Safety management method and system applied to video monitoring of ship unloading system

Abstract

The present invention relates to the technical field of monitoring of ship unloading, and disclosed are a safety management method and system applied to video monitoring of a ship unloading system. The method comprises: acquiring a plurality of standard factors of the ship unloading system, and determining levels of initial areas of the ship unloading system; on the basis of the levels of the initial areas of the ship unloading system, screening out key areas, and performing video monitoring of the key areas; on the basis of historical video content of the key areas, pre-defining standard actions in the key areas; on the basis of real-time video content of the key areas, sequentially performing object identification, motion action detection, and abnormal action identification, thereby determining a degree of abnormality of each key area; and on the basis of the degree of abnormality of each key area, performing risk assessment of the ship unloading system, and on the basis of a result of the risk assessment, formulating a safety strategy. Thus, the actions of ship unloading are linked with safety protection, thereby improving the performance of safety protection.

Description

A safety management method and system for monitoring video of ship unloading system Technical Field

[0001] The present application relates to the technical field of ship unloading monitoring, and more specifically, to a safety management method and system for monitoring videos of a ship unloading system. Background Art

[0002] The safety management of the ship unloading system is a critical task, and surveillance video plays an important role in it. By applying advanced technologies and methods, the working process of the ship unloader can be effectively managed and monitored to ensure the safety of operations and the lives of workers.

[0003] Video surveillance systems are fundamental to ship unloading monitoring technology. High-definition cameras and video surveillance systems capture real-time video images of the ship unloader's operating area, providing a data foundation for subsequent safety management.

[0004] The safety management method applied to the surveillance video of the ship unloading system involves the integrated application of multiple technologies, including video surveillance systems, motion detection, object recognition, and abnormal behavior recognition. These technologies combine to achieve comprehensive monitoring of the ship unloader working area, anomaly detection, and timely alarms, thereby ensuring the safety of workers and the smooth operation of the ship unloading operation.

[0005] In the existing technology, the safety management of the ship unloading system is not linked to the unloading behavior, resulting in low safety protection and failure to effectively ensure the reasonable operation of the ship unloading system.

[0006] Therefore, how to improve the safety and protection of the ship unloading system is a technical problem that needs to be solved.

[0007] Summary of the Invention

[0008] The present invention provides a safety management method for monitoring video of a ship unloading system, which is used to solve the technical problem of low safety protection of the ship unloading system in the prior art. The method comprises:

[0009] Obtain multiple standard factors for the ship unloading system and determine the initial area classification of the ship unloading system;

[0010] Screen out key areas based on the initial area level of the ship unloading system and conduct video surveillance of key areas;

[0011] Predefine standard actions within the key area based on historical video content of the key area;

[0012] Based on the real-time video content of key areas, object recognition, motion detection, and abnormal action recognition are performed in sequence to determine the degree of abnormality in each key area;

[0013] Conduct risk assessment of the ship unloading system according to the degree of abnormality in each key area, and formulate safety strategies based on the results of the risk assessment.

[0014] In some embodiments of the present application, before obtaining a plurality of standard factors of the ship unloading system, the method further includes:

[0015] Obtain functional and process information of the ship unloading system;

[0016] Determine the number of process steps based on process information, and determine the disassembly granularity based on the number of process steps;

[0017] The ship unloading system is divided into multiple functions according to the disassembly granularity and functional information, and the initial area involved in each function is determined, thereby dividing the ship unloading system into regions.

[0018] In some embodiments of the present application, the multiple standard factors include other factors, safety factors, and maintenance factors;

[0019] Among them, P is the initial regional level, γ1 is the influence weight of other factors, n is the number of other factors, α i is the weight corresponding to the other factors of item i, Q i is the parameter of the i-th other factor, γ2 is the influence weight of the safety factor, m is the number of safety factors, β j is the weight corresponding to the jth safety factor, W j is the parameter of the jth safety factor, k1 is the first constant, exp represents the exponential function, b is the number of maintenance times in the initial area, δ k is the weight of the kth maintenance, E k is the impact of the k-th maintenance, k2 is the second constant, and [] is the rounding symbol.

[0020] In some embodiments of the present application, critical areas are screened out based on the initial area level of the ship unloading system, including:

[0021] If the initial area level exceeds the first level, the initial area is divided into a key area;

[0022] If the initial area level does not exceed the first level and exceeds the second level, the initial area is divided into the area to be studied;

[0023] Calculate the degree of correlation between the area to be investigated and the key area, as well as the distance between the area to be investigated and the key area;

[0024] If the correlation degree exceeds the correlation degree threshold, or the distance is less than the distance threshold, or the correlation degree exceeds the correlation degree threshold and the distance is less than the distance threshold, the area to be studied is divided into a key area;

[0025] Otherwise, the area to be studied is not classified as a key area.

[0026] In some embodiments of the present application, standard actions within a key area are predefined based on historical video content of the key area, including:

[0027] Under normal conditions, the task instructions of the ship unloading system are obtained, and the motion trajectory of the target object in the key area is identified. The task instructions and the motion trajectory of the target object are combined as the standard action in the key area.

[0028] In some embodiments of the present application, object recognition, motion detection, and abnormal motion recognition are sequentially performed based on the real-time video content of the key areas to determine the abnormality level of each key area, including:

[0029] Object recognition, including:

[0030] Extract the features of the target object and describe them as feature vectors that can be matched and identified, and identify the target object through machine learning classification;

[0031] Motion detection, including:

[0032] Motion detection of target objects using optical flow method;

[0033] Abnormal action detection, including:

[0034] Anomalies are detected on the target object's motion using a preset neural network model to obtain a first abnormality degree, and a second abnormality degree is obtained by comparing the target object's motion with a standard motion;

[0035] If the difference between the first abnormality degree and the second abnormality degree is less than a preset difference, the first abnormality degree is used as the abnormality degree of the key area;

[0036] Otherwise, the average of the first abnormality degree and the second abnormality degree is taken as the abnormality degree of the key area.

[0037] In some embodiments of the present application, risk assessment of the ship unloading system is performed according to the abnormality level of each key area, including: determining the level of risk assessment according to the calibration amount L,

[0038] Where n is the number of key areas, ε i is the risk weight corresponding to the i-th key area, R i is the abnormality degree corresponding to the i-th key area, exp is the exponential function, n0 is the specified value, and k3 is the third constant.

[0039] Correspondingly, the present application also provides a safety management system for monitoring video of a ship unloading system, the system comprising:

[0040] The first module is used to obtain multiple standard factors of the ship unloading system and determine the initial regional level of the ship unloading system;

[0041] The second module is used to screen out key areas based on the initial area level of the ship unloading system and conduct video monitoring of the key areas;

[0042] The third module is used to predefine standard actions in the key area based on the historical video content of the key area;

[0043] The fourth module is used to perform object recognition, motion detection, and abnormal action recognition based on the real-time video content of the key areas, thereby determining the degree of abnormality in each key area;

[0044] The fifth module is used to conduct risk assessment of the ship unloading system according to the degree of abnormality in each key area and formulate safety strategies based on the results of the risk assessment.

[0045] By applying the above technical solutions, multiple standard factors of the ship unloading system are obtained, and the initial area level of the ship unloading system is determined; key areas are screened out according to the initial area level of the ship unloading system, and video surveillance is performed on the key areas; standard actions in the key areas are pre-defined based on the historical video content of the key areas; object recognition, motion detection, and abnormal action recognition are performed in sequence based on the real-time video content of the key areas, thereby determining the degree of abnormality of each key area; risk assessment of the ship unloading system is performed based on the degree of abnormality of each key area, and a safety strategy is formulated based on the results of the risk assessment. This application improves safety protection by pre-defining standard actions in the key areas according to the historical video content of the key areas, and performing object recognition, motion detection, and abnormal action recognition in sequence, thereby linking the unloading actions with safety protection. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0047] FIG1 is a schematic flow chart showing a method for managing security of monitoring videos of a ship unloading system according to an embodiment of the present invention;

[0048] FIG2 shows a schematic structural diagram of a safety management system for monitoring video of a ship unloading system proposed in an embodiment of the present invention. DETAILED DESCRIPTION

[0049] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0050] The present application provides a method for managing security of monitoring video of a ship unloading system, as shown in FIG1 , including the following steps:

[0051] Step S101: obtaining multiple standard factors of the ship unloading system and determining the initial regional level of the ship unloading system.

[0052] In this embodiment, the multiple areas of the ship unloading system are graded, and the higher the grade, the more important it is.

[0053] In some embodiments of the present application, before obtaining a plurality of standard factors of the ship unloading system, the method further includes:

[0054] Obtain functional and process information of the ship unloading system;

[0055] Determine the number of process steps based on process information, and determine the disassembly granularity based on the number of process steps;

[0056] The ship unloading system is divided into multiple functions according to the disassembly granularity and functional information, and the initial area involved in each function is determined, thereby dividing the ship unloading system into regions.

[0057] In this embodiment, the number of process steps corresponds to a disassembly granularity.

[0058] In some embodiments of the present application, the multiple standard factors include other factors, safety factors, and maintenance factors;

[0059] Among them, P is the initial regional level, γ1 is the influence weight of other factors, n is the number of other factors, α i is the weight corresponding to the other factors of item i, Q i is the parameter of the i-th other factor, γ2 is the influence weight of the safety factor, m is the number of safety factors, β j is the weight corresponding to the jth safety factor, W j is the parameter of the jth safety factor, k1 is the first constant, exp represents the exponential function, b is the number of maintenance times in the initial area, δ k is the weight of the kth maintenance, E kis the impact of the k-th maintenance, k2 is the second constant, and [] is the rounding symbol.

[0060] In this embodiment, Indicates the correction of maintenance factors to other factors and safety factors.

[0061] Step S102: screening out key areas according to the initial area level of the ship unloading system, and performing video monitoring on the key areas.

[0062] In this embodiment, relatively critical areas are screened out based on the initial area levels, and these areas are subjected to focused video surveillance.

[0063] In some embodiments of the present application, critical areas are screened out based on the initial area level of the ship unloading system, including:

[0064] If the initial area level exceeds the first level, the initial area is divided into a key area;

[0065] If the initial area level does not exceed the first level and exceeds the second level, the initial area is divided into the area to be studied;

[0066] Calculate the degree of correlation between the area to be investigated and the key area, as well as the distance between the area to be investigated and the key area;

[0067] If the correlation degree exceeds the correlation degree threshold, or the distance is less than the distance threshold, or the correlation degree exceeds the correlation degree threshold and the distance is less than the distance threshold, the area to be studied is divided into a key area;

[0068] Otherwise, the area to be studied is not classified as a key area.

[0069] Step S103 : pre-define standard actions in the key area according to the historical video content of the key area.

[0070] In some embodiments of the present application, standard actions within a key area are predefined based on historical video content of the key area, including:

[0071] Under normal conditions, the task instructions of the ship unloading system are obtained, and the motion trajectory of the target object in the key area is identified. The task instructions and the motion trajectory of the target object are combined as the standard action in the key area.

[0072] In this embodiment, the normal state means that the ship unloading system operates normally and the movement trajectory is normal.

[0073] Step S104 , performing object recognition, motion detection, and abnormal motion recognition in sequence based on the real-time video content of the key areas, thereby determining the abnormality level of each key area.

[0074] In some embodiments of the present application, object recognition, motion detection, and abnormal motion recognition are sequentially performed based on the real-time video content of the key areas to determine the abnormality level of each key area, including:

[0075] Object recognition, including:

[0076] Extract the features of the target object and describe them as feature vectors that can be matched and identified, and identify the target object through machine learning classification;

[0077] Motion detection, including:

[0078] Motion detection of target objects using optical flow method;

[0079] Abnormal action detection, including:

[0080] Anomalies are detected on the target object's motion using a preset neural network model to obtain a first abnormality degree, and a second abnormality degree is obtained by comparing the target object's motion with a standard motion;

[0081] If the difference between the first abnormality degree and the second abnormality degree is less than a preset difference, the first abnormality degree is used as the abnormality degree of the key area;

[0082] Otherwise, the average of the first abnormality degree and the second abnormality degree is taken as the abnormality degree of the key area.

[0083] In this embodiment, motion detection involves: differential image analysis: identifying moving objects by analyzing the differences between consecutive frames. Common methods include frame difference analysis, optical flow analysis, and background modeling. Optical flow analysis: detecting the trajectory and velocity of an object by analyzing the movement of pixels between consecutive frames. Region-based analysis: dividing video frames into different regions and performing pixel-level motion analysis on each region to detect moving targets. Object recognition: feature extraction and description: extracting features from video frames using computer vision algorithms (such as SIFT, SURF, and HOG) and describing them as feature vectors that can be identified and matched. Machine learning classifier: training a classifier using machine learning algorithms (such as support vector machines, random forests, and convolutional neural networks) to identify specific object categories from video frames. Abnormal behavior recognition: behavioral modeling: establishing a model or rules for normal behavior by observing samples of normal behavior. Abnormal behavior is then detected based on the differences between the behavior in the real-time video stream and the model. Statistical analysis: modeling and analyzing the behavior in the video based on statistical models and rules to identify abnormal behavior that does not conform to normal behavior patterns. Deep learning methods: Use methods such as deep neural networks and recurrent neural networks to learn behavioral patterns in video sequences and detect abnormal behaviors.

[0084] Step S105: Perform risk assessment on the ship unloading system according to the abnormality level of each key area, and formulate a safety strategy based on the results of the risk assessment.

[0085] In this embodiment, a safety policy of a corresponding level is formulated according to the level of risk assessment, so as to ensure the safe operation of the ship unloading system.

[0086] In some embodiments of the present application, risk assessment of the ship unloading system is performed according to the abnormality level of each key area, including: determining the level of risk assessment according to the calibration amount L,

[0087] Where n is the number of key areas, ε i is the risk weight corresponding to the i-th key area, R i is the abnormality degree corresponding to the i-th key area, exp is the exponential function, n0 is the specified value, and k3 is the third constant.

[0088] In this embodiment, each proofreading amount corresponds to a risk assessment level.

[0089] In this embodiment, there may be a certain degree of mutual influence between the multiple key areas, so Make corrections to key areas.

[0090] Through the description of the above embodiments, those skilled in the art can clearly understand that the present invention can be implemented through hardware or by using software plus the necessary general hardware platform. Based on this understanding, the technical solution of the present invention can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (such as a CD-ROM, a USB flash drive, a mobile hard disk, etc.) and includes a number of instructions for enabling a computer device (such as a personal computer, a server, or a network device) to execute the methods described in various implementation scenarios of the present invention.

[0091] By applying the above technical solutions, multiple standard factors of the ship unloading system are obtained, and the initial area level of the ship unloading system is determined; the key areas are screened out according to the initial area level of the ship unloading system, and video surveillance is performed on the key areas; standard actions in the key areas are pre-defined based on the historical video content of the key areas; object recognition, motion detection, and abnormal action recognition are performed in sequence based on the real-time video content of the key areas, thereby determining the degree of abnormality of each key area; risk assessment of the ship unloading system is performed based on the degree of abnormality of each key area, and a safety strategy is formulated based on the results of the risk assessment. This application improves safety protection by pre-defining standard actions in the key areas according to the historical video content of the key areas, and performing object recognition, motion detection, and abnormal action recognition in sequence, thereby linking the unloading actions with safety protection.

[0092] Correspondingly, the present application also provides a safety management system for monitoring video of a ship unloading system, as shown in FIG2 , the system includes:

[0093] The first module 201 is used to obtain multiple standard factors of the ship unloading system and determine the initial area level of the ship unloading system;

[0094] The second module 202 is used to screen out key areas according to the initial area level of the ship unloading system and perform video monitoring on the key areas;

[0095] The third module 203 is used to predefine standard actions in the key area based on the historical video content of the key area;

[0096] The fourth module 204 is configured to sequentially perform object recognition, motion detection, and abnormal motion recognition based on the real-time video content of the key areas, thereby determining the abnormality level of each key area;

[0097] The fifth module 205 is used to perform risk assessment on the ship unloading system according to the abnormality level of each key area and formulate a safety strategy based on the results of the risk assessment.

[0098] In some embodiments of the present application, the system is further used to:

[0099] Obtain functional and process information of the ship unloading system;

[0100] Determine the number of process steps based on process information, and determine the disassembly granularity based on the number of process steps;

[0101] The ship unloading system is divided into multiple functions according to the disassembly granularity and functional information, and the initial area involved in each function is determined, thereby dividing the ship unloading system into regions.

[0102] In some embodiments of the present application, the first module 201 is used to calculate the initial region level P based on multiple standard factors:

[0103] Several standard factors include other factors, safety factors and maintenance factors;

[0104] Among them, γ1 is the influence weight of other factors, n is the number of other factors, α i is the weight corresponding to the other factors of item i, Q i is the parameter of the i-th other factor, γ2 is the influence weight of the safety factor, m is the number of safety factors, β j is the weight corresponding to the jth safety factor, W j is the parameter of the jth safety factor, k1 is the first constant, exp represents the exponential function, b is the number of maintenance times in the initial area, δk is the weight of the kth maintenance, E k is the impact of the k-th maintenance, k2 is the second constant, and [] is the rounding symbol.

[0105] In some embodiments of the present application, the second module 202 is configured to:

[0106] If the initial area level exceeds the first level, the initial area is divided into a key area;

[0107] If the initial area level does not exceed the first level and exceeds the second level, the initial area is divided into the area to be studied;

[0108] Calculate the degree of correlation between the area to be investigated and the key area, as well as the distance between the area to be investigated and the key area;

[0109] If the correlation degree exceeds the correlation degree threshold, or the distance is less than the distance threshold, or the correlation degree exceeds the correlation degree threshold and the distance is less than the distance threshold, the area to be studied is divided into a key area;

[0110] Otherwise, the area to be studied is not classified as a key area.

[0111] In some embodiments of the present application, the third module 203 is used to:

[0112] Under normal conditions, the task instructions of the ship unloading system are obtained, and the motion trajectory of the target object in the key area is identified. The task instructions and the motion trajectory of the target object are combined as the standard action in the key area.

[0113] In some embodiments of the present application, the fourth module 204 is configured to:

[0114] Object recognition, including:

[0115] Extract the features of the target object and describe them as feature vectors that can be matched and identified, and identify the target object through machine learning classification;

[0116] Motion detection, including:

[0117] Motion detection of target objects using optical flow method;

[0118] Abnormal action detection, including:

[0119] Anomalies are detected on the target object's motion using a preset neural network model to obtain a first abnormality degree, and a second abnormality degree is obtained by comparing the target object's motion with a standard motion;

[0120] If the difference between the first abnormality degree and the second abnormality degree is less than a preset difference, the first abnormality degree is used as the abnormality degree of the key area;

[0121] Otherwise, the average of the first abnormality degree and the second abnormality degree is taken as the abnormality degree of the key area.

[0122] In some embodiments of the present application, the fifth module 205 is used to calculate the proofreading amount L:

[0123] Where n is the number of key areas, ε i is the risk weight corresponding to the i-th key area, R i is the abnormality degree corresponding to the i-th key area, exp is the exponential function, n0 is the specified value, and k3 is the third constant.

[0124] Those skilled in the art will appreciate that the modules in the system of the implementation scenario can be distributed in the system of the implementation scenario according to the implementation scenario description, or can be modified accordingly and located in one or more systems different from the implementation scenario. The modules of the above implementation scenario can be combined into one module or further divided into multiple submodules.

[0125] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A safety management method for monitoring video of a ship unloading system, characterized in that: The method comprises: Obtain multiple standard factors of the ship unloading system and determine the level of the initial area of ​​the ship unloading system; Screen out key areas according to the level of the initial area of ​​the ship unloading system and conduct video surveillance on the key areas; Predefine standard actions within the key area based on historical video content within the key area; Based on the real-time video content of the key area, object recognition, motion detection and abnormal action recognition are performed in sequence to determine the degree of abnormality in the key area; Conduct risk assessment of the ship unloading system according to the degree of abnormality in key areas, and formulate safety strategies based on the results of the risk assessment.

2. The safety management method for monitoring video of a ship unloading system according to claim 1, characterized in that: Before obtaining a plurality of standard factors of the ship unloading system, the method further comprises: Obtain functional and process information of the ship unloading system; Determine the number of process steps based on process information, and determine the disassembly granularity based on the number of process steps; The ship unloading system is divided into multiple functions according to the disassembly granularity and functional information, and the initial area involved in each function is determined, so as to divide the ship unloading system into regions.

3. The safety management method for monitoring video of a ship unloading system according to claim 2, characterized in that: Multiple standard factors include other factors, safety factors and maintenance factors; Among them, P is the level of the initial region, γ1 is the influence weight of other factors, n is the number of other factors, α i is the weight corresponding to the other factors of item i, Q i is the parameter of the i-th other factor, γ2 is the influence weight of the safety factor, m is the number of safety factors, β j is the weight corresponding to the jth safety factor, W j is the parameter of the jth safety factor, k1 is the first constant, exp represents the exponential function, b is the number of maintenance times in the initial area, δ k is the weight of the kth maintenance, E k is the impact of the kth maintenance, k2 is the second constant, and [] is the rounding symbol.

4. The safety management method for monitoring video of a ship unloading system according to claim 1, characterized in that: According to the level of the initial area of ​​the ship unloading system, the key areas are screened out, including: If the level of the initial area exceeds the first level, the initial area is divided into a critical area; If the level of the initial area does not exceed the first level and exceeds the second level, the initial area Divided into areas to be studied; Calculate the degree of correlation between the area to be investigated and the key area, as well as the distance between the area to be investigated and the key area; If the correlation degree exceeds the correlation degree threshold, or the distance is less than the distance threshold, or the correlation degree exceeds the correlation degree threshold and the distance is less than the distance threshold, the area to be studied is divided into a key area; Otherwise, the area to be investigated is not classified as a key area.

5. The safety management method for monitoring video of a ship unloading system according to claim 1, characterized in that: Standard actions in key areas are predefined based on historical video content in the key areas, including: Under normal conditions, the task instructions of the ship unloading system are obtained, and the motion trajectory of the target object in the key area is identified. The task instructions and the motion trajectory of the target object are combined as the standard action in the key area.

6. The safety management method for monitoring video of a ship unloading system according to claim 1, characterized in that: Based on the real-time video content of key areas, object recognition, motion detection, and abnormal action recognition are performed in sequence to determine the degree of abnormality in each key area, including: Object recognition, including: Extract the features of the target object and describe the features as feature vectors that can be matched and identified, and identify the target object through machine learning classification; Motion detection, including: Motion detection of target objects through optical flow method; Abnormal action detection, including: Anomaly detection is performed on the motion of the target object by using a preset neural network model to obtain a first abnormality degree, and a second abnormality degree is obtained by comparing the motion of the target object with a standard motion; If the difference between the first abnormality degree and the second abnormality degree is less than the preset difference, the first abnormality degree is used as the abnormality degree of the key area; Otherwise, the average of the first abnormality degree and the second abnormality degree is taken as the abnormality degree of the key area.

7. The method according to claim 1, characterized in that Conduct risk assessment of the ship unloading system based on the degree of abnormality in key areas, including: determining the level of risk assessment based on the amount of calibration, Among them, L is the proofreading amount, n is the number of key areas, ε i is the risk weight corresponding to the ith key area, R i is the abnormality degree corresponding to the i-th key area, exp is the exponential function, n0 is the specified value, and k3 is the third constant.

8. A safety management system for monitoring video of a ship unloading system, characterized in that: The system comprises: The first module is used to obtain multiple standard factors of the ship unloading system and determine the level of the initial area of ​​the ship unloading system; The second module is used to screen out key areas according to the level of the initial area of ​​the ship unloading system and conduct video monitoring of the key areas; The third module is used to predefine standard actions in the key area according to the historical video content of the key area; The fourth module is used to perform object recognition, motion detection and abnormal motion recognition in sequence based on the real-time video content of the key areas, so as to determine the abnormality level of each key area; The fifth module is used to conduct risk assessment of the ship unloading system according to the degree of abnormality in each key area and formulate safety strategies based on the results of the risk assessment.

Images