Chemical wharf operation intelligent monitoring system and method based on video AI technology

Abstract

The invention relates to a chemical wharf operation intelligent monitoring system and method based on a video AI technology, and the architecture of the chemical wharf operation intelligent monitoring system is divided into five layers: the lowest layer is an infrastructure layer, and the infrastructure layer is used for a video camera class and a data collection equipment class; the system specifically comprises an enterprise intranet, a camera, a machine room, a monitoring command place and data acquisition equipment, the middle level comprises a data layer and a platform layer, and the data layer and the platform layer are responsible for data acquisition and intelligent calculation; the intelligent monitoring system and method for chemical wharf operation based on the video AI technology are mainly used for constructing an integrated innovative full-management system of active sensing, monitoring, early warning and disposal based on machine vision, resource sharing and comprehensive utilization are promoted, repeated investment and construction are avoided, business management innovation is optimized, and the intelligent monitoring system and method for chemical wharf operation based on the video AI technology are suitable for industrial production. The economic benefit is created through an innovation mechanism, the cost is saved, the enterprise management service level is improved, the business management innovation is optimized, and the economic benefit is created through the innovation mechanism.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent monitoring systems, and in particular to an intelligent monitoring system and method for chemical terminal operations based on video AI technology. Background Art

[0002] Video surveillance technology: Video surveillance is a comprehensive technology that integrates front-end camera, mid-end transmission, control, and back-end display and recording. It captures images in the monitoring area in real time through cameras installed in different places, and sends these images to the central processing equipment through the transmission system for storage and management. With the rapid development of modern technology, video surveillance systems have become a key tool for improving security, improving production efficiency, and providing intelligent services.

[0003] Traditional video surveillance technology does face many challenges in current application scenarios. These problems not only affect the efficiency and accuracy of the monitoring system, but also limit its application in more fields. Its shortcomings include: 1. Difficulty in manual monitoring: The wide operating area and complex environment lead to heavy monitoring tasks, long inspection time and single data collection dimension, making manual identification difficult.

[0004] 2. No timely warning: Employees frequently violate regulations at the work site, and the system relies on manual review of surveillance videos, which is inflexible and results in untimely feedback on issues.

[0005] 3. Difficulty in collecting evidence after the fact: Playing back the video after the fact requires manual investigation to detect anomalies.

[0006] 4. Difficulty in data processing: Traditional video surveillance consumes a lot of storage space to store meaningless data. Unstructured data has low utilization rate and is difficult to extract. Summary of the invention

[0007] To achieve the above-mentioned purpose, the present invention provides the following technical solutions: an intelligent monitoring system for chemical terminal operations based on video AI technology, wherein the architecture of the intelligent monitoring system for chemical terminal operations is divided into five layers: the bottom layer is the infrastructure layer, which is used for video cameras and data acquisition equipment, and specifically includes an enterprise intranet, cameras, computer rooms, monitoring and command sites, and data acquisition equipment; The middle level is the data layer and platform layer, which are responsible for data collection and intelligent computing; the data layer includes security databases, visual analysis results, third-party system data, enterprise knowledge bases and other systems, and the platform layer is used for data logic organization, data collection, data storage, AI computing platform and system integration; The upper layer is the application layer and the display layer. The specific business applications of the application layer include personnel management, operation management, and environmental monitoring. Through the front-end video images, third-party system data, enterprise knowledge base, back-end intelligent servers, and platform layout, basic security prevention capabilities are built, the global perspective capability of security management is built, and the full perception and early warning capabilities are enhanced; The display layer includes mobile APP access, PC access, electronic work ticket system and SMS alarm platform.

[0008] Preferably, data collection and data processing collect unstructured data such as videos and pictures in real time through network cameras and NVRs, and analyze them through AI algorithms to form structured data, and abstract and model processes and specifications to form structured data.

[0009] Preferably, the construction of the algorithm model for the operation scene includes the whole domain, all elements and the whole process. The whole domain covers the spatial scope of the entire production scene, the full elements integrate all information on people, vehicles, objects and operation processes in the scene, and the whole process is the key operation point of the operation process in the scene.

[0010] Preferably, the data architecture uses a MySQL database, and cached data is stored in redis.

[0011] Preferably, the front-end authorized access adopts a token mechanism, and the token is returned to the front-end at login. The front-end caches it locally, and the back-end token is stored in Redis. The back-end user TOKEN is valid for 1 day and is authenticated through the API gateway.

[0012] Preferably, for access that exceeds the system's carrying capacity, current limiting and circuit breaking technology can be used to control the number of concurrency to ensure the smooth operation of the system. For simultaneous access to the same resource, such as consumption of message queues, etc., distributed locks are used to control it to ensure changes to the same resource at the same time, and only one caller can perform the operation at the same time.

[0013] Preferably, tasks involving timed calculations are implemented through the timed task framework of Spring Boot.

[0014] Preferably, the communication between the front-end and back-end is carried out using HTTPS, and the security of transmitted data is guaranteed by channel encryption to prevent data hijacking; the interface design adopts the Restful style to ensure the consistency of the interface specification and the readability of the interface definition; the response to the foreground data uniformly adopts the ApiResult object, which includes data body data, status status and message message.

[0015] An intelligent monitoring method for chemical terminal operations based on video AI technology, the intelligent monitoring method comprises the following steps: Step 1: Access the video stream of the surveillance camera of the established video surveillance system; the service pulls the video stream to the AI ​​intelligent security monitoring management platform; the video processing server processes the video stream; the algorithm model identifies and analyzes the video / image to obtain the anomaly detection results; the anomaly detection results are counted and analyzed and displayed on the intelligent monitoring display and analysis platform; the detailed results of the relevant anomaly detection are recorded and displayed in the video monitoring alarm; the platform pushes the anomaly detection results to the designated personnel through email, SMS, browser alarm or other sound and light alarm systems; the server cluster is required to provide web services, data services, MQ services, streaming media services, AI engines and other related system services; Step 2: In the petrochemical terminal safety production scenario, use the safety situation dashboard for daily management and as a unified window for emergency command, and build capabilities from safety production status visualization, business analysis & early warning, to auxiliary decision-making and execution, to achieve visualization, manageability, and controllability of petrochemical terminal production operations.

[0016] Step 3: Use real-time monitoring of hazardous sources to monitor the actual scene of hazardous sources in specific operation scenes; Step 4: Alarm processing and statistical analysis. Query historical operation records by time. In the details, you can view violations such as personnel, actions, and processes in the operation, which is convenient for closed-loop optimization management of the entire process before, during, and after the operation. For alarm events found in the operation, they are pushed to relevant responsible persons through emails, text messages, etc., and alarm information is displayed. The occurrence volume, processing volume, response speed, processing results, etc. of various types of alarm events are summarized and statistically analyzed to form statistical conclusions and provide decision-making support; Step 5: Rule configuration: customize various alarm rules according to business processes and security management standards, and support rule management. One rule is bound to multiple algorithms and multiple cameras, and the algorithm startup conditions are configured at the same time. Step 6: Interface management is used to connect to the algorithm warehouse, video surveillance platform and various other software systems. The connected system interfaces are placed in the docking module for unified management. It supports the full life cycle management of the interface, monitors the application, IP address, request time, request URL, request method, response status, response time, and detailed request content of the access interface, supports regular automatic requests to external interfaces, synchronizes external data, and supports custom settings for docking frequency.

[0017] Preferably, the security situation data dashboard displays content including: risk point overview, real-time alarm status, today's hidden danger distribution statistics, hidden danger trend statistics, personnel distribution, today's temporary operation overview and digital map.

[0018] Compared with the prior art, the present invention has the following beneficial effects: The present invention provides a chemical terminal operation intelligent monitoring system and method based on video AI technology. 1. Excellent economic benefits: To promote resource sharing and comprehensive utilization and avoid duplicate investment and construction, an intelligent algorithm software platform will be built in this project. Through open interfaces, the platform provides data exchange services, data integration services, directory management and services, data publishing services, etc. for existing and future other business application systems to share and use after being granted permissions. After the project is completed, it can realize resource sharing and business collaboration among heterogeneous systems of different departments in the enterprise, effectively avoid problems such as multiple investments, duplicate construction, and waste of resources, and effectively support the normal and healthy operation and management of the enterprise; optimize business management innovation, create economic benefits through innovative mechanisms, and create direct economic benefits through business innovation and service improvement. The platform can greatly reduce the occurrence of safety accidents at the terminal, improve the level of safety management at the terminal, reduce the probability of accidents, and create economic benefits; 2. It has good social benefits, can improve enterprise management efficiency, and ensure enterprise safety. After the intelligent algorithm platform system is completed, it will greatly improve the safety of terminal production, comprehensively improve the safety of staff access, vehicle access, ship access, production equipment operation, and various production and life safety levels, and greatly improve human resource efficiency and office efficiency; at the same time, the construction of this project ensures the timely update of enterprise security data, saves the company's financial expenditure, greatly reduces the occurrence of emergencies, reduces the risk of loss of life and property, and brings the company's safety index to a new height; saves costs and improves the level of enterprise management services. The construction of an intelligent algorithm platform can meet the company's increasing business data storage and management services, realize on-demand payment requirements for resources and services for multiple users, improve hardware utilization, and reduce energy consumption and facility costs. It can also simplify work processes and reduce manual intervention costs, which is conducive to reducing the company's operating costs and improving the company's economic operation level. It has important practical significance for improving the company's appearance and improving the living standards of workers. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 This is a schematic diagram of the architecture of the intelligent monitoring system for chemical terminal operations based on video AI technology of the present invention. DETAILED DESCRIPTION

[0020] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0021] See also Figure 1 , the present invention provides a technical solution: a chemical terminal operation intelligent monitoring system based on video AI technology, the chemical terminal operation intelligent monitoring system architecture is divided into five layers: the bottom is the infrastructure layer, the infrastructure layer is used for video cameras and data acquisition equipment, which specifically includes enterprise intranet, cameras, computer rooms and monitoring command places, data acquisition equipment; The middle level is the data layer and platform layer, which are responsible for data collection and intelligent computing; the data layer includes security databases, visual analysis results, third-party system data, enterprise knowledge bases and other systems, and the platform layer is used for data logic organization, data collection, data storage, AI computing platform and system integration; The upper layer is the application layer and the display layer. The specific business applications of the application layer include personnel management, operation management, and environmental monitoring. Through the front-end video images, third-party system data, enterprise knowledge base, back-end intelligent servers, and platform layout, basic security prevention capabilities are built, the global perspective capability of security management is built, and the full perception and early warning capabilities are enhanced; The display layer includes mobile APP access, PC access, electronic work ticket system and SMS alarm platform.

[0022] Among them, data collection and data processing collect unstructured data such as videos and pictures in real time through network cameras and NVRs, and analyze them through AI algorithms to form structured data, and abstract and model processes and specifications to form structured data.

[0023] Among them, the construction of the algorithm model for the operation scene includes the entire domain, all elements and the entire process. The entire domain covers the spatial scope of the entire production scene. The full elements integrate all information on people, vehicles, objects, and operation processes in the scene. The full process is the key operation point of the operation process in the scene.

[0024] Among them, the data architecture uses the MySQL database, and the cached data is stored in redis.

[0025] Among them, the front-end authorized access adopts the token mechanism. The token is returned to the front-end at login, and the front-end is cached locally. At the same time, the back-end token is stored in Redis. The back-end user TOKEN is valid for 1 day and is authenticated through the API gateway.

[0026] Among them, for access that exceeds the system's carrying capacity, current limiting and circuit breaker technology can be used to control the number of concurrency to ensure the smooth operation of the system. For simultaneous access to the same resource, such as consumption of message queues, distributed locks are used to control it to ensure changes to the same resource at the same time, and only one caller can perform the operation.

[0027] Among them, tasks involving timed calculations are implemented through the spring boot timed task framework.

[0028] Among them, the communication between the front and back ends is carried out using HTTPS, and channel encryption is used to ensure the security of transmitted data and prevent data hijacking; the interface design adopts the Restful style to ensure the consistency of the interface specification and the readability of the interface definition; the response to the foreground data uniformly adopts the ApiResult object, which includes data body data, status status and message message.

[0029] An intelligent monitoring method for chemical terminal operations based on video AI technology, the intelligent monitoring method comprises the following steps: Step 1: Access the video stream of the surveillance camera of the established video surveillance system; the service pulls the video stream to the AI ​​intelligent security monitoring management platform; the video processing server processes the video stream; the algorithm model identifies and analyzes the video / image to obtain the anomaly detection results; the anomaly detection results are counted and analyzed and displayed on the intelligent monitoring display and analysis platform; the detailed results of the relevant anomaly detection are recorded and displayed in the video monitoring alarm; the platform pushes the anomaly detection results to the designated personnel through email, SMS, browser alarm or other sound and light alarm systems; the server cluster is required to provide web services, data services, MQ services, streaming media services, AI engines and other related system services; Step 2: In the petrochemical terminal safety production scenario, use the safety situation dashboard for daily management and as a unified window for emergency command, and build capabilities from safety production status visualization, business analysis & early warning, to auxiliary decision-making and execution, to achieve visualization, manageability, and controllability of petrochemical terminal production operations.

[0030] Step 3: Use real-time monitoring of hazardous sources to monitor the actual scene of hazardous sources in specific operation scenes; Step 4: Alarm processing and statistical analysis. Query historical operation records by time. In the details, you can view violations such as personnel, actions, and processes in the operation, which is convenient for closed-loop optimization management of the entire process before, during, and after the operation. For alarm events found in the operation, they are pushed to relevant responsible persons through emails, text messages, etc., and alarm information is displayed. The occurrence volume, processing volume, response speed, processing results, etc. of various types of alarm events are summarized and statistically analyzed to form statistical conclusions and provide decision-making support; Step 5: Rule configuration: customize various alarm rules according to business processes and security management standards, and support rule management. One rule is bound to multiple algorithms and multiple cameras, and the algorithm startup conditions are configured at the same time. Step 6: Interface management is used to connect to the algorithm warehouse, video surveillance platform and various other software systems. The connected system interfaces are placed in the docking module for unified management. It supports the full life cycle management of the interface, monitors the application, IP address, request time, request URL, request method, response status, response time, and detailed request content of the access interface, supports regular automatic requests to external interfaces, synchronizes external data, and supports custom settings for docking frequency.

[0031] Among them, the security situation data dashboard displays: risk point overview, real-time alarm status, today's hidden danger distribution statistics, hidden danger trend statistics, personnel distribution, today's temporary operation overview and digital map.

[0032] It should be noted that this project relies on the existing video surveillance platform to achieve functions such as personnel management, leakage, smoke and fire identification, terminal operation supervision, inspection and testing, and vehicle management in the petrochemical terminal area. The construction content includes the petrochemical terminal video AI comprehensive management and control platform, video intelligent analysis algorithm and corresponding supporting hardware facilities.

[0033] Project implementation process requirements: 1) Access the video stream of the surveillance camera of the established video surveillance system; 2) The service pulls the video stream to the AI ​​smart security monitoring management platform; 3) The video processing server processes the video stream; 4) The algorithm model identifies and analyzes the video / image to obtain anomaly detection results; 5) Statistics and analysis of abnormal detection results are performed and displayed on the intelligent monitoring display and analysis platform; 6) Record and display detailed results of relevant abnormality detection in video monitoring alarm; 7) The platform pushes the abnormal detection results to designated personnel via email, SMS, browser alarm or other sound and light alarm systems (if any).

[0034] 8) The server cluster is required to provide web services, data services, MQ services, streaming media services, AI engines and other related system services.

[0035] The AI ​​integrated video management and control platform for petrochemical terminals needs to fully adapt to the operating scenarios of chemical companies. It can monitor and analyze the personnel status, operating behavior, environmental status, and equipment status in real time on-site, identify and alert existing problems, and automatically generate a job completion report after the job is completed.

[0036] B. Security situation dashboard of the petrochemical terminal video AI integrated management and control platform In the safety production scenario of petrochemical terminals, the safety situation dashboard is a unified window for daily management and emergency command. It can also build capabilities from safety production status visualization, business analysis & early warning, to auxiliary decision-making and execution, making the production operations of petrochemical terminals visible, manageable, and controllable.

[0037] The security situation data dashboard displays: risk point overview, real-time alarm status, today's hidden danger distribution statistics, hidden danger trend statistics, personnel distribution, today's temporary operation overview and digital map.

[0038] C. Real-time monitoring of hazardous sources Real-time monitoring of hazardous sources can realize real-time monitoring of hazardous sources in specific operation scenarios.

[0039] It supports regional and hierarchical viewing, supports multi-screen camera display with 4-split screen, 9-split screen and 16-split screen, and can view the algorithm type, alarm rules, alarm records and historical alarm records loaded in each scene.

[0040] It supports multiple event alarm modes such as real-time alarm box pop-up, alarm sound, alarm SMS, etc. The alarm event includes alarm information, alarm pictures, and videos before and after the alarm.

[0041] D. Alarm processing and statistical analysis Historical operation records can be queried by time, and the details include violations of personnel, actions, processes, etc. during the operation, which facilitates closed-loop optimization management of the entire process before, during, and after the operation.

[0042] For alarm events found during operation, the alarm information is pushed to relevant responsible persons through emails, text messages, etc. The occurrence volume, processing volume, response speed, processing results, etc. of various types of alarm events are summarized and statistically analyzed to form statistical conclusions and provide decision-making support.

[0043] E. Rule Configuration It supports the creation of alarm rules and can customize various alarm rules according to business processes and security management standards.

[0044] Supports rule management. One rule can be bound to multiple algorithms and multiple cameras.

[0045] Supports automatic switching algorithms and configures the start conditions of the algorithms, which can be the status of the device or the weather or any external information.

[0046] F. Interface Management It is used to connect to the algorithm warehouse, video surveillance platform and various other software systems, and put the connected system interfaces in the docking module for unified management, supporting the full life cycle management of the interface. It can monitor the application, IP address, request time, request URL, request method, response status, response time, detailed request content, etc. of the access interface.

[0047] Supports regular and automatic requests to external interfaces, synchronization of external data, and custom settings for docking frequency.

[0048] The above shows and describes the basic principles and main features of the present invention and the advantages of the present invention. It is obvious to those skilled in the art that the present invention is not limited to the details of the above exemplary embodiments, and the present invention can be implemented in other specific forms without departing from the spirit or basic features of the present invention. Therefore, no matter from which point of view, the embodiments should be regarded as exemplary and non-restrictive. The scope of the present invention is defined by the attached claims rather than the above description, and it is intended that all changes falling within the meaning and scope of the equivalent elements of the claims are included in the present invention. Any figure mark in the claims should not be regarded as limiting the claims involved.

[0049] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. Intelligent monitoring system for chemical terminal operations based on video AI technology, characterized by: The chemical terminal operation intelligent monitoring system architecture is divided into five layers: the bottom is the infrastructure layer, which is used for video cameras and data acquisition equipment, and specifically includes the enterprise intranet, cameras, computer rooms and monitoring command sites, and data acquisition equipment; The middle level is the data layer and platform layer, which are responsible for data collection and intelligent computing; the data layer includes security databases, visual analysis results, third-party system data, enterprise knowledge bases and other systems, and the platform layer is used for data logic organization, data collection, data storage, AI computing platform and system integration; The upper layer is the application layer and the display layer. The specific business applications of the application layer include personnel management, operation management, and environmental monitoring. Through the front-end video images, third-party system data, enterprise knowledge base, back-end intelligent servers, and platform layout, basic security prevention capabilities are built, the global perspective capability of security management is built, and the full perception and early warning capabilities are enhanced; The display layer includes mobile APP access, PC access, electronic work ticket system and SMS alarm platform.

2. The intelligent monitoring system for chemical terminal operations based on video AI technology according to claim 1 is characterized by: Data collection and data processing collect unstructured data such as videos and pictures in real time through network cameras and NVRs, and analyze them through AI algorithms to form structured data. Processes and specifications are abstracted and modeled to form structured data.

3. The intelligent monitoring system for chemical terminal operations based on video AI technology according to claim 1 is characterized by: The construction of the algorithm model for the operation scene includes the entire domain, all elements and the entire process. The entire domain covers the spatial scope of the entire production scene. The full elements integrate all information on people, vehicles, objects and operation processes in the scene. The full process is the key operation point of the operation process in the scene.

4. The intelligent monitoring system for chemical terminal operations based on video AI technology according to claim 1 is characterized by: The data architecture uses the MySQL database, and cached data is stored in redis.

5. The intelligent monitoring system for chemical terminal operations based on video AI technology according to claim 1 is characterized by: The front-end authorized access adopts the token mechanism. The token is returned to the front-end at login. The front-end caches it locally, and the back-end token is stored in Redis. The back-end user TOKEN is valid for 1 day and is authenticated through the API gateway.

6. The intelligent monitoring system for chemical terminal operations based on video AI technology according to claim 1 is characterized by: For access that exceeds the system's carrying capacity, current limiting and circuit breaking technology can be used to control the number of concurrent requests to ensure the smooth operation of the system. For simultaneous access to the same resource, such as consumption of message queues, distributed locks are used to control changes to the same resource at the same time, and only one caller can perform the operation.

7. The intelligent monitoring system for chemical terminal operations based on video AI technology according to claim 1 is characterized by: Tasks involving timed calculations are implemented through the spring boot timed task framework.

8. The intelligent monitoring system for chemical terminal operations based on video AI technology according to claim 1 is characterized by: The communication between the front and back ends is carried out using HTTPS, and channel encryption is used to ensure the security of transmitted data and prevent data hijacking. The interface design adopts the Restful style to ensure the consistency of the interface specification and the readability of the interface definition. The response to the front-end data uniformly adopts the ApiResult object, which includes data body data, status status, and message message.

9. Intelligent monitoring method for chemical terminal operations based on video AI technology, characterized by: The intelligent monitoring method comprises the following steps: Step 1: Access the video stream of the surveillance camera of the established video surveillance system; the service pulls the video stream to the AI ​​intelligent security monitoring management platform; the video processing server processes the video stream; the algorithm model identifies and analyzes the video / image to obtain the anomaly detection results; the anomaly detection results are counted and analyzed and displayed on the intelligent monitoring display and analysis platform; the detailed results of the relevant anomaly detection are recorded and displayed in the video monitoring alarm; the platform pushes the anomaly detection results to the designated personnel through email, SMS, browser alarm or other sound and light alarm systems; the server cluster is required to provide web services, data services, MQ services, streaming media services, AI engines and other related system services; Step 2: In the petrochemical terminal safety production scenario, use the safety situation dashboard for daily management and as a unified window for emergency command, and build capabilities from safety production status visualization, business analysis & early warning, to auxiliary decision-making and execution, to achieve visualization, manageability, and controllability of petrochemical terminal production operations. Step 3: Use real-time monitoring of hazardous sources to monitor the actual scene of hazardous sources in specific operation scenes; Step 4: Alarm processing and statistical analysis. Query historical operation records by time. In the details, you can view violations such as personnel, actions, and processes in the operation, which is convenient for closed-loop optimization management of the entire process before, during, and after the operation. For alarm events found in the operation, they are pushed to relevant responsible persons through emails, text messages, etc., and alarm information is displayed. The occurrence volume, processing volume, response speed, processing results, etc. of various types of alarm events are summarized and statistically analyzed to form statistical conclusions and provide decision-making support; Step 5: Rule configuration: customize various alarm rules according to business processes and security management standards, and support rule management. One rule is bound to multiple algorithms and multiple cameras, and the algorithm startup conditions are configured at the same time. Step 6: Interface management is used to connect to the algorithm warehouse, video surveillance platform and various other software systems. The connected system interfaces are placed in the docking module for unified management. It supports the full life cycle management of the interface, monitors the application, IP address, request time, request URL, request method, response status, response time, and detailed request content of the access interface, supports regular automatic requests to external interfaces, synchronizes external data, and supports custom settings for docking frequency.

10. The intelligent monitoring method for chemical terminal operations based on video AI technology according to claim 9 is characterized in that: The security situation data dashboard displays: risk point overview, real-time alarm status, today's hidden danger distribution statistics, hidden danger trend statistics, personnel distribution, today's temporary operation overview and digital map.

Images