Alarm system for automatically identifying nonstandard lightering loading operation based on AI image

Abstract

The invention discloses an alarm system for automatically identifying nonstandard lightering loading operation based on an AI image, which relates to the technical field of ship alarm systems and comprises an image acquisition module, an image preprocessing module, an AI image identification module, an alarm triggering module, a linkage alarm module, an alarm setting module, a work responsibility supervision module and a system management module. The alarm system for automatically identifying the non-standard lightering loading operation based on the AI image can automatically identify the non-standard lightering loading operation based on the AI image, has a flexible alarm mechanism, can set grades according to alarm types, can perform setting according to a first grade, a second grade, a third grade and a fourth grade, and meanwhile, customizes advanced processing time efficiency and an alarm notification mode, so that the alarm efficiency is improved. The work responsibilities of related personnel are effectively supervised by setting the timeliness of alarm confirmation and processing, the platform can check alarm information, short messages or mails can be set to receive alarms, and the supervision situation is concerned at any time when people work outdoors.

Description

Technical Field

[0001] The present invention relates to the technical field of ship alarm systems, and in particular to an alarm system based on AI images to automatically identify irregularities in transshipment loading operations. Background Art

[0002] Ship is a general term for various vessels, which are means of transportation that can sail or anchor in waters for transportation or operations. Ships have different technical performance, equipment and structural types according to different usage requirements. The appearance of a ship is generally a streamlined envelope that is conducive to overcoming fluid resistance. Its materials are constantly updated with the advancement of science and technology, from early natural materials such as wood, bamboo, and hemp to modern steel, aluminum, fiberglass, acrylic and various composite materials. Ships occupy a very important position in national defense, national economy and marine development. According to their use, ships are generally divided into two categories: military and civilian. Among them, civilian ships are generally divided into transport ships, engineering ships, fishing boats, port ships, etc.

[0003] Irregularity in ship transshipment and loading operations refers to the failure to follow established operating procedures, safety standards or industry norms during the transshipment and loading of cargoes, including operating procedure errors, insufficient safety measures, improper use of equipment, personnel operating errors, poor environmental management, etc. Once these occur, they may cause serious safety, environmental and economic problems. Therefore, it is necessary to strengthen the supervision and management of ship transshipment and loading operations to ensure the standardization and safety of the operation process. The ship alarm system is a comprehensive safety system that monitors the various states of the ship during operation and issues warnings when abnormalities or potential dangers are found.

[0004] However, in terms of automatic identification of irregularities in transshipment loading operations, since the existing ship alarm system mainly focuses on monitoring the physical state of the ship (such as liquid level, pressure, etc.), the automatic identification capability of the complex transshipment loading operation process and its standardization is limited. It is technically difficult to achieve real-time monitoring and automatic judgment of all loading operation links. Different types of ships and different loading operation scenarios have different regulatory requirements. The existing alarm system is difficult to adapt to these diverse rule changes, resulting in false alarms or missed alarms.

[0005] Therefore, it is urgent to improve this shortcoming. The present invention studies and improves the existing technology and its shortcomings, and provides an alarm system based on AI images to automatically identify irregularities in transshipment loading operations. Summary of the invention

[0006] The purpose of the present invention is to provide an alarm system based on AI images for automatically identifying irregularities in transshipment loading operations, so as to solve the problems raised in the above-mentioned background technology.

[0007] To achieve the above object, the present invention provides the following technical solution: an alarm system for automatically identifying irregularities in transshipment loading operations based on AI images, comprising:

[0008] An image acquisition module, which is responsible for capturing images of the transshipment loading operation site in real time through a high-definition camera or other image acquisition equipment;

[0009] An image preprocessing module, which is responsible for preprocessing the collected image or video data;

[0010] AI image recognition module, which uses AI technologies such as deep learning to analyze and recognize collected images in real time, and automatically identifies irregular behaviors in transshipment loading operations, such as improper operation, wrong loading sequence, unstable cargo stacking, etc.;

[0011] An alarm trigger module, which is responsible for automatically triggering the alarm mechanism according to the results of the AI ​​image recognition module, and supports turning on or off the corresponding alarm according to the navigation status and operation status of the ship;

[0012] The linkage alarm module is responsible for integrating the interfaces of multiple alarm systems to achieve data sharing and interaction. When the AI ​​image recognition module detects irregular behavior and triggers an alarm, it will link other related systems (such as the navigation monitoring system, safety management system, etc.) for further confirmation and alarm.

[0013] An alarm setting module is responsible for flexibly configuring alarm parameters according to user needs and actual conditions, including setting alarm levels, alarm processing timeliness, and notification methods;

[0014] The work responsibility supervision module is responsible for effectively supervising the work responsibilities of relevant personnel by setting the time limit for alarm confirmation and processing, recording and counting the processing of alarms, and providing a basis for subsequent assessment and evaluation;

[0015] System management module: The system management module is responsible for the configuration, maintenance and management of the entire system, including user authority management, system log recording, data backup and recovery functions, to ensure the stable operation of the system and the security of data.

[0016] Furthermore, the preprocessing steps of the image preprocessing module include:

[0017] Image denoising: Use median filtering or bilateral filtering to restore a clean image from a noisy image to reduce or eliminate the impact of noise on image quality while retaining detailed information in the image;

[0018] Image enhancement: The image is processed in sequence by histogram equalization (improving the contrast of the image by adjusting the grayscale distribution of the image), gamma correction (improving the brightness of the image by adjusting the gamma value of the image), and sharpening (retaining the details of the image by enhancing the edge information of the image) to improve the contrast, brightness, clarity and other characteristics of the image, improve the visual effect and recognizability of the image, and make the image more suitable for subsequent analysis and recognition tasks;

[0019] Image segmentation: Divide an image into several regions or objects with similar characteristics through threshold segmentation (dividing pixels in an image into different categories by setting one or more thresholds), edge detection (using edge detection operators such as Sobel and Canny to detect edges with obvious brightness changes in the image to identify objects), region growing (starting from a set of seed points, gradually merging surrounding pixels according to predetermined criteria such as pixel similarity until specific conditions are met), and deep learning methods (such as U-Net, Mask R-CNN and other deep learning models).

[0020] Furthermore, the AI ​​image recognition module specifically includes the following submodules:

[0021] State judgment submodule: used to judge the current navigation state (such as sailing, anchoring, mooring) and operation state (such as loading, unloading, idle state) of the ship;

[0022] Behavior recognition submodule: used to identify the specific behaviors of ships and operators and determine whether there are irregular operations;

[0023] Risk assessment submodule: used to assess the risk level in the current operation scenario and determine whether the threshold for triggering an alarm has been reached;

[0024] Time series analysis submodule: used to analyze the time series data of ship and operation status, identify abnormal changes and trigger alarms.

[0025] Furthermore, the status judgment submodule obtains real-time navigation status information by integrating a ship navigation data interface (such as an AIS system), and analyzes on-site video or image data through a deep learning model to judge the operation status.

[0026] Furthermore, the behavior recognition submodule uses a convolutional neural network to perform real-time analysis on image or video data, identify the actions of ships and operators, and compare the recognition results with a preset irregular behavior library to determine whether there is any irregularity in the behavior.

[0027] Furthermore, the risk assessment submodule comprehensively considers multiple dimensions such as the ship's navigation status, operating status, environmental factors (such as weather, sea conditions) and historical data to conduct risk assessment, and uses machine learning or statistical models to assess and predict risks.

[0028] Furthermore, the time series analysis submodule monitors and records the time series data of the ship's navigation trajectory, operating efficiency, and equipment status in real time, and uses time series analysis technology (such as ARIMA model, LSTM neural network, etc.) to detect abnormal changes.

[0029] Furthermore, the alarm trigger module includes four sections of trigger logic, which are used for triggering the alarm mechanism of the state judgment submodule, behavior recognition submodule, risk assessment submodule and timing analysis submodule respectively, and the trigger logic content is as follows:

[0030] Trigger logic 1: According to the preset rules or algorithms, when the ship is in a specific navigation state (such as sailing) and the operation state does not meet the specifications (such as illegal loading), an alarm is triggered;

[0031] Trigger logic 2: When irregular behavior is identified (such as not wearing safety equipment, operating equipment in violation of regulations, etc.), an alarm is triggered immediately;

[0032] Trigger logic three: When the risk assessment result reaches or exceeds the preset alarm threshold, an alarm is triggered;

[0033] Trigger logic 4: When an abnormal change is detected (such as a sudden change in speed, a significant drop in operating efficiency, etc.), determine whether an alarm needs to be triggered based on other factors.

[0034] Furthermore, the alarm setting module specifically includes the following submodules:

[0035] Level setting submodule: allows users to customize the alarm level according to the severity and urgency of different alarm types. Through fine-grained level division, the system can allocate resources more effectively and ensure that high-priority alarms are handled in a timely manner;

[0036] Time limit setting submodule: allows users to customize the processing time limit of alarms according to actual needs to ensure timely response by relevant personnel. The settings include alarm validity period (set the validity period of alarm information. Alarm information exceeding this time range will be considered expired and no more reminders will be given), processing time limit (set different processing time limits for alarms of different levels, requiring managers to respond to and process alarms within the specified time), and automatic closing function (for alarms that have not been processed for a long time, the system can automatically close and generate corresponding records for subsequent tracing and analysis);

[0037] Notification settings submodule: supports multiple alarm notification methods, such as SMS, email, etc., allowing users to customize the alarm notification method and recipients, so that people working away from home can receive alarm information in a timely manner. The settings include notification method (such as mobile phone SMS, email, APP push, system pop-up window, etc. Users can choose the appropriate notification method or combine multiple methods according to actual needs), recipients (allowing users to specify the recipients of alarm information, which can be individual personnel, specific departments or all members. In addition, the recipient can be automatically determined according to the alarm level or type), notification template (provides custom notification template function, users can set the format and style of notification content according to actual needs, so as to better convey alarm information).

[0038] Furthermore, the level setting submodule divides the alarm level into four levels, namely level 1, level 2, level 3, and level 4, from high to low, and each level sets clear judgment criteria. These criteria are determined based on factors such as the type, frequency, and potential hazards of irregular behaviors identified by the AI ​​image recognition module. It can be a composite condition based on the above factors or a threshold setting for a single condition. Different levels correspond to different severities and processing priorities. At the same time, the judgment criteria are further refined in combination with similar behaviors in historical data and their processing results. For example, if a certain type of irregular behavior has caused serious accidents many times in the past, then its alarm level may be set to a higher level.

[0039] The present invention provides an alarm system for automatically identifying irregularities in transshipment loading operations based on AI images, which has the following beneficial effects:

[0040] The present invention can automatically identify irregularities in transshipment loading operations based on AI images, and has a flexible alarm mechanism. It can set levels according to the type of alarm, and can be set at level one, level two, level three, and level four. At the same time, it can customize advanced processing time limits and alarm notification methods, and effectively supervise the work responsibilities of relevant personnel by setting the time limits for alarm confirmation and processing. In addition to the platform being able to view alarm information, it can also set SMS or email to receive alarms, and always pay attention to the supervision situation while working outside. This alarm system not only improves the level of supervision of operating specifications, but also provides strong protection for the safe navigation of ships through flexible alarm mechanisms and efficient navigation services. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 This is a logic block diagram of an alarm system for automatically identifying irregularities in transshipment loading operations based on AI images according to the present invention;

[0042] Figure 2 The present invention is a schematic diagram of an AI image recognition module-alarm trigger module of an alarm system for automatically identifying irregularities in transshipment loading operations based on AI images. DETAILED DESCRIPTION

[0043] The following embodiments of the present invention are described in further detail in conjunction with the accompanying drawings and examples. The following examples are used to illustrate the present invention, but are not intended to limit the scope of the present invention.

[0044] like Figure 1-Figure 2 As shown, an alarm system based on AI images to automatically identify irregularities in transshipment loading operations includes:

[0045] Image acquisition module: responsible for capturing the images of the transshipment loading operation site in real time through a high-definition camera;

[0046] Image preprocessing module: responsible for preprocessing the collected image or video data. The preprocessing steps include:

[0047] Image denoising: Use median filtering or bilateral filtering to restore a clean image from a noisy image to reduce or eliminate the impact of noise on image quality while retaining detailed information in the image;

[0048] Image enhancement: The image is processed in sequence by histogram equalization (improving the contrast of the image by adjusting the grayscale distribution of the image), gamma correction (improving the brightness of the image by adjusting the gamma value of the image), and sharpening (retaining the details of the image by enhancing the edge information of the image) to improve the contrast, brightness, clarity and other characteristics of the image, improve the visual effect and recognizability of the image, and make the image more suitable for subsequent analysis and recognition tasks;

[0049] Image segmentation: divide an image into several regions or objects with similar characteristics through threshold segmentation (by setting one or more thresholds to divide the pixels in the image into different categories), edge detection (using edge detection operators such as Sobel and Canny to detect edges with obvious brightness changes in the image to identify objects), region growing (starting from a set of seed points, gradually merging surrounding pixels according to predetermined criteria such as pixel similarity until specific conditions are met), and deep learning methods (such as U-Net, Mask R-CNN and other deep learning models);

[0050] AI image recognition module: It is responsible for using AI technologies such as deep learning to analyze and recognize the collected images in real time, and automatically identify irregular behaviors in transshipment loading operations, such as improper operation, wrong loading sequence, unstable cargo stacking, etc. The AI ​​image recognition module specifically includes the following sub-modules:

[0051] Status judgment submodule: used to judge the current navigation status (such as sailing, anchoring, mooring) and operating status (such as loading, unloading, idle) of the ship; the status judgment submodule obtains real-time navigation status information through the AIS system, and analyzes the on-site video or image data through the deep learning model to judge the operating status;

[0052] Behavior recognition submodule: used to identify the specific behaviors of ships and operators and determine whether there are irregular operations. The behavior recognition submodule uses convolutional neural networks to perform real-time analysis on image or video data, identify the actions of ships and operators, and compare the recognition results with the preset irregular behavior library to determine whether there are irregular behaviors.

[0053] Risk assessment submodule: used to assess the risk level in the current operation scenario and determine whether the threshold for triggering an alarm has been reached. The risk assessment submodule comprehensively considers multiple dimensions such as the ship's navigation status, operation status, environmental factors (such as weather and sea conditions), and historical data to conduct risk assessment, and uses machine learning or statistical models to assess and predict risks.

[0054] Time series analysis submodule: used to analyze the time series data of ships and operating status, identify abnormal changes and trigger alarms; the time series analysis submodule monitors and records the time series data of ship navigation trajectory, operating efficiency and equipment status in real time, and uses time series analysis technology (such as ARIMA model, LSTM neural network, etc.) to detect abnormal changes;

[0055] Alarm trigger module: responsible for automatically triggering the alarm mechanism according to the results of the AI ​​image recognition module, and supports turning on or off the corresponding alarm according to the navigation status and operation status of the ship; the alarm trigger module includes four sections of trigger logic, which are used for triggering the alarm mechanism of the state judgment submodule, behavior recognition submodule, risk assessment submodule and timing analysis submodule, and the trigger logic content is as follows:

[0056] Trigger logic 1: According to the preset rules or algorithms, when the ship is in a specific navigation state (such as sailing) and the operation state does not meet the specifications (such as illegal loading), an alarm is triggered;

[0057] Trigger logic 2: When irregular behavior is identified (such as not wearing safety equipment, operating equipment in violation of regulations, etc.), an alarm is triggered immediately;

[0058] Trigger logic three: When the risk assessment result reaches or exceeds the preset alarm threshold, an alarm is triggered;

[0059] Trigger logic 4: When an abnormal change is detected (such as a sudden change in speed, a sharp drop in operating efficiency, etc.), determine whether an alarm needs to be triggered in combination with other factors;

[0060] Linkage alarm module: responsible for integrating the interfaces of multiple alarm systems to achieve data sharing and interaction. When the AI ​​image recognition module detects irregular behavior and triggers an alarm, it links other related systems (such as the navigation monitoring system, safety management system, etc.) for further confirmation and alarm.

[0061] Alarm setting module: responsible for flexibly configuring alarm parameters according to user needs and actual conditions, including setting alarm levels, alarm processing timeliness, and notification methods; the alarm setting module specifically includes the following sub-modules:

[0062] Level setting submodule: allows users to customize the alarm level according to the severity and urgency of different alarm types. Through fine level division, the system can allocate resources more effectively to ensure that high-priority alarms are handled in a timely manner; the level setting submodule divides the alarm level into four levels, from high to low: level 1, level 2, level 3, and level 4, and each level has clear judgment criteria. These criteria are determined based on the type, frequency, potential hazards and other factors of the irregular behavior identified by the AI ​​image recognition module. They can be composite conditions based on the above factors or threshold settings for a single condition. Different levels correspond to different severity and processing priorities. In addition, the judgment criteria are further refined in combination with similar behaviors in historical data and their processing results. For example, if a certain type of irregular behavior has caused serious accidents many times in the past, its alarm level may be set to a higher level;

[0063] Time limit setting submodule: allows users to customize the processing time limit of alarms according to actual needs to ensure timely response by relevant personnel. The settings include alarm validity period (set the validity period of alarm information. Alarm information exceeding this time range will be considered expired and no more reminders will be given), processing time limit (set different processing time limits for alarms of different levels, requiring managers to respond to and process alarms within the specified time), and automatic closing function (for alarms that have not been processed for a long time, the system can automatically close and generate corresponding records for subsequent tracing and analysis);

[0064] Notification setting submodule: supports multiple alarm notification methods, such as SMS, email, etc., allowing users to customize the alarm notification method and recipients, so that people working away from home can receive alarm information in time. The settings include notification method (such as mobile phone SMS, email, APP push, system pop-up window, etc. Users can choose the appropriate notification method or combine multiple methods according to actual needs), recipients (allowing users to specify the recipients of alarm information, which can be individual personnel, specific departments or all members. In addition, the recipients can be automatically determined according to the alarm level or type), and notification templates (providing the function of customizing notification templates. Users can set the format and style of notification content according to actual needs to better convey alarm information);

[0065] Job Responsibility Supervision Module: Responsible for effectively supervising the job responsibilities of relevant personnel by setting the time limit for alarm confirmation and processing, recording and counting the processing of alarms, and providing a basis for subsequent assessment and evaluation;

[0066] System management module: responsible for the configuration, maintenance and management of the entire system, including user authority management, system log recording, data backup and recovery functions, to ensure the stable operation of the system and the security of data.

[0067] In summary, combined Figure 1-Figure 2 As shown, the use process of the alarm system based on AI image automatic recognition of irregular transshipment loading operations is as follows:

[0068] S1, the image acquisition module uses a high-definition camera or other image acquisition equipment to capture the images of the transshipment loading operation site in real time, and sends them to the image preprocessing module to perform image denoising, image enhancement, and image segmentation on the original image data;

[0069] S2, AI image recognition module analyzes and recognizes the collected images in real time through the state judgment submodule, behavior recognition submodule, risk assessment submodule and timing analysis submodule, automatically identifies irregular behaviors in the transshipment loading operation, and sends the recognition results to the alarm trigger module;

[0070] S3, the alarm trigger module selects whether to trigger the alarm mechanism based on the results of the AI ​​image recognition module and the alarm trigger logic, and the linkage alarm module will link other related systems for further confirmation and alarm; at the same time, users can flexibly configure alarm parameters through the alarm setting module, including setting the alarm level, alarm processing time limit and notification method;

[0071] S4. The job responsibility supervision module effectively supervises the job responsibilities of relevant personnel by setting the time limit for alarm confirmation and processing, records and compiles statistics on the processing of alarms, and provides a basis for subsequent assessment and evaluation.

[0072] The embodiments of the present invention are given for the purpose of illustration and description, and are not intended to be exhaustive or to limit the invention to the disclosed forms. Many modifications and variations will be apparent to those of ordinary skill in the art. The embodiments are selected and described in order to better illustrate the principles and practical applications of the present invention and to enable those of ordinary skill in the art to understand the present invention and thereby design various embodiments with various modifications suitable for specific uses.

Claims

1. An alarm system based on AI images to automatically identify irregularities in transshipment loading operations, characterized in that: include: An image acquisition module, which is responsible for capturing images of the transshipment loading operation site in real time through a high-definition camera or other image acquisition equipment; An image preprocessing module, which is responsible for preprocessing the collected image or video data; An AI image recognition module, which is responsible for using AI technology to perform real-time analysis and recognition of the collected images, and automatically identify irregular behaviors in the transshipment loading operation; An alarm trigger module, which is responsible for automatically triggering the alarm mechanism according to the results of the AI ​​image recognition module, and supports turning on or off the corresponding alarm according to the navigation status and operation status of the ship; The linkage alarm module is responsible for integrating the interfaces of multiple alarm systems to achieve data sharing and interaction. When the AI ​​image recognition module detects irregular behavior and triggers an alarm, it links other related systems for further confirmation and alarm. An alarm setting module, which is responsible for flexibly configuring alarm parameters according to user needs and actual conditions, including setting alarm levels, alarm processing timeliness, and notification methods; The work responsibility supervision module is responsible for effectively supervising the work responsibilities of relevant personnel by setting the time limit for alarm confirmation and processing, recording and counting the processing of alarms, and providing a basis for subsequent assessment and evaluation; System management module: The system management module is responsible for the configuration, maintenance and management of the entire system, including user authority management, system log recording, data backup and recovery functions, to ensure the stable operation of the system and the security of data.

2. According to claim 1, the alarm system for automatically identifying irregularities in transshipment loading operations based on AI images is characterized in that: The preprocessing steps of the image preprocessing module include: Image denoising: Use median filtering or bilateral filtering to restore a clean image from a noisy image to reduce or eliminate the impact of noise on image quality while retaining detailed information in the image; Image enhancement: Use histogram equalization, gamma correction, and sharpening to process the image in sequence to improve the contrast, brightness, and clarity of the image, improve the visual effect and recognizability of the image, and make the image more suitable for subsequent analysis and recognition tasks; Image segmentation: Divide the image into several regions or objects with similar characteristics through threshold segmentation, edge detection, region growing, and deep learning methods.

3. The alarm system for automatically identifying irregularities in transshipment loading operations based on AI images according to claim 1 is characterized in that: The AI ​​image recognition module specifically includes the following submodules: Status judgment submodule: used to judge the current navigation status and operation status of the ship; Behavior recognition submodule: used to identify the specific behaviors of ships and operators and determine whether there are irregular operations; Risk assessment submodule: used to assess the risk level in the current operation scenario and determine whether the threshold for triggering an alarm has been reached; Time series analysis submodule: used to analyze the time series data of ship and operation status, identify abnormal changes and trigger alarms.

4. The alarm system for automatically identifying irregularities in transshipment loading operations based on AI images according to claim 3 is characterized in that: The status judgment submodule obtains real-time navigation status information through an integrated ship navigation data interface, and analyzes on-site video or image data through a deep learning model to judge the operation status.

5. The AI ​​image-based automatic recognition and warning system for irregular transshipment loading operations according to claim 3 is characterized in that: The behavior recognition submodule uses a convolutional neural network to perform real-time analysis on image or video data, identify the actions of ships and operators, and compare the recognition results with a preset irregular behavior library to determine whether the behavior is irregular.

6. The AI ​​image-based automatic recognition and warning system for irregular transshipment loading operations according to claim 3 is characterized in that: The risk assessment submodule comprehensively considers the ship's navigation status, operating status, environmental factors and historical data to conduct risk assessment, and uses machine learning or statistical models to evaluate and predict risks.

7. The AI ​​image-based automatic recognition and warning system for irregular transshipment loading operations according to claim 3 is characterized in that: The time series analysis submodule monitors and records the time series data of the ship's navigation track, operating efficiency, and equipment status in real time, and uses time series analysis technology to detect abnormal changes.

8. The AI ​​image-based automatic recognition and warning system for irregular transshipment loading operations according to claim 1 is characterized in that: The alarm trigger module includes four sections of trigger logic, which are used for triggering the alarm mechanism of the state judgment submodule, behavior recognition submodule, risk assessment submodule and timing analysis submodule respectively, and the trigger logic content is as follows: Trigger logic 1: According to the preset rules or algorithms, when the ship is in a specific navigation state and the operating status does not meet the specifications, an alarm is triggered; Trigger logic 2: When irregular behavior is identified, an alarm is triggered immediately; Trigger logic three: When the risk assessment result reaches or exceeds the preset alarm threshold, an alarm is triggered; Trigger logic 4: When an abnormal change is detected, determine whether an alarm needs to be triggered based on other factors.

9. The AI ​​image-based automatic recognition and warning system for irregular transshipment loading operations according to claim 1 is characterized in that: The alarm setting module specifically includes the following submodules: Level setting submodule: allows users to customize the alarm level according to the severity and urgency of different alarm types. Through detailed level division, the system can allocate resources more effectively to ensure that high-priority alarms are handled in a timely manner; Time limit setting submodule: allows users to customize the alarm processing time limit according to actual needs to ensure timely response by relevant personnel. The settings include alarm validity period, processing time limit, and automatic shutdown function; Notification settings submodule: supports multiple alarm notification methods, allowing users to customize the alarm notification method and recipients, so that people working away from home can receive alarm information in a timely manner. The settings include notification method, recipients, and notification templates.

10. The AI ​​image-based automatic recognition and warning system for irregular transshipment loading operations according to claim 9 is characterized in that: The level setting submodule divides the alarm level into four levels, namely level 1, level 2, level 3, and level 4 from high to low, and sets clear judgment criteria for each level. Different levels correspond to different severities and processing priorities.

Images