CN115314685A - AI video structured analysis-based hazardous chemical substance availability management system and method - Google Patents
AI video structured analysis-based hazardous chemical substance availability management system and method Download PDFInfo
- Publication number
- CN115314685A CN115314685A CN202211238562.0A CN202211238562A CN115314685A CN 115314685 A CN115314685 A CN 115314685A CN 202211238562 A CN202211238562 A CN 202211238562A CN 115314685 A CN115314685 A CN 115314685A
- Authority
- CN
- China
- Prior art keywords
- video
- hazardous chemical
- chemical substance
- personnel
- detected
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
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
- H04N7/181—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast for receiving images from a plurality of remote sources
-
- 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
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/172—Classification, e.g. identification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/20—Movements or behaviour, e.g. gesture recognition
- G06V40/23—Recognition of whole body movements, e.g. for sport training
- G06V40/25—Recognition of walking or running movements, e.g. gait recognition
-
- G—PHYSICS
- G07—CHECKING-DEVICES
- G07C—TIME OR ATTENDANCE REGISTERS; REGISTERING OR INDICATING THE WORKING OF MACHINES; GENERATING RANDOM NUMBERS; VOTING OR LOTTERY APPARATUS; ARRANGEMENTS, SYSTEMS OR APPARATUS FOR CHECKING NOT PROVIDED FOR ELSEWHERE
- G07C1/00—Registering, indicating or recording the time of events or elapsed time, e.g. time-recorders for work people
- G07C1/10—Registering, indicating or recording the time of events or elapsed time, e.g. time-recorders for work people together with the recording, indicating or registering of other data, e.g. of signs of identity
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B31/00—Predictive alarm systems characterised by extrapolation or other computation using updated historic data
Abstract
The invention discloses a dangerous chemical article availability management system and method based on AI video structured analysis, wherein the system comprises a management terminal, a video processing module, a plurality of cameras, a user terminal and a card punching device, wherein the management terminal is used for collecting personnel information with the authority of contacting the dangerous chemical article, and authorizing and differentiating the authority of the personnel for division of labor; the card punching equipment is arranged in the hazardous chemical storage warehouse, and a claimant performs card punching confirmation through a user terminal before the claimant; after the adoption is checked, the video processing module carries out video structuralization analysis and detection on video frames to be detected, camera pictures along the transportation line of the hazardous chemical substances are used as key points to detect the video frames to be detected, whether personnel entering the hazardous chemical substance storage warehouse are confirmed in the adoption application event is analyzed and identified, the detection process is finished after the personnel are used for transporting the hazardous chemical substances to a use workshop and completing checking the card, and the detection result of the video frames to be detected is sent to the management terminal in a system early warning mode.
Description
Technical Field
The invention belongs to the technical field of computer vision, and particularly relates to a dangerous chemical article availability management system and method based on AI video structured analysis.
Background
At present, for monitoring the flow direction of hazardous chemicals enterprises, a main management mode comprises offline paper ledger registration and a software management system mode. And (3) off-line paper ledger registration, namely, recording paper-pen records of the dangerous chemical articles entering and exiting the warehouse every time, recording the time of entering and exiting the warehouse every time, the personnel to be accepted, the field custodians and the managers as well as the types and the quantity of the dangerous chemical articles, signing by enterprise managers or security personnel, and using the records as credit endorsements of the dangerous chemical articles for accepting and exiting the warehouse. The software management system is a mainstream technical scheme at present, and system registration is carried out on the information related to the dangerous chemical substance in and out of the warehouse every time in a pure software mode, so that an online account for dangerous chemical substance in and out of the warehouse is formed.
The existing software management system technology essentially transfers the function of the paper standing book to the online, the software management platform realizes the online registration of the hazardous chemical substance standing book, and managers can directly check the hazardous chemical substance in and out of the warehouse standing book of the enterprise online without checking the paper standing book of the enterprise. However, the supervision of software management on the hazardous chemical substance adoption still remains in a posterior recording method based on the autonomy, and in the actual operation, whether the warehouse entry and exit operation is carried out according to the standard requirements is unknown, so that an obvious management vulnerability exists.
Disclosure of Invention
In view of the above technical problems, the present invention is to provide a hazardous chemical substance availability management system and method based on AI video structured analysis, which are used to improve the security of hazardous chemical substance availability.
In order to solve the technical problem, the invention adopts the following technical scheme:
the embodiment of the invention provides a dangerous chemical article availability management system based on AI video structured analysis, which comprises a management terminal, a video processing module, a plurality of cameras arranged along the transportation line of the dangerous chemical articles, a user terminal and a card punching device,
the management terminal is used for collecting personnel information with the authority of contacting hazardous chemicals, and authorizing and authority division of personnel;
the card punching equipment is arranged in the hazardous chemical storage warehouse, and a claimant performs card punching confirmation through a user terminal before the claimant;
after the collar is punched, the video processing module carries out video structuralization analysis and detection on video frames to be detected, the video frames to be detected are taken as key video frames to be detected by cameras along the transportation line of the hazardous chemical substances, whether personnel entering the hazardous chemical substance storage warehouse are confirmed in the collar application event is analyzed and identified, and in the process of transporting the hazardous chemical substances, whether a hazardous chemical substance container leaves a monitoring picture is detected by the personnel, the detection process is finished after the hazardous chemical substances are transported to a use workshop by the personnel until the personnel are transported, the punching is finished, and the detection results of the video frames to be detected are sent to the management terminal in a system early warning mode.
In one possible design, the video processing module performs incremental machine learning through the monitoring camera, performs gait feature analysis on personnel in a monitoring range, and confirms the personnel ID corresponding to the gait feature in a management terminal marking mode.
In one possible design, the video processing module identifies the person ID based on the face and gait features when the video structured analysis event is received.
In one possible design, when an unlearned application event is triggered, a camera picture arranged in a dangerous chemical storage warehouse is used as a video frame to be detected of a video frame to be detected, a video processing module carries out real-time detection on the frame to be detected, whether unauthorized people break in or not is detected, and if the unauthorized people break in, an early warning message is sent to a management terminal.
In one possible design, under the condition that the acceptance start card punching signal is triggered and the acceptance completion card punching signal is not triggered, the video processing module takes a picture of a storage camera of hazardous chemical substances as a video to be analyzed, frame-by-frame analysis is carried out in the time period, whether personnel of a target main body are consistent with the personnel appointed for acceptance is detected, and if the personnel lack or the information of the personnel is inconsistent, an early warning message is sent to the management terminal.
In one possible design, the taking a camera picture along the transportation line of the hazardous chemical substance as a key video frame to be detected includes: storing all camera information in a directed acyclic graph DAG, wherein each camera is recorded as a node in the DAG; according to a preset route for hazardous chemical articles, the cameras covered by the route establish DAG paths between two adjacent camera nodes according to the passing sequence of the route.
In one possible design, when a route is newly created, BFS is searched for DAG with breadth first, and no cyclic path appears in a DAG data structure.
In one possible design, the hazardous chemical substance warehouse is bound with a DAG path, when an analysis event is triggered, a camera list corresponding to the DAG path is found out according to the binding relation, and video characteristics in camera video streams on the path are sequentially analyzed according to a time sequence.
In one possible design, the structural characteristics of the video structural analysis comprise the transportation track of the hazardous chemical container, the state of the hazardous chemical container, smoke, flame and personnel information.
In another aspect, an embodiment of the present invention provides a hazardous chemical substance availability management method based on AI video structured analysis, including:
the method comprises the steps that a hazardous chemical substance receiving management system which comprises a management terminal, a video processing module, a plurality of cameras arranged along a hazardous chemical substance transportation line, a user terminal and a card punching device and is based on AI video structural analysis is arranged, wherein the management terminal is used for collecting personnel information with the authority of contacting the hazardous chemical substance, and authorizing and distinguishing the authority of the personnel for division of labor; the card punching equipment is arranged in the hazardous chemical storage warehouse, and a claimant performs card punching confirmation through a user terminal before the claimant;
after the adoption is checked, the video processing module carries out video structured analysis and detection on video frames to be detected, camera pictures along the transportation line of the hazardous chemical substances are used as key video frames to be detected, whether personnel entering a hazardous chemical substance storage warehouse are confirmed in the adoption application event is analyzed and identified, in the process of transporting the hazardous chemical substances, whether a hazardous chemical substance container leaves a monitoring picture is judged, the detection process is finished after the hazardous chemical substances are transported to a use workshop by the leading personnel and the checking is finished, and the detection results of the video frames to be detected are sent to the management terminal in a system early warning mode.
In one possible design, the video processing module is arranged to perform incremental machine learning through the monitoring camera, perform gait feature analysis on personnel in a monitoring range, and confirm the personnel ID corresponding to the gait feature in a management terminal marking mode.
In one possible design, the video processing module is configured to identify the person ID based on the face and gait characteristics when receiving the video structured analysis event.
In one possible design, when an unlearned application event is triggered, a camera picture arranged in a dangerous chemical storage warehouse is used as a video frame to be detected of a video frame to be detected, a video processing module carries out real-time detection on the frame to be detected, whether unauthorized people break in or not is detected, and if the unauthorized people break in, an early warning message is sent to a management terminal.
In one possible design, under the condition that the acceptance start card punching signal is triggered and the acceptance completion card punching signal is not triggered, the video processing module takes a picture of a storage camera of hazardous chemical substances as a video to be analyzed, frame-by-frame analysis is carried out in the time period, whether personnel of a target main body are consistent with the personnel appointed for acceptance is detected, and if the personnel lack or the information of the personnel is inconsistent, an early warning message is sent to the management terminal.
In one possible design, the taking a camera picture along the transportation line of the hazardous chemical substance as a key video frame to be detected includes: storing all camera information in a directed acyclic graph DAG, wherein each camera is recorded as a node in the DAG; according to a preset route for hazardous chemical substance receiving, cameras covered by the route establish a DAG path between two adjacent camera nodes according to the sequence of the route.
In one possible design, when a route is newly created, the BFS is searched for the DAG in a breadth-first mode, and no circular path appears in a DAG data structure.
In one possible design, the hazardous chemical warehouse is bound with a DAG path, when an analysis event is triggered, a camera list corresponding to the DAG path is found out according to the binding relation, and video features in camera video streams on the path are sequentially analyzed according to the time sequence.
In one possible design, the structural features of the video structural analysis include hazardous chemical container transportation trajectories, hazardous chemical container states, smoke, flame and personnel information.
The invention has the following beneficial effects:
(1) The personnel information with the access to hazardous chemical substances is collected through the management terminal, the personnel labor division is authorized and the access is distinguished, so that the flow can be better standardized, and the omission on the personnel is avoided.
(2) By setting the receiving event trigger of the user terminal and receiving and punching, the operation of the receiver can be standardized, and the safety and the alertness of the receiver can be improved.
(3) Through a plurality of cameras that set up along the line at dangerous chemicals transportation, and set up through the DAG route, can only carry out the analysis to the camera video stream in the relevant route of incident, reducible video processing module analysis's work load to reduce the calculation power consumption, improve video frame analysis efficiency.
(4) The personnel ID is confirmed based on gait recognition and human faces, and the accuracy of personnel recognition can be improved.
(5) The system can be used for receiving dangerous chemicals from a warehouse until the video structural analysis and early warning of operation standardization during transportation to a use workshop, and the realization that the dangerous chemicals can be accurately, efficiently and traceably carried out comprises whether field personnel meet the standard requirements or not and whether the transportation operation meets the specified management requirements or not.
Drawings
Fig. 1 is a schematic structural diagram of a hazardous chemical substance availability management system based on AI video structural analysis according to an embodiment of the present invention;
fig. 2 is a flowchart of steps of a hazardous chemical substance availability management method based on AI video structured analysis according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, not all, embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Referring to fig. 1, a schematic structural diagram of a hazardous chemical substance availability management system based on AI video structural analysis according to an embodiment of the present invention includes a management terminal 10, a video processing module 50, a plurality of cameras 40 arranged along a hazardous chemical substance transportation line, a user terminal 20, and a card swiping device 30, where the management terminal 10 is configured to collect information of a person having an authority to contact the hazardous chemical substance, and authorize and distinguish division of authority of the person; the card punching equipment 30 is arranged in a dangerous chemical storage warehouse, and a claimant confirms the card punching through the user terminal 20 before the claimant; after the adoption is started, the video processing module 50 carries out video structuralization analysis and detection on video frames to be detected, the pictures of the cameras 40 along the transportation line of the hazardous chemical substances are used as key video frames to be detected, whether personnel entering the hazardous chemical substance storage warehouse are confirmed in the adoption application event is analyzed and identified, and whether the hazardous chemical substance container leaves a monitoring picture is carried out during the transportation of the hazardous chemical substances by the personnel, the detection process is finished after the hazardous chemical substances are transported to a use workshop by the personnel until the personnel are transported to finish the card punching, and the detection result of the video frames to be detected is sent to the management terminal in a system early warning mode. For example, when a card punching signal is triggered when a card is started to be input into or output from a warehouse and a card punching signal is not triggered when transportation is completed, images of dangerous chemical transport lines along cameras are grouped to serve as videos to be analyzed, frame extraction analysis is performed on each image of the camera, the detected target subjects are personnel (namely two persons to be input and one operator) and dangerous chemical containers, when the target personnel are detected, a frame-by-frame analysis mode is entered, whether the monitored images are only the personnel to be transported or not is detected, the dangerous chemical containers are not detected, and if the detected target personnel are detected, container separation early warning is triggered.
In an embodiment of the present invention, the management terminal 10 may be provided with a plurality of management terminals, and a person having management and approval authorities holds the management terminal. The management terminal records the working personnel and the hazardous chemical substance warehouse, examines and approves the working personnel and the hazardous chemical substance warehouse, the corresponding personnel and the hazardous chemical substance warehouse enter the database of the supervision terminal, and key information such as personnel names, posts, photos and the like is input into the system to form a basic information base. The operation personnel applies for the receiving operation at the user terminal, and the management terminal performs examination and approval and at least comprises two pieces of receiving personnel information and two pieces of custodian information.
In one embodiment of the invention, the distribution situation of the cameras in a factory is complex, and the situation of a long-distance visual angle may exist, so that the accuracy of face recognition is reduced, and even the face recognition cannot be realized. Therefore, the video processing module 50 performs incremental machine learning through the monitoring camera, performs gait feature analysis on people in the monitoring range of the plant area, and confirms the person ID corresponding to the gait feature in a management terminal marking mode. After receiving the analysis event, the video processing module 50 identifies the person ID based on the face and gait features, thereby improving the accuracy of person identification.
In an embodiment of the present invention, the video processing module 50 may be specifically an AI video box, which has functions of incremental machine learning, gait analysis, structured analysis, and the like.
In one embodiment of the present invention, the video structuring analysis used follows the following basic principles and methods: performing background modeling by a preset background modeling method, performing frame extraction analysis detection on a video to be detected, and if two continuous frame extraction analysis results have no target main body, no longer performing video structural analysis on video frames in the period so as to reduce the consumption of computing resources under the condition of no abnormal state; and starting a frame-by-frame analysis mode when the target subject is detected in the frame extraction, namely starting the frame-by-frame analysis after the frame extraction, and stopping the frame-by-frame analysis mode until the target subject is not detected in the continuous 2S video frames. The structural characteristics of the video structural analysis comprise the transportation track of the hazardous chemical container, the state (whether the hazardous chemical container is inclined or not and whether the hazardous chemical container is turned on one side or not), smoke, flame and personnel information.
In an embodiment of the present invention, when the non-claiming application event is triggered, the camera picture set in the hazardous chemical substance storage warehouse is used as a video frame to be detected, the video processing module 50 performs real-time detection on the frame to be detected, detects whether an unauthorized person enters, and sends an early warning message to the management terminal 10 if the unauthorized person enters. For example, frames are extracted from a monitoring picture of a warehouse camera, the extracted video frames are video frames to be analyzed, and a target subject detected in the state is a person, namely, an illegal intrusion early warning is triggered when the fact that the person enters the warehouse is detected.
In an embodiment of the present invention, under the condition that the reception start card punching signal is triggered and the reception completion card punching signal is not triggered, the video processing module 50 uses the picture of the storage camera of the hazardous chemical substance as a video to be analyzed, performs frame-by-frame analysis within the time period, detects whether the personnel of the target subject is consistent with the reception designated personnel, and sends an early warning message to the management terminal 10 if the personnel lack or the personnel information is inconsistent. For example, frame-by-frame analysis is performed in the time period, a personnel group of a target subject is detected, the number of persons arriving at the site of the warehouse is judged to be all persons authorized to receive through continuous detection analysis, the information is consistent with the designated personnel information in the warehouse-in and warehouse-out application event, and if no personnel exist, a post missing early warning is triggered. For another example, frame-by-frame analysis is performed in the time period, the detection target subject is a person, and if the person is detected to be a person confirmed in the non-warehouse-in/out application event, a "person unqualified" warning is triggered.
In one embodiment of the invention, in order to reduce the computational power consumption, the analyzed cameras are dynamically grouped when the camera pictures along the transportation line of the hazardous chemical substances are taken as key points of the video frames to be detected. Storing all camera information in a Directed Acyclic Graph (DAG), wherein each camera is recorded as a node in the DAG; according to a preset route for hazardous chemical articles, the cameras covered by the route establish DAG paths between two adjacent camera nodes according to the passing sequence of the route. When a new route is created, a Breadth First Search (BFS) is performed on the DAG to ensure that no cyclic path appears in the DAG data structure. And binding the dangerous chemical warehouse with the DAG path, finding out a camera list corresponding to the DAG path according to the binding relation when an analysis event is triggered, and sequentially analyzing the video characteristics in the camera video stream on the path according to the time sequence. By utilizing the DAG data structure to store the cameras on the transportation path of the extremely toxic dangerous chemicals, when an analysis event is triggered, only the video stream of the camera in the path related to the event is analyzed, so that the computational power consumption is reduced.
Corresponding to the embodiment of the system, referring to fig. 2, a hazardous chemical substance availability management method based on AI video structured analysis according to the embodiment of the present invention is shown, including:
the management system for hazardous chemical articles acceptance based on AI video structured analysis is provided, wherein the management system comprises a management terminal 10, a video processing module 50, a plurality of cameras 40 arranged along the transportation line of the hazardous chemical articles, a user terminal 20 and a card punching device 30, and is shown in figure 1, the management terminal is used for collecting personnel information with the authority of contacting the hazardous chemical articles and authorizing and distinguishing the authority for the division of labor of the personnel; the card punching equipment is arranged in the dangerous chemical storage warehouse, and a claimant confirms the card punching through a user terminal before the claimant;
after the adoption is started, the video processing module 50 carries out video structuralization analysis and detection on video frames to be detected, camera pictures along the transportation line of the hazardous chemical substances are used as key video frames to be detected, whether personnel entering the hazardous chemical substance storage warehouse are confirmed in the adoption application event is analyzed and identified, and whether personnel leave monitoring pictures in the process of transporting the hazardous chemical substances is detected by the personnel, the detection process is finished after the hazardous chemical substances are transported to a use workshop by the personnel until the personnel are transported to finish the card punching, and the detection results of the video frames to be detected are sent to the management terminal in a system early warning mode.
In an embodiment of the present invention, the video processing module 50 may be specifically an AI video box, and has functions of incremental machine learning, gait analysis, structural analysis, and the like.
The video structuring analysis employed follows the following basic principles and methods: performing background modeling by a preset background modeling method, performing frame extraction analysis detection on a video to be detected, and if two continuous frame extraction analysis results have no target main body, no longer performing video structural analysis on video frames in the period, so that the consumption of computing resources under the condition of no abnormal state is reduced; and starting a frame-by-frame analysis mode when the target subject is detected in the frame extraction, namely starting the frame-by-frame analysis after the frame extraction, and stopping the frame-by-frame analysis mode until the target subject is not detected in the continuous 2S video frames. The structural characteristics of the video structural analysis comprise the transportation track of the hazardous chemical container, the state (whether the hazardous chemical container is inclined or not and whether the hazardous chemical container is turned on one side or not), smoke, flame and personnel information.
In an embodiment of the present invention, a plurality of management terminals 10 may be provided, and a person having authority to manage and approve holds the management terminal 10. The management terminal records the working personnel and the hazardous chemical substance warehouse, examines and approves the working personnel and the hazardous chemical substance warehouse, the corresponding personnel and the hazardous chemical substance warehouse enter the database of the supervision terminal, and key information such as personnel names, posts, photos and the like is input into the system to form a basic information base. The method comprises the steps that an operator applies for a receiving operation at a user terminal, a receiving event is triggered, and a management terminal performs approval and at least comprises two pieces of receiving personnel information and two pieces of custodian information.
In an embodiment of the invention, the distribution situation of the cameras in the factory is complex, and the situation of a long-distance visual angle may exist, so that the accuracy of face recognition is reduced, and even the face recognition cannot be realized. Therefore, the video processing module 50 is configured to perform incremental machine learning through the monitoring camera, perform gait feature analysis on people in the monitoring range of the plant area, and confirm the person ID corresponding to the gait feature in a management terminal marking mode. After receiving the analysis event, the video processing module identifies the personnel ID based on the human face and the gait characteristics, and improves the accuracy of personnel identification.
In an embodiment of the present invention, when the non-claiming application event is triggered, the camera picture set in the hazardous chemical substance storage warehouse is used as a video frame to be detected, the video processing module 50 performs real-time detection on the frame to be detected, detects whether an unauthorized person intrudes, and sends an early warning message to the management terminal if the unauthorized person intrudes. For example, frames are extracted from a monitoring picture of a warehouse camera, the extracted video frames are video frames to be analyzed, and a target subject detected in the state is a person, namely, an illegal intrusion early warning is triggered when a person is detected to enter the warehouse.
In an embodiment of the invention, under the condition that the acceptance start card punching signal is triggered and the acceptance completion card punching signal is not triggered, the video processing module 50 is set to take a picture of a storage camera of hazardous chemical substances as a video to be analyzed, frame-by-frame analysis is carried out in the time period, whether personnel of a target main body are consistent with those appointed by the acceptance is detected, and if personnel lack or personnel information is inconsistent, an early warning message is sent to the management terminal. For example, frame-by-frame analysis is performed in the time period, a staff group of a target subject is detected, the number of persons on site in the warehouse is judged to be all persons authorized to receive through continuous detection analysis, the number of persons is consistent with the designated staff information in the warehouse entering and exiting application event, and if the persons are lacked, a post missing early warning is triggered. For another example, frame-by-frame analysis is performed in the time period, the detection target subject is a person, and if the person is detected to be a person confirmed in the non-warehouse-in/out application event, a "person unqualified" early warning is triggered.
In one embodiment of the invention, in order to reduce the computational power consumption, the analyzed cameras are dynamically grouped when the camera pictures along the transportation line of the hazardous chemical substances are taken as key video frames to be detected. Storing all camera information in a Directed Acyclic Graph (DAG), wherein each camera is recorded as a node in the DAG; according to a preset route for hazardous chemical substance receiving, cameras covered by the route establish a DAG path between two adjacent camera nodes according to the sequence of the route. When a route is newly established, the Breadth-First Search (BFS) is carried out on DAG, and no circular path appears in a DAG data structure. And binding the dangerous chemical warehouse with the DAG path, finding out a camera list corresponding to the DAG path according to the binding relation when an analysis event is triggered, and analyzing the video characteristics in the camera video stream on the path in sequence according to the time sequence. By utilizing the DAG data structure to store the cameras on the transportation path of the extremely toxic dangerous chemicals, when an analysis event is triggered, only the video stream of the camera in the path related to the event is analyzed, so that the computational power consumption is reduced.
The method realizes the purpose of receiving hazardous chemicals from a warehouse until the video structural analysis and early warning of operation standardization during transportation to a use workshop, and the method can realize accurate, efficient and traceable realization including whether field personnel meet the standard requirements or not and whether transportation operation meets the specified management requirements or not.
It is to be understood that the exemplary embodiments described herein are illustrative and not restrictive. Although one or more embodiments of the present invention have been described with reference to the accompanying drawings, it will be understood by those of ordinary skill in the art that various changes in form and details may be made therein without departing from the spirit and scope of the present invention as defined by the following claims.
Claims (18)
1. A dangerous chemical article availability management system based on AI video structured analysis is characterized by comprising a management terminal, a video processing module, a plurality of cameras arranged along the transportation line of dangerous chemicals, a user terminal and a card punching device,
the management terminal is used for collecting personnel information with the authority of contacting hazardous chemicals, and authorizing and authority division of personnel;
the card punching equipment is arranged in a dangerous chemical storage warehouse, and a claimant confirms the card punching through a user terminal before the claimant;
after the collar is punched, the video processing module carries out video structuralization analysis and detection on video frames to be detected, the video frames to be detected are taken as key video frames to be detected by cameras along the transportation line of the hazardous chemical substances, whether personnel entering the hazardous chemical substance storage warehouse are confirmed in the collar application event is analyzed and identified, and in the process of transporting the hazardous chemical substances, whether a hazardous chemical substance container leaves a monitoring picture is detected by the personnel, the detection process is finished after the hazardous chemical substances are transported to a use workshop by the personnel until the personnel are transported, the punching is finished, and the detection results of the video frames to be detected are sent to the management terminal in a system early warning mode.
2. The AI video structured analysis based hazardous chemical substance adoption management system of claim 1, wherein the video processing module performs incremental machine learning via a monitoring camera, performs gait feature analysis on persons within a monitoring range, and confirms a person ID corresponding to the gait feature in a management terminal marking manner.
3. The AI video structured analysis-based hazardous chemical item availability management system of claim 2, wherein the video processing module identifies the person ID based on face and gait characteristics upon receiving a video structured analysis event.
4. The AI video structured analysis-based hazardous chemical substance availability management system of claim 1, wherein when no availability application event is triggered, a camera picture arranged in a hazardous chemical substance storage warehouse is used as a video frame to be detected of a video frame to be detected, the video processing module detects the frame to be detected in real time, detects whether unauthorized people intrude, and sends an early warning message to the management terminal if the intrusion of unauthorized people is detected.
5. The AI video structured analysis based hazardous chemical substance availability management system of claim 1, wherein when the availability start card punch signal is triggered and the availability completion card punch signal is not triggered, the video processing module takes the hazardous chemical substance storage camera picture as the video to be analyzed, performs frame-by-frame analysis within the time period, detects whether the personnel of the target subject are consistent with the availability designated personnel, and sends an early warning message to the management terminal if the personnel lack or the personnel information is inconsistent.
6. The AI video structured analysis based hazardous chemical substance availability management system of claim 1, wherein said taking the camera pictures along the hazardous chemical substance transportation line as the key video frames to be detected comprises: storing all camera information in a directed acyclic graph DAG, wherein each camera is recorded as a node in the DAG; according to a preset route for hazardous chemical articles, the cameras covered by the route establish DAG paths between two adjacent camera nodes according to the passing sequence of the route.
7. The AI video structured analysis based hazardous chemical substance availability management system of claim 6, wherein when a route is newly created, BFS is searched for DAG breadth first to ensure that no cyclic path occurs in the DAG data structure.
8. The AI video structured analysis-based hazardous chemical substance availability management system according to claim 6 or 7, wherein the hazardous chemical substance warehouse is bound to a DAG path, when an analysis event is triggered, a camera list corresponding to the DAG path is found according to the binding relationship, and video features in camera video streams on the path are sequentially analyzed according to a time sequence.
9. The AI video structured analysis-based hazardous chemical availability management system of claim 1, wherein the structured features of the video structured analysis comprise hazardous chemical container transportation trajectories, hazardous chemical container status, smoke, fire and personnel information.
10. A dangerous chemical article receiving management method based on AI video structured analysis is characterized by comprising the following steps:
the method comprises the steps that a hazardous chemical substance receiving management system which comprises a management terminal, a video processing module, a plurality of cameras arranged along a hazardous chemical substance transportation line, a user terminal and a card punching device and is based on AI video structural analysis is arranged, wherein the management terminal is used for collecting personnel information with the authority of contacting the hazardous chemical substance, and authorizing and distinguishing the authority of the personnel for division of labor; the card punching equipment is arranged in a dangerous chemical storage warehouse, and a claimant confirms the card punching through a user terminal before the claimant;
after the collar is punched, the video processing module carries out video structuralization analysis and detection on video frames to be detected, the video frames to be detected are taken as key video frames to be detected by cameras along the transportation line of the hazardous chemical substances, whether personnel entering the hazardous chemical substance storage warehouse are confirmed in the collar application event is analyzed and identified, and in the process of transporting the hazardous chemical substances, whether a hazardous chemical substance container leaves a monitoring picture is detected by the personnel, the detection process is finished after the hazardous chemical substances are transported to a use workshop by the personnel until the personnel are transported, the punching is finished, and the detection results of the video frames to be detected are sent to the management terminal in a system early warning mode.
11. The AI video structured analysis-based hazardous chemical substance adoption management method according to claim 10, wherein the video processing module is configured to perform incremental machine learning through a monitoring camera, perform gait feature analysis on persons within a monitoring range, and confirm a person ID corresponding to the gait feature in a management terminal marking manner.
12. The AI video structured analysis-based hazardous chemical item availability management method of claim 11, wherein the video processing module is configured to identify a person ID based on face and gait characteristics upon receipt of a video structured analysis event.
13. The AI video structured analysis-based hazardous chemical substance availability management method according to claim 10, wherein when an availability application event is triggered, a camera picture arranged in a hazardous chemical substance storage warehouse is used as a video frame to be detected of a video frame to be detected, the video processing module detects the frame to be detected in real time, detects whether an unauthorized person enters the hazardous chemical substance storage warehouse or not, and sends an early warning message to the management terminal if the unauthorized person enters the hazardous chemical substance storage warehouse.
14. The AI video structured analysis-based hazardous chemical substance availability management method according to claim 10, wherein under the condition that availability start card punching signal is triggered and availability completion card punching signal is not triggered, the video processing module takes a hazardous chemical substance storage camera picture as a video to be analyzed, performs frame-by-frame analysis within the time period, detects whether a person of a target subject is consistent with an availability designated person, and sends an early warning message to the management terminal if the person is absent or the information of the person is inconsistent.
15. The AI video structured analysis-based hazardous chemical substance availability management method according to claim 10, wherein the processing of taking the camera pictures along the hazardous chemical substance transportation line as key points for the video frames to be detected comprises: storing all camera information in a directed acyclic graph DAG, wherein each camera is recorded as a node in the DAG; according to a preset route for hazardous chemical substance receiving, cameras covered by the route establish a DAG path between two adjacent camera nodes according to the sequence of the route.
16. The AI video structured analysis-based hazardous chemical substance availability management method of claim 15, wherein when a route is newly created, BFS is searched for in a breadth-first manner for DAG, ensuring that no cyclic paths appear in a DAG data structure.
17. The AI video structured analysis-based hazardous chemical substance availability management method according to claim 15 or 16, wherein the hazardous chemical substance warehouse is bound to a DAG path, when an analysis event is triggered, a camera list corresponding to the DAG path is found according to the binding relationship, and video features in camera video streams on the path are sequentially analyzed according to a time sequence.
18. The AI video structured analysis-based hazardous chemical item adoption management method of claim 10, wherein the structured features of the video structured analysis include hazardous chemical item container transportation trajectory, hazardous chemical item container status, smoke, fire and personnel information.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211238562.0A CN115314685A (en) | 2022-10-11 | 2022-10-11 | AI video structured analysis-based hazardous chemical substance availability management system and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211238562.0A CN115314685A (en) | 2022-10-11 | 2022-10-11 | AI video structured analysis-based hazardous chemical substance availability management system and method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115314685A true CN115314685A (en) | 2022-11-08 |
Family
ID=83867918
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211238562.0A Pending CN115314685A (en) | 2022-10-11 | 2022-10-11 | AI video structured analysis-based hazardous chemical substance availability management system and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115314685A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115578827A (en) * | 2022-12-07 | 2023-01-06 | 杭州宇链科技有限公司 | Method and system for tracking dangerous chemical article receiving container based on AI video |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101943902A (en) * | 2010-08-18 | 2011-01-12 | 北京物资学院 | Safety monitoring system and method for dangerous goods logistics |
| CN103632044A (en) * | 2013-11-19 | 2014-03-12 | 北京环境特性研究所 | Camera topology building method and device based on geographic information system |
| CN105574700A (en) * | 2015-12-30 | 2016-05-11 | 方正宽带网络服务有限公司 | Hazardous material management system and method |
| CN106791700A (en) * | 2017-01-20 | 2017-05-31 | 辽宁科技大学 | A kind of enterprise's key area personnel path safety monitoring system and method |
| CN109558292A (en) * | 2017-09-26 | 2019-04-02 | 阿里巴巴集团控股有限公司 | A kind of monitoring method and device |
| CN209486860U (en) * | 2019-01-28 | 2019-10-11 | 武汉恩特拉信息技术有限公司 | A kind of personnel safety monitoring device based on recognition of face |
| CN110889662A (en) * | 2019-12-25 | 2020-03-17 | 上海耀客物联网有限公司 | Dangerous chemical management system and management method |
| CN111179551A (en) * | 2019-12-17 | 2020-05-19 | 西安工程大学 | Real-time monitoring method for dangerous chemical transport driver |
| CN112073527A (en) * | 2020-09-14 | 2020-12-11 | 华南农业大学 | Campus hazardous chemical substance overall process management system and method |
| CN114037228A (en) * | 2021-10-27 | 2022-02-11 | 四川易链科技有限公司 | Hazardous chemical substance full-flow supervision method and system based on block chain |
-
2022
- 2022-10-11 CN CN202211238562.0A patent/CN115314685A/en active Pending
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101943902A (en) * | 2010-08-18 | 2011-01-12 | 北京物资学院 | Safety monitoring system and method for dangerous goods logistics |
| CN103632044A (en) * | 2013-11-19 | 2014-03-12 | 北京环境特性研究所 | Camera topology building method and device based on geographic information system |
| CN105574700A (en) * | 2015-12-30 | 2016-05-11 | 方正宽带网络服务有限公司 | Hazardous material management system and method |
| CN106791700A (en) * | 2017-01-20 | 2017-05-31 | 辽宁科技大学 | A kind of enterprise's key area personnel path safety monitoring system and method |
| CN109558292A (en) * | 2017-09-26 | 2019-04-02 | 阿里巴巴集团控股有限公司 | A kind of monitoring method and device |
| CN209486860U (en) * | 2019-01-28 | 2019-10-11 | 武汉恩特拉信息技术有限公司 | A kind of personnel safety monitoring device based on recognition of face |
| CN111179551A (en) * | 2019-12-17 | 2020-05-19 | 西安工程大学 | Real-time monitoring method for dangerous chemical transport driver |
| CN110889662A (en) * | 2019-12-25 | 2020-03-17 | 上海耀客物联网有限公司 | Dangerous chemical management system and management method |
| CN112073527A (en) * | 2020-09-14 | 2020-12-11 | 华南农业大学 | Campus hazardous chemical substance overall process management system and method |
| CN114037228A (en) * | 2021-10-27 | 2022-02-11 | 四川易链科技有限公司 | Hazardous chemical substance full-flow supervision method and system based on block chain |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115578827A (en) * | 2022-12-07 | 2023-01-06 | 杭州宇链科技有限公司 | Method and system for tracking dangerous chemical article receiving container based on AI video |
| CN115578827B (en) * | 2022-12-07 | 2023-03-14 | 杭州宇链科技有限公司 | Method and system for tracking dangerous chemical article receiving container based on AI video |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109980781A (en) | A kind of transformer substation intelligent monitoring system | |
| CN112785798B (en) | Behavior analysis method for constructors of power substation engineering construction project | |
| CN109767121A (en) | A kind of substation distant place intelligent operation managing and control system and method | |
| CN110287859A (en) | A kind of personnel's material dispatching system based on artificial intelligence technology | |
| CN104866982A (en) | Internet+hazardous waste management system and method | |
| CN108515991B (en) | Receiving and dispatching train standardized operation detecting system | |
| CN106302758A (en) | Supervision department's management informationalized to demolition site system and using method thereof | |
| CN115314685A (en) | AI video structured analysis-based hazardous chemical substance availability management system and method | |
| CN204791136U (en) | System for internet + dangerous discarded object management | |
| CN111460985A (en) | On-site worker track statistical method and system based on cross-camera human body matching | |
| CN112800910A (en) | Communication machine room maintenance operation efficiency evaluation method and system | |
| CN112381435A (en) | Gridding directional pushing management method for dynamic risk in hydropower station operation process | |
| CN113393607B (en) | Intelligent electric power security method and system | |
| CN107169627A (en) | A kind of electric energy meter quality management-control method and system based on augmented reality | |
| CN107194265A (en) | A kind of device of the warehouse compartment supervision based on block chain image processing techniques | |
| CN208588688U (en) | A kind of urban track traffic safe examination system | |
| CN113038079A (en) | Positioning video linkage system and method based on 5G + MEC | |
| CN111947758A (en) | Method for acquiring real weighing data of material | |
| WO2023116250A1 (en) | Vehicle maintenance method and system | |
| CN111723725A (en) | Multi-dimensional analysis system based on video AI | |
| CN109858829A (en) | A kind of accident risks method for early warning based on block chain | |
| CN114048346B (en) | GIS-based safety production integrated management and control platform and method | |
| CN213241249U (en) | Railway station six-in-one passenger real-name system verification system based on artificial intelligence | |
| CN112153067B (en) | Edge computing safety system based on block chain | |
| CN114257607B (en) | Intelligent dangerous chemical full life cycle management system |
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 |