corresponding abnormal coordinates of the abnormal areas in the monitoring images;
detecting whether the abnormal coordinates are coordinate areas corresponding to the monitored area coordinate set or not through the safety monitoring module, and if so, performing abnormal category probability prediction on the abnormal areas to obtain abnormal identification results;
and matching a corresponding alarm event for the abnormal object in the abnormal area through the IOT platform based on the abnormal identification result, generating alarm information according to the alarm event through the ispM platform, and pushing the alarm information to a corresponding monitoring terminal.
2. The safety monitoring method for a construction area according to claim 1, wherein the obtaining a plurality of initial monitoring images by performing image framing processing on the monitoring video through the IOT platform comprises:
establishing a single-thread framing channel for each acquisition device through the IOT platform;
and performing image framing on each monitoring video by using the single-thread framing channel, and performing data stream packaging on the framed images to obtain a plurality of initial monitoring images.
3. The safety monitoring method for a construction area according to claim 1, wherein the detecting of the abnormal area in the monitoring image by the safety monitoring module includes:
selecting a preset number of monitoring images in the same acquisition device through the safety monitoring module to obtain a plurality of groups of primary monitoring image sets, and performing image data enhancement processing on each primary monitoring image set to obtain a plurality of groups of enhanced monitoring image sets;
performing element slicing on each enhanced monitoring image set, and performing weighted average processing on the result of the element slicing by preset pixel size to obtain a plurality of monitoring characteristic graphs;
carrying out image grid division on each monitoring feature map, and carrying out abnormal feature identification on the monitoring feature maps of each grid after division to obtain a plurality of abnormal image grids;
and respectively calculating the abnormal probability corresponding to each abnormal image grid, and selecting the abnormal image grid exceeding the preset abnormal probability based on the result of the abnormal probability calculation to generate an abnormal region.
4. The method for monitoring the safety of the construction area according to claim 3, wherein the extracting the corresponding abnormal coordinates of the abnormal area in the monitoring image comprises:
determining a plurality of abnormal image grids corresponding to the abnormal regions;
identifying a plurality of grid coordinate points of each abnormal image grid based on a preset characteristic image coordinate axis;
and selecting coordinate points belonging to the boundary of the abnormal area from the grid coordinate points to obtain abnormal coordinates.
5. The safety monitoring method for the construction area according to claim 3, wherein if the abnormal area is detected, performing abnormal type probability prediction on the abnormal area to obtain an abnormal recognition result, and the method comprises:
if yes, calculating a boundary frame predicted value of each abnormal image grid in the abnormal area;
performing coordinate calculation on the object in each abnormal image grid in the abnormal area by using a full-connection layer in a preset construction monitoring model, and detecting whether the result of the coordinate calculation belongs to a predicted value of a bounding box;
and if the abnormal recognition result belongs to the abnormal recognition result, performing confidence calculation on the coordinate calculation result and each boundary box predicted value to obtain the abnormal recognition result corresponding to the abnormal category accuracy.
6. The method for monitoring the safety of the construction area according to claim 1, wherein the matching, based on the abnormal recognition result, of the corresponding alarm event to the abnormal object of the abnormal area through the IOT platform comprises:
if the abnormal recognition result contains an abnormal object, marking the abnormal object in the monitoring image according to the abnormal recognition result;
and generating an alarm image set according to the labeled monitoring image, calling back the alarm image set to the IOT platform, and generating an alarm event corresponding to the abnormal type according to the called-back alarm image set through the IOT platform.
7. The safety monitoring method for the construction area according to claim 1, wherein the generating of the alarm information according to the alarm event through the isPM platform and the pushing of the alarm information to the corresponding monitoring terminal comprise:
the ispM platform selects a corresponding data format type according to a monitoring scene corresponding to the alarm event, and performs preprocessing on the alarm event;
generating corresponding alarm content according to the preprocessing result, and generating a plurality of alarm information by adopting the alarm content by utilizing a preset monitoring object template;
and pushing each warning message to a corresponding monitoring terminal.
8. The utility model provides a construction area's safety monitoring device, is applied to construction area's safety monitoring system, a serial communication port, safety monitoring system includes safety monitoring module, IOT platform and ispM platform, wherein, the IOT platform includes a plurality of collection equipment, construction area's safety monitoring device includes:
the video acquisition module is used for acquiring a monitoring area coordinate set of the directional area corresponding to each acquisition device and acquiring a monitoring video corresponding to the construction area through the acquisition device;
the coordinate extraction module is used for performing image framing processing on the monitoring video through the IOT platform to obtain a plurality of initial monitoring images, detecting abnormal areas in the monitoring images through the safety monitoring module and extracting corresponding abnormal coordinates of the abnormal areas in the monitoring images;
the abnormal recognition module is used for detecting whether the abnormal coordinates are coordinate regions corresponding to the monitored region coordinate set through the safety monitoring module, and if so, performing abnormal type probability prediction on the abnormal regions to obtain abnormal recognition results;
and the alarm pushing module is used for matching a corresponding alarm event to the abnormal object in the abnormal area through the IOT platform based on the abnormal identification result, generating alarm information according to the alarm event through the ispM platform, and pushing the alarm information to a corresponding monitoring terminal.
9. A safety monitoring device for a construction area, the safety monitoring device comprising: a memory and at least one processor, the memory having instructions stored therein;
the at least one processor invoking the instructions in the memory to cause a safety monitoring device of the construction area to perform the steps of the method of safety monitoring of a construction area of any of claims 1-7.
10. A computer-readable storage medium having instructions stored thereon, wherein the instructions, when executed by a processor, implement the steps of a method for safety monitoring a construction area according to any one of claims 1-7.
CN202211095238.8A 2022-09-08 2022-09-08 Safety monitoring method, device and equipment for construction area and storage medium Active CN115190277B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202211095238.8A CN115190277B (en) 2022-09-08 2022-09-08 Safety monitoring method, device and equipment for construction area and storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202211095238.8A CN115190277B (en) 2022-09-08 2022-09-08 Safety monitoring method, device and equipment for construction area and storage medium

Publications (2)

Publication Number Publication Date
CN115190277A CN115190277A (en) 2022-10-14
CN115190277B true CN115190277B (en) 2022-12-13

Family

ID=83523372

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202211095238.8A Active CN115190277B (en) 2022-09-08 2022-09-08 Safety monitoring method, device and equipment for construction area and storage medium

Country Status (1)

Country Link
CN (1) CN115190277B (en)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116502897B (en) * 2023-06-26 2023-10-27 中兴软件技术(南昌)有限公司 Method and system for monitoring infrastructure safety based on image recognition

Family Cites Families (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH07288802A (en) * 1994-04-18 1995-10-31 Nippon Telegr & Teleph Corp <Ntt> Intruded object supervisory equipment
JP2004234324A (en) * 2003-01-30 2004-08-19 Nisshin Steel Co Ltd Method for monitoring facility
JP6016295B2 (en) * 2012-08-10 2016-10-26 株式会社日立国際電気 Monitoring system
CN107230310A (en) * 2017-06-26 2017-10-03 地壳机器人科技有限公司 Relay-type monitoring method and pilotless automobile
CN109784672A (en) * 2018-12-25 2019-05-21 上海交通大学 A kind of warning system for real time monitoring and method for power grid exception
CN110866908B (en) * 2019-11-12 2021-03-26 腾讯科技（深圳）有限公司 Image processing method, image processing apparatus, server, and storage medium
CN111191631A (en) * 2020-01-08 2020-05-22 哈尔滨融智爱科智能科技有限公司 Video monitoring supervision area multi-device linkage method based on image domain template matching
CN111444854B (en) * 2020-03-27 2021-03-02 科大讯飞(苏州)科技有限公司 Abnormal event detection method, related device and readable storage medium
CN111680556B (en) * 2020-04-29 2024-06-07 平安国际智慧城市科技股份有限公司 Method, device, equipment and storage medium for identifying traffic gate vehicle type
US11887358B2 (en) * 2020-06-10 2024-01-30 Ancestry.Com Operations Inc. Systems and methods for identifying and segmenting objects from images
CN113837001A (en) * 2021-08-18 2021-12-24 中国科学院自动化研究所 Method and device for detecting abnormal intruding object in real time under monitoring scene
CN114390260A (en) * 2022-01-26 2022-04-22 中煤科工集团重庆智慧城市科技研究院有限公司 Hazardous area and important place monitoring platform applied to smart city
CN114973140A (en) * 2022-06-10 2022-08-30 广西北投公路建设投资集团有限公司 Dangerous area personnel intrusion monitoring method and system based on machine vision

Also Published As

Publication number Publication date
CN115190277A (en) 2022-10-14

Similar Documents

Publication Publication Date Title
CN107679471B (en) Indoor personnel air post detection method based on video monitoring platform
CN110659391A (en) Video detection method and device
CN115190277B (en) Safety monitoring method, device and equipment for construction area and storage medium
CN110889339A (en) Head and shoulder detection-based dangerous area grading early warning method and system
CN112819068A (en) Deep learning-based real-time detection method for ship operation violation behaviors
CN113807240A (en) Intelligent transformer substation personnel dressing monitoring method based on uncooperative face recognition
CN116311081B (en) Medical laboratory monitoring image analysis method and system based on image recognition
CN111178424A (en) Petrochemical production site safety compliance real-time detection system and method
CN115099724A (en) Monitoring and early warning method, device and equipment for construction scene and storage medium
CN111860187A (en) High-precision worn mask identification method and system
CN114926778A (en) Safety helmet and personnel identity recognition system under production environment
Sharadhi et al. Face mask recogniser using image processing and computer vision approach
CN111191498A (en) Behavior recognition method and related product
CN117523806A (en) Fire disaster early warning method, system, equipment and storage medium
CN111652128B (en) High-altitude power operation safety monitoring method, system and storage device
CN113239766A (en) Behavior recognition method based on deep neural network and intelligent alarm device
CN116665419B (en) Intelligent fault early warning system and method based on AI analysis in power production operation
CN112528825A (en) Station passenger recruitment service method based on image recognition
CN114764895A (en) Abnormal behavior detection device and method
CN111582183A (en) Mask identification method and system in public place
CN110751125A (en) Wearing detection method and device
CN111723725A (en) Multi-dimensional analysis system based on video AI
CN116977900A (en) Intelligent laboratory monitoring alarm system and method thereof
CN115829324A (en) Personnel safety risk silent monitoring method
CN115578731A (en) Method and device for generating safety control information, electronic equipment and storage medium

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