Safety early warning method and system based on port monitoring data analysis

Abstract

The invention discloses a safety early warning method and system based on port monitoring data analysis, and relates to the technical field of port safety monitoring, human bodies and mobile working equipment in an onshore working area are identified from a port monitoring video, and the human bodies and the mobile working equipment are mapped to a planar graph of the onshore working area; synchronously acquiring ship access and cargo loading and unloading data of the underwater working area of the port, and determining the tonnage and cargo loading and unloading volume of each berthing ship to form a ship cargo parameter group; integrating and classifying the land working area planar graphs corresponding to the ship cargo parameter groups of which the closeness degrees are greater than or equal to a preset value, and analyzing expression characteristics of human body and mobile equipment mapping points to obtain characteristic parameters; the feature parameters are mapped back to a land working area plane graph to form a plane template for analyzing real-time monitoring data of the port and performing safety early warning based on an analysis result, the technical scheme of the invention realizes safety early warning of the port, and improves the safety of port operation.

Description

Technical Field

[0001] The present invention relates to the technical field of port safety monitoring, and in particular to a safety early warning method and system based on port monitoring data analysis. Background Art

[0002] With the rapid development of global trade, ports, as important hubs of international trade, are increasingly concerned about their safe, efficient and environmentally friendly operation. However, traditional port monitoring methods can no longer meet the needs of modern port operations. Problems such as the dispersion and isolation of multi-source monitoring data, low data utilization, delayed safety warnings, and single monitoring methods are becoming increasingly prominent. Therefore, a safety warning method and system based on port monitoring data is urgently needed. Summary of the invention

[0003] The purpose of the present invention is to provide a method and system capable of providing timely safety warning based on port monitoring data analysis.

[0004] The present invention discloses a safety early warning method based on port monitoring data analysis, comprising: Obtain port surveillance video, and use human body recognition technology and landmark object recognition technology to respectively identify human bodies and mobile working equipment in the port's land work area. A land work area plan is set for the port's road work area, and the identified human bodies are mapped on the land work area plan in the form of human body mapping points, and the identified mobile working equipment is mapped on the land work area plan in the form of mobile equipment mapping points; The ship entry and exit data of the port water work area and the cargo loading and unloading data of each ship are simultaneously obtained, the tonnage parameter of each berthing ship in the ship entry and exit data is determined, and the cargo loading and unloading volume parameter of the cargo loading and unloading data is determined, and the combination of the tonnage parameter and the cargo loading and unloading volume parameter of each berthing ship is recorded as a ship cargo parameter group; Integrate and classify the land work area plan views corresponding to the ship cargo parameter groups whose proximity is greater than or equal to a preset value, and perform human body mapping point performance feature analysis and mobile equipment mapping point performance feature analysis on the integrated and classified land work area plan views to determine human body mapping point performance feature parameters and mobile equipment mapping point performance feature parameters; Mapping the human body mapping point performance characteristic parameters and the mobile equipment mapping point performance characteristic parameters corresponding to each category of ship cargo parameter group onto the onshore work area plane map to obtain the onshore work area plane template; The onshore workspace template is used to analyze the real-time monitoring data of the port, and safety warnings are issued based on the analysis results.

[0005] In some embodiments disclosed in the present invention, a method for respectively identifying a human body and a mobile working equipment in a port land work area by using a human body recognition technology and a marker recognition technology includes: Detecting human bodies using human body recognition technology, including edge detection, shape analysis, and motion detection of surveillance video images, determining image areas in surveillance video images that match human body characteristics, and marking image areas that match human body characteristics; The mobile working equipment is detected by using landmark recognition technology, including edge detection, shape analysis and motion detection on the monitoring video image, determining the image area in the monitoring video image that meets the characteristics of the working equipment, and marking the image area that meets the characteristics of the working equipment.

[0006] In some embodiments disclosed in the present invention, a land work area plan is set for the port land work area, and a method for mapping the identified human body and mobile working equipment to the land work area plan in the form of mapping points includes: Based on the relevant design drawings of the port, construct a plan of the onshore work area, divide the plan of the onshore work area into several core work areas, and set a plan coordinate system for the plan of the onshore work area; Configure an image coordinate system for each surveillance video image, determine the correspondence between the area captured by the surveillance video image and the plan view of the onshore work area, and match the image coordinate system with the plan view coordinate system; Determine a human image area and a working equipment image area on the monitoring video image, and determine a first area center point of the human image area and a second area center point of the working equipment image area; The position coordinates of the center point of the first area relative to the monitoring video image are converted into position coordinates in the land work area plan, and mapped on the land work area plan in the form of human body mapping points. The position coordinates of the center point of the second area relative to the monitoring video image are converted into position coordinates in the land work area plan, and mapped on the land work area plan in the form of mobile equipment mapping points.

[0007] In some embodiments disclosed herein, a method for determining the proximity between ship cargo parameter groups includes: Determine the tonnage parameters of the ship at each berth in the port, sort the tonnage parameters according to the order between the berths, and configure the cargo loading and unloading volume parameters of each ship after the corresponding tonnage parameters to obtain the ship cargo parameter group; Based on the time sequence, the ship cargo parameter groups are time-sorted to obtain a ship cargo parameter group sequence; Corresponding to the change of tonnage parameters of each ship position in the ship cargo parameter group sequence, a tonnage change curve is constructed. According to the change of cargo loading and unloading volume parameters corresponding to each ship position, a cargo loading and unloading volume change curve is constructed. The tonnage change curve is analyzed to determine the first mutation section in the tonnage change curve where the tonnage parameter change rate is greater than or equal to a preset value, and the midpoint of the first mutation section is marked as the first mutation reference node. The cargo loading and unloading volume change curve is analyzed to determine the second mutation section in the cargo loading volume change curve where the loading and unloading volume parameter change rate is greater than or equal to the preset value, and the midpoint of the second mutation section is marked as the second mutation reference node; A first inter-node distance sequence is constructed for the distance between the first mutation reference nodes, and a second inter-node distance sequence is constructed for the distance between the second mutation reference nodes. By comparing the symmetry between the first inter-node distance sequences and the symmetry between the second inter-node distance sequences, the corresponding tonnage change curves and cargo loading and unloading volume change curves are aligned, and the curve matching parameters between each corresponding curve are determined. Based on the curve matching parameters of each corresponding curve, the degree of proximity between the ship cargo parameter groups is determined.

[0008] In some embodiments disclosed in the present invention, a method for determining curve matching parameters between corresponding curves includes: Determine a mutation section and a non-mutation section on the curve, and set a first attention weight coefficient for the mutation section and a second attention weight coefficient for the non-mutation section; For the longitudinal difference values ​​of different position nodes of the curve, if the longitudinal difference value is less than or equal to the preset value, the corresponding position node is identified as the matching position node, and the sub-curve matching parameters of the segment are determined based on the ratio of the number of matching position nodes of each segment to the number of all position nodes; Based on the attention weight coefficient of each segment on the curve and the sub-matching parameter, the curve matching parameters between the curves are determined; The expression for calculating the curve fitting parameter is: ; Where W is the curve fitting parameter, is the sub-matching parameter standard judgment function, which is used to judge whether the sub-matching parameter meets the preset standard. If it meets the preset standard, then Output 1, otherwise output 0. is the number of nodes at the corresponding positions of the i-th segment in the curve, is the number of all position nodes corresponding to the i-th segment in the curve, n is the total number of mutation segments and non-mutation segments in the curve, is the attention weight coefficient of the ith segment, and C is the sub-fitting parameter fitting adjustment constant.

[0009] In some embodiments disclosed in the present invention, a method for determining the degree of proximity between ship cargo parameter groups based on the curve matching parameters of each corresponding curve includes: Several sub-matching parameter intervals are set for each curve, and a sub-closeness is set for each sub-matching parameter interval. The sub-closeness corresponding to each curve is determined by judging the sub-matching parameter interval to which the sub-matching parameter of each curve belongs, and the cumulative sum of the sub-closeness of all curves is calculated to obtain the closeness between the ship cargo parameter groups.

[0010] In some embodiments disclosed in the present invention, a method for performing human body mapping point performance feature analysis on an integrated and classified land work area plan and a method for performing mobile device mapping point performance feature analysis includes: Establishing a time reference line, analyzing the performance status of the land work area plan map corresponding to different time nodes on the time reference line, wherein the analysis method includes randomly selecting a number of random coordinates relative to the land work area plan map, and determining the number of human body mapping points of human body mapping points within a preset range near each random coordinate, and the number of mobile device mapping points of mobile device mapping points, and calculating the average number of human body mapping points and the average number of mobile device mapping points of all random coordinates; Determine a human body mapping point multiple of the number of human body mapping points of each random coordinate compared to the average number of human body mapping points, and if the human body mapping point multiple is greater than or equal to a preset value, identify the random coordinate as a human body mapping point dense coordinate; determine a mobile device mapping point multiple of the number of mobile device mapping points of each random coordinate compared to the average number of mobile device mapping points, and if the mobile device mapping point multiple is greater than or equal to a preset value, identify the random coordinate as a mobile device mapping point dense coordinate; The dense coordinates of the human body mapping points and the corresponding number of human body mapping points are identified as sub-human body mapping point performance characteristic parameters, and the sub-human body mapping point performance characteristic parameters are sorted according to the time reference line to obtain the human body mapping point performance characteristic parameters, and the combination of the dense coordinates of the mobile device mapping points and the corresponding number of mobile device mapping points is identified as the sub-mobile device mapping point performance characteristic parameters, and the sub-mobile device mapping point performance characteristic parameters are sorted according to the time reference line to obtain the mobile device mapping point performance characteristic parameters.

[0011] In some embodiments disclosed in the present invention, a method for mapping a human body mapping point performance characteristic parameter and a mobile device mapping point performance characteristic parameter onto a land work plane map includes: A land work area plan is constructed for each time node of the reference time line, and a number of marked human body mapping points are set on the land work area plan based on the dense coordinates of the human body mapping points corresponding to different time nodes and the corresponding number of human body mapping points, and the number of human body mapping points is recorded in the corresponding marked human body mapping points; Based on the dense coordinates of the mobile equipment mapping points corresponding to different time nodes and the corresponding number of mobile equipment mapping points, several marking mobile equipment mapping points are set on the land work area plan, and the number of mobile equipment mapping points is recorded in the corresponding marking mobile equipment mapping points to obtain the land work area plan template.

[0012] In some embodiments disclosed in the present invention, there is also disclosed a safety early warning system based on port monitoring data analysis, including: The first module is used to obtain port surveillance videos, and use human body recognition technology and landmark object recognition technology to respectively identify human bodies and mobile working equipment in the port's land work area. A land work area plan is set for the port's road work area, and the identified human bodies are mapped on the land work area plan in the form of human body mapping points, and the identified mobile working equipment is mapped on the land work area plan in the form of mobile equipment mapping points; The second module is used to synchronously obtain the ship entry and exit data of the port water work area and the cargo loading and unloading data of each ship, determine the tonnage parameters of each docked ship in the ship entry and exit data, and determine the cargo loading and unloading volume parameters of the cargo loading and unloading data, and record the combination of the tonnage parameters and cargo loading and unloading volume parameters of each docked ship as a ship cargo parameter group; The third module is used to integrate and classify the land work area plan views corresponding to the ship cargo parameter groups whose proximity is greater than or equal to the preset value, and perform human body mapping point performance feature analysis and mobile device mapping point performance feature analysis on the integrated and classified land work area plan views to determine the human body mapping point performance feature parameters and the mobile device mapping point performance feature parameters; The fourth module is used to map the human body mapping point performance characteristic parameters and the mobile equipment mapping point performance characteristic parameters corresponding to each category of ship cargo parameter group to the land work area plane map to obtain the land work area plane template; The fifth module is used to analyze the real-time monitoring data of the port using the onshore workspace template and issue safety warnings based on the analysis results.

[0013] The present invention discloses a safety warning method and system based on port monitoring data analysis, which relates to the technical field of port safety monitoring. Human bodies and mobile working equipment in an onshore work area are identified from port monitoring videos and mapped onto a plan view of the onshore work area. Ship entry and exit and cargo loading and unloading data in an underwater work area of ​​the port are simultaneously acquired to determine the tonnage and cargo loading and unloading volume of each docked ship, forming a ship cargo parameter group. The onshore work area plan views corresponding to the ship cargo parameter groups whose proximity is greater than or equal to a preset value are integrated and classified, and the performance characteristics of the mapping points of human bodies and mobile equipment are analyzed to obtain characteristic parameters. These characteristic parameters are mapped back to the onshore work area plan view to form a plane template for analyzing the real-time monitoring data of the port, and performing safety warnings based on the analysis results. The above technical solution of the present invention realizes safety warnings for ports and improves the safety of port operations.

[0014] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 It is a method step diagram of the safety early warning method based on port monitoring data analysis disclosed in an embodiment of the present invention. DETAILED DESCRIPTION

[0016] The technical solution of the present invention is further described below through the accompanying drawings and embodiments.

[0017] The following will be combined with the accompanying drawings and specific embodiments to clearly and completely describe the technical solution of the present invention. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention and cannot be understood as limiting the scope of protection of the present invention. Those skilled in the art in this field can make some non-essential improvements and adjustments based on the content of the present invention described below. In the present invention, unless otherwise clearly specified and limited, the technical terms used in the present invention should be the common meanings understood by the technical personnel described in the present invention.

[0018] Example:

[0019] The purpose of the present invention is to provide a method and system capable of providing timely safety warning based on port monitoring data analysis.

[0020] The present invention discloses a safety early warning method based on port monitoring data analysis, see Figure 1 ,include: Step S100, obtain port surveillance video, and use human body recognition technology and landmark object recognition technology to respectively identify human bodies and mobile work equipment in the port's land work area. A land work area plan is set for the port's road work area. The identified human bodies are mapped on the land work area plan in the form of human body mapping points, and the identified mobile work equipment is mapped on the land work area plan in the form of mobile equipment mapping points.

[0021] This step mainly involves the acquisition and processing of port surveillance videos. First, the video images of the port are captured in real time through surveillance cameras installed in various locations in the port. Then, advanced human body recognition technology is used to accurately identify the human body in the video. This technology is usually based on deep learning algorithms, which can analyze the pixel information in the video frame, identify the shape, movement and other characteristics of the human body, and accurately mark the position of the human body in the video. At the same time, the mobile work equipment (such as cranes, forklifts, etc.) in the port's land work area is identified using landmark object recognition technology. These devices usually have unique shapes and logos, allowing the recognition algorithm to easily distinguish them from other objects. The identified human body and equipment information is then mapped to a pre-set land work area plan, displayed in the form of human body mapping points and mobile equipment mapping points, providing basic data support for subsequent safety warnings.

[0022] In some embodiments disclosed in the present invention, a method for respectively identifying a human body and a mobile working equipment in a port land work area by using a human body recognition technology and a marker recognition technology includes: Step S101, using human body recognition technology to detect the human body, including edge detection, shape analysis and motion detection of the monitoring video image, determining the image area in the monitoring video image that meets the human body characteristics, and marking the image area that meets the human body characteristics; Step S102, using landmark recognition technology to detect the mobile working equipment, including edge detection, shape analysis and motion detection on the monitoring video image, determining the image area in the monitoring video image that meets the characteristics of the working equipment, and marking the image area that meets the characteristics of the working equipment.

[0023] In some embodiments disclosed in the present invention, a land work area plan is set for the port land work area, and a method for mapping the identified human body and mobile working equipment to the land work area plan in the form of mapping points includes: Step S103, based on the relevant design drawings of the port, construct a plan view of the onshore work area, divide the onshore work area plan view into several core work areas, and set a plan view coordinate system for the onshore work area plan view.

[0024] This step is the basic process of constructing the onshore work area plan. First, based on the relevant design drawings of the port, these drawings usually contain detailed information such as the port's layout, facility location, and dimensions, which are an accurate basis for constructing the plan. By carefully analyzing these drawings, a plan of the onshore work area can be constructed. On the plan, in order to further refine the work area, several core work areas will be divided. These areas may be divided according to function, equipment type, or workflow. At the same time, in order to accurately locate and analyze later, a plan coordinate system will be set for the onshore work area plan. Step S104, configuring an image coordinate system for each surveillance video image, determining the correspondence between the area captured by the surveillance video image and the plan view of the onshore work area, and performing corresponding matching between the image coordinate system and the plan view coordinate system.

[0025] This step is the key process to establish the connection between the surveillance video image and the onshore work area plan. First, an image coordinate system is configured for each surveillance video image. This coordinate system is set based on the pixel position of the video image and is used to describe the position of each pixel in the image. Then, it is necessary to determine the correspondence between the area captured by the surveillance video image and the onshore work area plan. This usually requires on-site investigation, measurement, or alignment using known landmark objects. Finally, the image coordinate system and the plan coordinate system are matched. This process may involve operations such as coordinate transformation, scaling, and rotation to ensure that the points in the image accurately correspond to the points on the plan.

[0026] Step S105, determining a human image area and a working equipment image area on the monitoring video image, and determining a first area center point of the human image area and a second area center point of the working equipment image area.

[0027] This step is the process of identifying and locating the human body and working equipment in the surveillance video image. First, the human body image area and the working equipment image area on the surveillance video image are determined through image processing technology (such as target detection, image segmentation, etc.). These areas are sets of pixels in the image that contain the human body or working equipment. Then, in order to simplify subsequent processing, the first area center point of the human body image area and the second area center point of the working equipment image area are determined. These center points can be the centroid, geometric center or representative points determined according to other rules of the area, which are used to represent the position of the entire area.

[0028] Step S106, converting the position coordinates of the center point of the first area relative to the monitoring video image into position coordinates in the land work area plan, and mapping them on the land work area plan in the form of human body mapping points, converting the position coordinates of the center point of the second area relative to the monitoring video image into position coordinates in the land work area plan, and mapping them on the land work area plan in the form of mobile device mapping points.

[0029] This step is the process of mapping the points in the monitoring video image and the points on the land work area plan. First, the position coordinates of the center point of the first area (representing the human body) relative to the monitoring video image are converted into position coordinates in the land work area plan. This conversion process is based on the coordinate system correspondence established in step S104, and may involve coordinate transformation and calculation. Then, mapping is performed on the land work area plan in the form of human body mapping points, that is, the position of the human body is marked on the plan. Similarly, for the center point of the second area of ​​the work equipment image area, the same coordinate conversion and mapping process is performed, and the position of the equipment is marked on the land work area plan in the form of mobile equipment mapping points. In this way, the human body and the mobile working equipment in the monitoring video image are mapped to the land work area plan in the form of mapping points.

[0030] Step S200, synchronously obtain the ship entry and exit data of the port's water work area and the cargo loading and unloading data of each ship, determine the tonnage parameters of each docked ship in the ship entry and exit data, and determine the cargo loading and unloading volume parameters of the cargo loading and unloading data, and record the combination of the tonnage parameters of each docked ship and the cargo loading and unloading volume parameters as a ship cargo parameter group.

[0031] This step focuses on the acquisition and processing of ship entry and exit and cargo loading and unloading data in the port's water work area. Through the port's information management system, basic information such as the ship's entry and exit time and tonnage can be obtained in real time. At the same time, through the cargo loading and unloading record system, the cargo loading and unloading situation of each ship can be recorded in detail, including the type, quantity, loading and unloading time of the loaded and unloaded cargo. These data are used to calculate the tonnage parameters and cargo loading and unloading volume parameters of each docked ship to form a ship cargo parameter group. These parameter groups not only reflect the transportation capacity of the ship, but also reflect the port's busyness and cargo handling capacity, providing important data basis for subsequent safety warnings.

[0032] In some embodiments disclosed herein, a method for determining the proximity between ship cargo parameter groups includes: Step S201, determining the tonnage parameters of the ship at each berth of the port, sorting the tonnage parameters according to the sequence between the berths, and configuring the cargo loading and unloading volume parameters of each ship after the corresponding tonnage parameters to obtain a ship cargo parameter group.

[0033] This step aims to collate and organize the tonnage and cargo handling volume data of ships at port berths. First, for each ship at a port berth, determine its tonnage parameter, which is usually one of the basic attributes of a ship and reflects the ship's carrying capacity. Then, sort these tonnage parameters according to the sequence between berths (such as the order from port entrance to exit, or the order according to certain management regulations). Next, configure the cargo handling volume parameters of each ship after the corresponding tonnage parameters to form a ship cargo parameter group containing tonnage and loading and unloading volume. In this way, each ship at a berth has a corresponding parameter group, which is convenient for subsequent analysis and processing.

[0034] Step S202, based on the time sequence, the ship cargo parameter groups are time-sorted to obtain a ship cargo parameter group sequence.

[0035] Step S203, corresponding to the change of the tonnage parameter of each ship position in the ship cargo parameter group sequence, construct a tonnage change curve, construct a cargo loading and unloading volume change curve for the change of the cargo loading and unloading volume parameter corresponding to each ship position, analyze the tonnage change curve, determine the first mutation section in the tonnage change curve where the tonnage parameter change rate is greater than or equal to a preset value, mark the midpoint of the first mutation section as the first mutation reference node, analyze the cargo loading and unloading volume change curve, determine the second mutation section in the cargo loading and unloading volume change curve where the loading and unloading volume parameter change rate is greater than or equal to the preset value, mark the midpoint of the second mutation section as the second mutation reference node.

[0036] Step S204, constructing a first inter-node distance sequence for the distance between the first mutation reference nodes, constructing a second inter-node distance sequence for the distance between the second mutation reference nodes, and aligning the corresponding tonnage change curves and cargo loading and unloading volume change curves by comparing the symmetry between the first inter-node distance sequences and the symmetry between the second inter-node distance sequences, and determining the curve matching parameters between each corresponding curve, and determining the degree of proximity between the ship cargo parameter groups based on the curve matching parameters of each corresponding curve.

[0037] In some embodiments disclosed in the present invention, a method for determining curve matching parameters between corresponding curves includes: Step S2041 , determining a mutation segment and a non-mutation segment on the curve, and setting a first attention weight coefficient for the mutation segment and a second attention weight coefficient for the non-mutation segment.

[0038] Step S2042, for the longitudinal difference values ​​of different position nodes of the curve, if the longitudinal difference value is less than or equal to the preset value, the corresponding position node is identified as the matching position node, and the sub-curve matching parameters of the segment are determined based on the ratio of the number of matching position nodes in each segment to the number of all position nodes.

[0039] Step S2043, based on the attention weight coefficient of each segment on the curve and the sub-matching parameter, determine the curve matching parameters between the curves.

[0040] The expression for calculating the curve fitting parameter is: .

[0041] Where W is the curve fitting parameter, is the sub-matching parameter standard judgment function, which is used to judge whether the sub-matching parameter meets the preset standard. If it meets the preset standard, then Output 1, otherwise output 0. is the number of nodes at the corresponding positions of the i-th segment in the curve, is the number of all position nodes corresponding to the i-th segment in the curve, n is the total number of mutation segments and non-mutation segments in the curve, is the attention weight coefficient of the ith segment, and C is the sub-fitting parameter fitting adjustment constant.

[0042] In some embodiments disclosed in the present invention, a method for determining the degree of proximity between ship cargo parameter groups based on the curve matching parameters of each corresponding curve includes: Step S2044, a number of sub-matching parameter intervals are set for each curve, and a sub-closeness is set for each sub-matching parameter interval. The sub-matching parameter interval to which the sub-matching parameter of each curve belongs is judged, the sub-closeness corresponding to each curve is determined, and the cumulative sum of the sub-closeness of all curves is calculated to obtain the closeness between the ship cargo parameter groups.

[0043] Step S300, integrating and classifying the land work area plan views corresponding to the ship cargo parameter groups whose proximity is greater than or equal to a preset value, and performing human body mapping point performance feature analysis and mobile device mapping point performance feature analysis on the integrated and classified land work area plan views to determine the human body mapping point performance feature parameters and the mobile device mapping point performance feature parameters.

[0044] The core of this step is to cluster the ship cargo parameter groups and extract features from the clustered land work area plan. First, according to the similarity of the ship cargo parameter groups (such as tonnage, loading and unloading volume, etc.), the parameter groups are clustered, and the ship cargo parameter groups whose proximity is greater than or equal to the preset value are classified into one category. Then, the performance characteristics of the human body mapping points and mobile equipment mapping points are analyzed for the land work area plan corresponding to each category. This analysis may include the density, distribution pattern, movement trajectory, etc. of the mapping points, aiming to extract characteristic parameters that can reflect the working status of the port. These characteristic parameters will provide input for the subsequent safety warning model.

[0045] In some embodiments disclosed in the present invention, a method for performing human body mapping point performance feature analysis on an integrated and classified land work area plan and a method for performing mobile device mapping point performance feature analysis includes: Step S301, establish a time reference line, and analyze the performance status of the land work area plan corresponding to different time nodes on the time reference line. The analysis method includes randomly selecting a number of random coordinates relative to the land work area plan, and determining the number of human body mapping points of human body mapping points within a preset range near each random coordinate, and the number of mobile device mapping points of mobile device mapping points, and calculating the average number of human body mapping points and the average number of mobile device mapping points of all random coordinates.

[0046] Step S302, determining the human body mapping point multiple of the number of human body mapping points of each random coordinate compared to the average number of human body mapping points, if the human body mapping point multiple is greater than or equal to a preset value, the random coordinate is identified as a human body mapping point dense coordinate, determining the mobile device mapping point multiple of the number of mobile device mapping points of each random coordinate compared to the average number of mobile device mapping points, if the mobile device mapping point multiple is greater than or equal to the preset value, the random coordinate is identified as a mobile device mapping point dense coordinate.

[0047] Step S303, identifying the dense coordinates of the human body mapping points and the corresponding number of human body mapping points as sub-human body mapping point performance characteristic parameters, and sorting the sub-human body mapping point performance characteristic parameters according to the time reference line to obtain the human body mapping point performance characteristic parameters, identifying the combination of the dense coordinates of the mobile device mapping points and the corresponding number of mobile device mapping points as sub-mobile device mapping point performance characteristic parameters, and sorting the sub-mobile device mapping point performance characteristic parameters according to the time reference line to obtain the mobile device mapping point performance characteristic parameters.

[0048] Step S400, mapping the human body mapping point performance characteristic parameters and the mobile equipment mapping point performance characteristic parameters corresponding to each category of the ship cargo parameter group to the land work area plane map, to obtain the land work area plane template.

[0049] This step is the process of mapping the feature parameters extracted in the previous step back to the land work area plan to form a plane template. Specifically, for each category of ship cargo parameter group, the corresponding human body mapping point performance feature parameters and mobile equipment mapping point performance feature parameters are graphically displayed on the land work area plan. These feature parameters may be represented by marker points of different colors, shapes or sizes to form an intuitive plane template. These templates not only reflect the characteristics of the port under different working conditions, but also provide a visual reference for subsequent safety warnings.

[0050] In some embodiments disclosed in the present invention, a method for mapping a human body mapping point performance characteristic parameter and a mobile device mapping point performance characteristic parameter onto a land work plane map includes: Step S401, constructing a land work area plan for each time node of the reference timeline, setting a number of marked human body mapping points on the land work area plan based on the dense coordinates of the human body mapping points corresponding to different time nodes and the corresponding number of human body mapping points, and recording the number of human body mapping points at the corresponding marked human body mapping points.

[0051] Step S402, based on the dense coordinates of the mobile equipment mapping points corresponding to different time nodes and the corresponding number of mobile equipment mapping points, set a number of marker mobile equipment mapping points on the land work area plan, and record the number of mobile equipment mapping points in the corresponding marker mobile equipment mapping points to obtain the land work area plan template.

[0052] Step S500: Analyze the real-time monitoring data of the port using the onshore workspace template, and issue a safety warning based on the analysis results.

[0053] This step is to use the onshore work area plane template generated in the previous step to analyze the real-time monitoring data of the port and issue a safety warning. Specifically, by comparing the real-time monitoring data with the plane template, abnormalities or potential risks in the port's working status can be discovered in a timely manner. For example, when the distribution of human body mapping points or mobile equipment mapping points shown in the real-time monitoring data is significantly different from the plane template, it may mean that there are safety hazards or reduced work efficiency in the port. At this time, the system will trigger the safety warning mechanism to remind relevant personnel to take timely measures to intervene or adjust to ensure the normal operation and safety of the port.

[0054] In some embodiments disclosed in the present invention, there is also disclosed a safety early warning system based on port monitoring data analysis, including: The first module is used to obtain port surveillance videos, and use human body recognition technology and landmark object recognition technology to respectively identify human bodies and mobile working equipment in the port's land work area. A land work area plan is set for the port's road work area, and the identified human bodies are mapped on the land work area plan in the form of human body mapping points, and the identified mobile working equipment is mapped on the land work area plan in the form of mobile equipment mapping points; The second module is used to synchronously obtain the ship entry and exit data of the port water work area and the cargo loading and unloading data of each ship, determine the tonnage parameters of each docked ship in the ship entry and exit data, and determine the cargo loading and unloading volume parameters of the cargo loading and unloading data, and record the combination of the tonnage parameters and cargo loading and unloading volume parameters of each docked ship as a ship cargo parameter group; The third module is used to integrate and classify the land work area plan views corresponding to the ship cargo parameter groups whose proximity is greater than or equal to the preset value, and perform human body mapping point performance feature analysis and mobile device mapping point performance feature analysis on the integrated and classified land work area plan views to determine the human body mapping point performance feature parameters and the mobile device mapping point performance feature parameters; The fourth module is used to map the human body mapping point performance characteristic parameters and the mobile equipment mapping point performance characteristic parameters corresponding to each category of ship cargo parameter group to the land work area plane map to obtain the land work area plane template; The fifth module is used to analyze the real-time monitoring data of the port using the onshore workspace template and issue safety warnings based on the analysis results.

[0055] The present invention discloses a safety warning method and system based on port monitoring data analysis, which relates to the technical field of port safety monitoring. Human bodies and mobile working equipment in an onshore work area are identified from port monitoring videos and mapped onto a plan view of the onshore work area. Ship entry and exit and cargo loading and unloading data in an underwater work area of ​​the port are simultaneously acquired to determine the tonnage and cargo loading and unloading volume of each docked ship, forming a ship cargo parameter group. The onshore work area plan views corresponding to the ship cargo parameter groups whose proximity is greater than or equal to a preset value are integrated and classified, and the performance characteristics of the mapping points of human bodies and mobile equipment are analyzed to obtain characteristic parameters. These characteristic parameters are mapped back to the onshore work area plan view to form a plane template for analyzing the real-time monitoring data of the port, and performing safety warnings based on the analysis results. The above technical solution of the present invention realizes safety warnings for ports and improves the safety of port operations.

[0056] Through the description of the above implementation methods, those skilled in the art can clearly understand that the present invention can be implemented by hardware, or by software plus a necessary general hardware platform. Based on such an 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 (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.), including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each implementation scenario of the present invention.

[0057] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that they can still modify or replace the technical solution of the present invention with equivalents, and these modifications or equivalent replacements cannot cause the modified technical solution to deviate from the spirit and scope of the technical solution of the present invention.

Claims

1. A safety early warning method based on port monitoring data analysis, characterized in that: include: Obtain port surveillance video, and use human body recognition technology and landmark object recognition technology to respectively identify human bodies and mobile working equipment in the port's land work area. A land work area plan is set for the port's road work area, and the identified human bodies are mapped on the land work area plan in the form of human body mapping points, and the identified mobile working equipment is mapped on the land work area plan in the form of mobile equipment mapping points; The ship entry and exit data of the port water work area and the cargo loading and unloading data of each ship are simultaneously obtained, the tonnage parameter of each berthing ship in the ship entry and exit data is determined, and the cargo loading and unloading volume parameter of the cargo loading and unloading data is determined, and the combination of the tonnage parameter and the cargo loading and unloading volume parameter of each berthing ship is recorded as a ship cargo parameter group; Integrate and classify the land work area plan views corresponding to the ship cargo parameter groups whose proximity is greater than or equal to a preset value, and perform human body mapping point performance feature analysis and mobile equipment mapping point performance feature analysis on the integrated and classified land work area plan views to determine human body mapping point performance feature parameters and mobile equipment mapping point performance feature parameters; Mapping the human body mapping point performance characteristic parameters and the mobile equipment mapping point performance characteristic parameters corresponding to each category of ship cargo parameter group onto the onshore work area plane map to obtain the onshore work area plane template; The onshore workspace template is used to analyze the real-time monitoring data of the port, and safety warnings are issued based on the analysis results.

2. The safety early warning method based on port monitoring data analysis according to claim 1 is characterized in that: The method of using human body recognition technology and marker recognition technology to respectively identify human bodies and mobile working equipment in the port land work area includes: Detecting human bodies using human body recognition technology, including edge detection, shape analysis, and motion detection of surveillance video images, determining image areas in surveillance video images that match human body characteristics, and marking image areas that match human body characteristics; The mobile working equipment is detected by using landmark recognition technology, including edge detection, shape analysis and motion detection on the monitoring video image, determining the image area in the monitoring video image that meets the characteristics of the working equipment, and marking the image area that meets the characteristics of the working equipment.

3. The safety early warning method based on port monitoring data analysis according to claim 2 is characterized in that: A land work area plan is set for the land work area of ​​the port. The method of mapping the identified human body and mobile work equipment to the land work area plan in the form of mapping points includes: Based on the relevant design drawings of the port, construct a plan of the onshore work area, divide the plan of the onshore work area into several core work areas, and set a plan coordinate system for the plan of the onshore work area; Configure an image coordinate system for each surveillance video image, determine the correspondence between the area captured by the surveillance video image and the plan view of the onshore work area, and match the image coordinate system with the plan view coordinate system; Determine a human image area and a working equipment image area on the monitoring video image, and determine a first area center point of the human image area and a second area center point of the working equipment image area; The position coordinates of the center point of the first area relative to the monitoring video image are converted into position coordinates in the land work area plan, and mapped on the land work area plan in the form of human body mapping points. The position coordinates of the center point of the second area relative to the monitoring video image are converted into position coordinates in the land work area plan, and mapped on the land work area plan in the form of mobile equipment mapping points.

4. The safety early warning method based on port monitoring data analysis according to claim 1 is characterized in that: Methods for determining the proximity between ship cargo parameter groups include: Determine the tonnage parameters of the ship at each berth in the port, sort the tonnage parameters according to the order between the berths, and configure the cargo loading and unloading volume parameters of each ship after the corresponding tonnage parameters to obtain the ship cargo parameter group; Based on the time sequence, the ship cargo parameter groups are time-sorted to obtain a ship cargo parameter group sequence; Corresponding to the change of tonnage parameters of each ship position in the ship cargo parameter group sequence, a tonnage change curve is constructed. According to the change of cargo loading and unloading volume parameters corresponding to each ship position, a cargo loading and unloading volume change curve is constructed. The tonnage change curve is analyzed to determine the first mutation section in the tonnage change curve where the tonnage parameter change rate is greater than or equal to a preset value, and the midpoint of the first mutation section is marked as the first mutation reference node. The cargo loading and unloading volume change curve is analyzed to determine the second mutation section in the cargo loading volume change curve where the loading and unloading volume parameter change rate is greater than or equal to the preset value, and the midpoint of the second mutation section is marked as the second mutation reference node; A first inter-node distance sequence is constructed for the distance between the first mutation reference nodes, and a second inter-node distance sequence is constructed for the distance between the second mutation reference nodes. By comparing the symmetry between the first inter-node distance sequences and the symmetry between the second inter-node distance sequences, the corresponding tonnage change curves and cargo loading and unloading volume change curves are aligned, and the curve matching parameters between each corresponding curve are determined. Based on the curve matching parameters of each corresponding curve, the degree of proximity between the ship cargo parameter groups is determined.

5. The safety early warning method based on port monitoring data analysis according to claim 4 is characterized in that: Methods for determining curve fitting parameters between corresponding curves include: Determine a mutation section and a non-mutation section on the curve, and set a first attention weight coefficient for the mutation section and a second attention weight coefficient for the non-mutation section; For the longitudinal difference values ​​of different position nodes of the curve, if the longitudinal difference value is less than or equal to the preset value, the corresponding position node is identified as the matching position node, and the sub-curve matching parameters of the segment are determined based on the ratio of the number of matching position nodes of each segment to the number of all position nodes; Based on the attention weight coefficient of each segment on the curve and the sub-matching parameter, the curve matching parameters between the curves are determined; The expression for calculating the curve fitting parameter is: ； Where W is the curve fitting parameter, is the sub-matching parameter standard judgment function, which is used to judge whether the sub-matching parameter meets the preset standard. If it meets the preset standard, then Output 1, otherwise output 0. is the number of nodes at the corresponding positions of the i-th segment in the curve, is the number of all position nodes corresponding to the i-th segment in the curve, n is the total number of mutation segments and non-mutation segments in the curve, is the attention weight coefficient of the ith segment, and C is the sub-fitting parameter fitting adjustment constant.

6. The safety early warning method based on port monitoring data analysis according to claim 5 is characterized in that: Based on the curve matching parameters of each corresponding curve, the method for determining the closeness between the ship cargo parameter groups includes: Several sub-matching parameter intervals are set for each curve, and a sub-closeness is set for each sub-matching parameter interval. The sub-closeness corresponding to each curve is determined by judging the sub-matching parameter interval to which the sub-matching parameter of each curve belongs, and the cumulative sum of the sub-closeness of all curves is calculated to obtain the closeness between the ship cargo parameter groups.

7. The safety early warning method based on port monitoring data analysis according to claim 1 is characterized in that: The method for analyzing the human body mapping point performance characteristics of the integrated and classified land work area plan and the mobile equipment mapping point performance characteristics analysis includes: Establishing a time reference line, analyzing the performance status of the land work area plan map corresponding to different time nodes on the time reference line, wherein the analysis method includes randomly selecting a number of random coordinates relative to the land work area plan map, and determining the number of human body mapping points of human body mapping points within a preset range near each random coordinate, and the number of mobile device mapping points of mobile device mapping points, and calculating the average number of human body mapping points and the average number of mobile device mapping points of all random coordinates; Determine a human body mapping point multiple of the number of human body mapping points of each random coordinate compared to the average number of human body mapping points, and if the human body mapping point multiple is greater than or equal to a preset value, identify the random coordinate as a human body mapping point dense coordinate; determine a mobile device mapping point multiple of the number of mobile device mapping points of each random coordinate compared to the average number of mobile device mapping points, and if the mobile device mapping point multiple is greater than or equal to a preset value, identify the random coordinate as a mobile device mapping point dense coordinate; The dense coordinates of the human body mapping points and the corresponding number of human body mapping points are identified as sub-human body mapping point performance characteristic parameters, and the sub-human body mapping point performance characteristic parameters are sorted according to the time reference line to obtain the human body mapping point performance characteristic parameters, and the combination of the dense coordinates of the mobile device mapping points and the corresponding number of mobile device mapping points is identified as the sub-mobile device mapping point performance characteristic parameters, and the sub-mobile device mapping point performance characteristic parameters are sorted according to the time reference line to obtain the mobile device mapping point performance characteristic parameters.

8. The safety early warning method based on port monitoring data analysis according to claim 7 is characterized in that: The method for mapping the characteristic parameters of the human body mapping points and the characteristic parameters of the mobile equipment mapping points onto the land work plane map includes: A land work area plan is constructed for each time node of the reference time line, and a number of marked human body mapping points are set on the land work area plan based on the dense coordinates of the human body mapping points corresponding to different time nodes and the corresponding number of human body mapping points, and the number of human body mapping points is recorded in the corresponding marked human body mapping points; Based on the dense coordinates of the mobile equipment mapping points corresponding to different time nodes and the corresponding number of mobile equipment mapping points, several marking mobile equipment mapping points are set on the land work area plan, and the number of mobile equipment mapping points is recorded in the corresponding marking mobile equipment mapping points to obtain the land work area plan template.

9. The safety early warning system based on port monitoring data analysis is characterized by: include: The first module is used to obtain port surveillance videos, and use human body recognition technology and landmark object recognition technology to respectively identify human bodies and mobile working equipment in the port's land work area. A land work area plan is set for the port's road work area, and the identified human bodies are mapped on the land work area plan in the form of human body mapping points, and the identified mobile working equipment is mapped on the land work area plan in the form of mobile equipment mapping points; The second module is used to synchronously obtain the ship entry and exit data of the port water work area and the cargo loading and unloading data of each ship, determine the tonnage parameters of each docked ship in the ship entry and exit data, and determine the cargo loading and unloading volume parameters of the cargo loading and unloading data, and record the combination of the tonnage parameters and cargo loading and unloading volume parameters of each docked ship as a ship cargo parameter group; The third module is used to integrate and classify the land work area plan views corresponding to the ship cargo parameter groups whose proximity is greater than or equal to the preset value, and perform human body mapping point performance feature analysis and mobile device mapping point performance feature analysis on the integrated and classified land work area plan views to determine the human body mapping point performance feature parameters and the mobile device mapping point performance feature parameters; The fourth module is used to map the human body mapping point performance characteristic parameters and the mobile equipment mapping point performance characteristic parameters corresponding to each category of ship cargo parameter group to the land work area plane map to obtain the land work area plane template; The fifth module is used to analyze the real-time monitoring data of the port using the onshore workspace template and issue safety warnings based on the analysis results.

Images