Port functional area mining method and system based on geographical hash coding

Abstract

The invention provides a port function area mining method and system based on geographical hash coding. The method comprises the following steps: determining a ship berthing function area; performing berthing identification and classification on track points; different Geohash hierarchies are selected for the in-port area and the out-port area, and a clustering method based on connectivity is adopted to segment the parking area into a plurality of sub-areas; an outlier grid and an abnormal value are removed, and a functional area with actual operation significance is formed; determining a ship entry and exit main channel; identifying an area with significantly unbalanced entry and exit point distribution; extracting a track section of the ship from the port entry point to the first mooring point, and displaying the track section in a thermodynamic diagram form; and segmenting the region into a plurality of sub-regions by adopting a connectivity-based clustering method, removing an outlier grid, and finally dividing a main channel. The method has the advantages that accurate recognition and dynamic segmentation of multiple types of functional areas in a complex port environment are achieved, and the spatial index efficiency and the expandability of practical application are remarkably improved.

Description

Technical Field

[0001] The present application belongs to the field of ship behavior recognition, and specifically relates to a port functional area mining method and system based on geographic hash coding. Background Art

[0002] Maritime transport holds a central position in the global trade system, with the vast majority of international cargo transport relying on the efficient movement of ships between major ports worldwide. As global trade continues to expand, the number, types, and activities of ships are becoming increasingly diverse, significantly increasing the density and dynamic complexity of traffic in ports and their adjacent waters. This trend has not only exacerbated waterway congestion but also posed unprecedented challenges to port management and shipping safety. Traditional port monitoring and management methods, relying on manual inspections and static rules, are no longer able to meet the demands of modern ports for intelligent, real-time, and efficient operations. To this end, an increasing number of ports are accelerating their transformation towards digitalization and intelligentization, driving the development of smart ports and smart shipping.

[0003] Automatic identification and extraction of ship functional zones is crucial for port management and intelligent scheduling. Accurate delineation of port functional zones not only enhances port supervision and optimizes ship scheduling, but also effectively reduces collision risks and improves port operational efficiency and safety. AIS (Automatic Identification System) data provides rich dynamic and static information for mining ship functional zones, reflecting the distribution and spatial utilization of ships within ports and adjacent waters. Spatial aggregation and analysis of large amounts of AIS trajectory data reveal the actual distribution patterns of functional zones within the port area, such as berthing, anchoring, and navigation channels, for different types of ships. However, due to factors such as large data volumes, complex spatial distribution, and diverse ship operations, efficient and accurate automatic extraction of port functional zones from AIS data still faces technical challenges, such as spatial indexing efficiency, regional boundary determination, and multi-scale adaptability. Therefore, an automatic functional zone mining method that fully utilizes the spatial characteristics of AIS trajectory data and possesses efficient spatial aggregation and regional connectivity determination capabilities is urgently needed to meet the practical needs of smart ports and intelligent shipping.

[0004] Existing methods for mining ship functional areas mainly focus on two aspects: feature area extraction and channel extraction. Feature area extraction mainly targets areas where the ship's navigation status changes, such as berthing areas, turning areas, and high and low speed areas. Some methods identify these areas through statistical analysis, but they often rely on manual interpretation of heat maps and subjective analysis of ship motion characteristics. In order to improve the level of automation, subsequent studies have attempted to divide these feature point areas with geometric boundaries, such as using polygons to envelop berthing points or subdividing berthing areas for different types of ships. The polygonal representation is intuitive, but it has certain limitations: when the number of edges is small, it is difficult to accurately describe the shape of a complex area, and when the number of edges is too large, the computational complexity of trajectory point attribution determination will be significantly increased, affecting real-time performance.

[0005] Channel extraction methods can be broadly categorized into three types: clustering-based, grid-based, and statistical. Clustering methods typically utilize vectorization techniques to group trajectory data based on spatial similarities, thereby identifying major shipping routes. Grid methods divide the space into equal-sized grids, count the number of trajectory points within each grid, and filter core channel regions using cumulative importance functions. Statistical methods, based on AIS trajectory data, utilize kernel density estimation, regression models, and other techniques to extract channels with significant flow characteristics. While these methods have their own advantages and disadvantages in practical applications, they all suffer from common issues such as parameter sensitivity, limited spatial representation capabilities, and difficulty adapting to complex port environments. With the advancement of smart port development, achieving efficient, accurate, and automated extraction of ship functional areas without manual labeling has become a key challenge in current research and practical applications. Unsupervised methods are attracting increasing attention due to their low reliance on data annotation and adaptability to complex environments. However, existing unsupervised methods still lack the necessary parameters for adaptability, spatial indexing efficiency, and functional area connectivity, making them difficult to meet the high standards required for practical port management. Summary of the Invention

[0006] The purpose of this application is to overcome the defects of limited spatial expression ability, poor real-time performance and weak adaptability in the current automatic extraction process of functional areas of ports and adjacent waters.

[0007] To achieve the above objectives, this application proposes a port functional area mining method based on geo-hash coding, including: Determine the functional areas for ship berthing: identify and classify trajectory points; select different Geohash levels for the port and the outer port areas, and use a connectivity-based clustering method to segment the berthing area into multiple sub-areas; use Geohash grids to screen the berthing points, remove outliers and abnormal values, and form functional areas with practical operational significance; Determine the main channel for ships entering and leaving the port: Identify areas with significantly uneven distribution of entry and exit points; extract the trajectory segments of ships from the port entry point to the first berth point and display them in the form of a heat map; use a connectivity-based clustering method to split the area into multiple sub-regions, remove outlier grids, and finally divide the main channel.

[0008] As an improvement to the above method, the use of Geohash grids to screen the anchor points and remove outliers and abnormal values ​​includes: Calculate the mean number of trajectory points in each sub-region and remove sub-regions with fewer trajectory points than the mean. Remove sub-regions with fewer than a set threshold number of track points.

[0009] As an improvement to the above method, the method of segmenting the parking area using a connectivity-based clustering method includes: For each unvisited grid, a new sub-region is initialized, and all grids adjacent to the grid that have not been visited are recursively searched and added to the sub-region. During the traversal process, the visited grids are marked to avoid repeated searches. When all reachable neighbors are traversed, the current sub-region is added to the sub-region list. This process is repeated until all grids are assigned to a sub-region.

[0010] As an improvement to the above method, the method of identifying areas with significantly uneven distribution of entry and exit points includes: If the number of port entry points in the same grid is a multiple of the number of port exit points or there are only port entry points, the waterway is an entry waterway; if the number of port exit points in the same grid is a multiple of the number of port entry points or there are only port exit points, the waterway is an exit waterway.

[0011] As an improvement to the above method, the method for determining the port entry point and the port exit point includes: First, the trajectory points within the boundary grid are marked as target points. For each target point, the time difference between it and the previous and next trajectory points is calculated. If the time interval before the target point exceeds the set time threshold, or it is the first point and the subsequent interval is within the set time threshold, it is determined to be an entry point. If the time interval after the target point exceeds the set time threshold, or it is the last point and the previous interval is within the set time threshold, it is determined to be a departure point.

[0012] As an improvement to the above method, the method for dividing the main channel includes: The sub-region with the most connected grids is selected as the main channel.

[0013] The present application also provides a port functional area mining system based on geographic hash coding, which is implemented based on the above method and includes: A module for determining ship berthing functional areas is used to identify and classify trajectory points. Different Geohash levels are selected for the port and outer port areas, and a connectivity-based clustering method is used to segment the berthing area into multiple sub-areas. The berthing points are screened using Geohash grids, outliers and abnormal values ​​are removed, and functional areas with practical operational significance are formed. The module for determining the main channel for ships entering and leaving the port is used to identify areas with significantly uneven distribution of entry and exit points; the trajectory segments of ships from the entry point to the first berthing point are extracted and displayed in the form of a heat map; a connectivity-based clustering method is used to divide the area into multiple sub-areas, and outlier grids and outliers are removed to finally divide the main channel.

[0014] Compared with the prior art, the advantages of this application are: 1. This application innovatively introduces multi-level Geohash coding to spatially discretize the port and adjacent waters, enabling efficient spatial indexing and data aggregation at different scales. By adaptively selecting Geohash levels for areas inside and outside the port, combined with low-frequency mesh removal and outlier filtering, the spatial resolution and computational efficiency of functional area extraction are significantly improved.

[0015] 2. This application proposes a depth-first search (DFS) clustering algorithm based on Geohash grid connectivity, which can automatically identify spatially connected functional sub-areas with practical operational significance. This method effectively overcomes the inability of traditional polygon or clustering methods to express regional boundaries in complex port environments, and improves the accuracy and robustness of functional area division.

[0016] 3. This application constructs an end-to-end unsupervised extraction process from port entry and exit identification of raw AIS track points to spatial aggregation of functional zones. This framework, which requires no complex parameter tuning and automatically adapts to diverse port environments and ship operation modes, enables efficient, accurate, and automated extraction of functional zones, significantly enhancing the practicality and scalability of smart port management.

[0017] These three innovations together form an end-to-end, efficient, and highly generalizable framework for the automatic extraction of port functional areas. This framework not only enables accurate identification and dynamic segmentation of multiple functional areas in complex port environments, but also significantly improves spatial indexing efficiency and scalability for practical applications, providing a solid data foundation and technical support for intelligent port management and maritime safety. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 Shown is a schematic diagram of the trajectory point area boundary; Figure 2 The figure shows the extraction flow chart of the ship berthing functional area; Figure 3 Shown is the extraction flow chart of the ship entry and exit channels; Figure 4 Shown is the anchorage extraction result diagram; Figure 5 Shown is a diagram of the main channel for ships entering and leaving the port; Figure 6 Shown are the main entry and exit channels extracted by the proposed method. DETAILED DESCRIPTION

[0019] The technical solution of this application is described in detail below with reference to the accompanying drawings.

[0020] The present invention aims to solve the problems of limited spatial expression ability, poor real-time performance and weak adaptability in the current automatic extraction process of functional areas of ports and adjacent waters, and provides a method and system for mining ship functional areas based on geographic hash coding. With the continuous expansion of the scale of global maritime transportation, the types and activities of ships are becoming increasingly diverse, and the spatial distribution and dynamic changes of port functional areas are becoming more and more complex. The traditional functional area division method that relies on manual drawing or simple geometric envelopes is difficult to meet the needs of smart ports for efficient, accurate and automated management. Existing methods have obvious deficiencies in spatial indexing efficiency, regional boundary determination and multi-scale adaptability, and are difficult to cope with real-time functional area extraction and dynamic updates under large-scale AIS data.

[0021] This application proposes a method and system for mining ship functional areas based on geohash coding. This method achieves efficient spatial indexing through multi-level geohash coding. Combining annotated stop trajectories with connectivity-based functional area clustering, it can automatically and accurately extract the functional areas of ports and adjacent waters without the need for manual labeling. This method not only improves the automation and intelligence level of functional area extraction but also provides a solid data foundation and technical support for port management, intelligent scheduling, and maritime safety.

[0022] This application proposes a method and system for mining ship functional areas based on geohash coding, which is used to mine ship functional areas near ports, mainly including anchorage areas and port entry and exit channels. When mining anchorage areas, the marked stop trajectory points are used. For the extraction of port entry and exit channels, this application proposes a port area entry and exit trajectory point identification algorithm to support subsequent channel mining.

[0023] Before describing the detailed method steps, let's first explain the principles of geohashing. Geohashing (Geohash) is a spatial indexing method that encodes longitude and latitude into short alphanumeric strings, enabling efficient storage and querying of location data. It works by recursively dividing the Earth's surface into rectangular grids, assigning each grid a unique code. Longer codes indicate higher spatial resolution and finer regions. This hierarchical structure supports fast spatial queries, scalable aggregation, and efficient spatial joins. In functional area mining, Geohash grids are used to organize ship trajectory data, enabling efficient extraction of functional areas and real-time spatial matching of AIS tracks with coastline features.

[0024] The method for identifying arrival and departure events (as shown in Table 1) is to analyze the Figure 1 ) is used to identify the ship's entry and exit events. The regional boundary is the boundary of the area where the analyzed data is located. Among them, the method input includes the trajectory point sequence and the trajectory point index located at the regional boundary. In addition to the trajectory sequence, this method also inputs the index set of trajectory points in the boundary grid, so as to efficiently identify entry and exit events. In the study, 30 minutes was set as the time threshold to distinguish the significant time intervals that occur when crossing the boundary. According to the AIS report, it is extremely rare for the AIS signal to be interrupted for more than 30 minutes, so this threshold can be used as a reasonable standard for distinguishing true entry and exit events. The specific process is: first, mark the trajectory points in the boundary grid as target points (steps 1-3); for each target point, calculate its time difference with the previous and next trajectory points, and if it is the first and last points of the sequence, the corresponding value is set to None (steps 6-9). If the time interval before the target point exceeds the threshold (or it is the first point and the subsequent interval is within the threshold), it is determined to be an arrival event (steps 10-11); conversely, if the time interval after the target point exceeds the threshold (or it is the last point and the preceding interval is within the threshold), it is determined to be a departure event (steps 12-13).

[0025] Table 1 Identification methods for arrival and departure events

[0026] During functional area mining, anchorage points or high-frequency trajectory points are often distributed across multiple adjacent Geohash grids. To accurately identify spatial connectivity, these discrete grids need to be aggregated. As shown in Table 2, Algorithm 2 is designed to address this requirement. By extracting spatially connected Geohash subregions, it automatically segments functional areas and ensures spatial consistency, thereby improving the accuracy and practicality of functional area extraction. The algorithm takes as input a collection of geohash grids with different practical operational significance (such as Geohash codes corresponding to ship anchorage points and port entry and exit route grids. For ship anchorage points, the primary focus is to remove small, isolated areas and divide them into different stopover subregions; for port entry and exit route grids, the primary focus is to ensure connectivity within the main channel). The algorithm outputs a number of subregions, each consisting of a set of spatially connected Geohash grids. The algorithm uses a depth-first search (DFS) strategy to traverse the Geohash grid network. For each unvisited Geohash code, a new subregion is initialized. All adjacent (i.e., four-connected neighbors with shared edges) Geohash codes belonging to the subregion and not yet visited are recursively searched and added to the subregion (steps 11-19). During the traversal, visited codes are marked to avoid repeated searches (step 12). Once all reachable neighbors have been traversed, the current subregion is added to the subregion list. This process is repeated until all Geohash codes in the subregion have been assigned to a subregion.

[0027] Table 2 Aggregate discrete grid method

[0028] like Figure 2As shown in the figure, functional anchorage areas are mined based on annotated anchorage points. Ships' stopover behavior outside the harbor exhibits diverse and complex characteristics. Some ships make brief stops to avoid larger vessels, while fishing boats may anchor for operational purposes. To more accurately extract the actual anchorage and reduce interference from other activities, we focus only on large ships with a length greater than 50 meters. Specifically, we extract the level-7 Geohash codes corresponding to the anchorage trajectory points, count the number of anchorage points within each grid, and remove grids with the lowest 25% of anchorage points as low-frequency areas to reduce noise. Given the wider scope of anchorage areas outside the harbor, a level-6 Geohash grid is used to reduce computational overhead. To further improve the accuracy of functional area boundaries and facilitate management, a connectivity-based clustering method is used to aggregate the Geohash grid into spatially connected subregions. The mean number of small grid cells within each connected subregion is calculated, and connected regions with values ​​below the mean are removed, thereby eliminating small isolated regions caused by drift or noise (using Algorithm 2). This method ensures that the extracted parking functional areas have both practical operational significance and strong robustness to data fluctuations.

[0029] Figure 2 The figure shows the process for extracting ship berthing functional zones. First, track points are identified and classified as mooring or anchoring (Step 1; existing methods for identifying and classifying ship stopover behavior include methods based on ship motion feature constraints, clustering, and deep learning). For the harbor and offshore areas, appropriate Geohash levels are selected (level 7 for harbor and level 6 for offshore) to balance spatial resolution and computational efficiency (Step 2). A connectivity-based clustering method is used to segment the berthing area into functional zones with practical operational significance (Step 3). Finally, a Geohash grid is used to filter the berthing points, removing outliers and outliers (Step 4; Outlier grids: Calculate the mean number of track points within each connected subregion and remove connected areas with fewer track points than the mean. Outliers are grids with few track points. When extracting frequently used stopover areas, temporary stops can be considered anomalies in the extraction process). The method for determining whether a grid is inside or outside the harbor is: when the number of trajectory points within the grid that are inside the harbor is greater than the number of trajectory points that are outside the harbor, the window is inside the harbor; otherwise, the window is outside the harbor. The method for determining whether a trajectory point is inside or outside the harbor is: when the level 6 Geohash grid to which the trajectory point belongs is within the predefined coastline Geohash set, the trajectory point is inside the harbor; otherwise, it is outside the harbor.

[0030] By analyzing the ship entry and exit points obtained by Algorithm 1, it can be found that there are multiple areas with extremely uneven distribution of entry and exit events in the port, indicating that there are multiple main entry and exit channels in the port (see Figure 3 , step 1). To facilitate efficient retrieval and analysis of large maritime areas, a level-6 geohash grid is used to encode relevant areas. Taking Port Entry Area 1 as an example, the method involves extracting the vessel's trajectory segments from the port entry point to the first anchorage, as well as the path it takes into the port area. Heatmap visualization of these trajectories reveals that many ships prioritize heading to the anchorage upon arrival to await further dispatch, while some ships depart without entering the port area or anchorage.

[0031] To extract the primary channels, the number of trajectory points within each Geohash grid is counted. Algorithm 2 is then used to remove grids with values ​​below the mean or isolated locations to reduce noise and highlight the primary channels. This process effectively isolates the primary approach channel associated with Port 1. The same method can be applied to other approach and exit regions, achieving robust extraction of the primary approach and exit channels. This method ensures that the ultimately extracted channels accurately reflect the primary ship flow directions and are robust against outliers or sparse data.

[0032] Figure 3 The figure shows the extraction process for the main channels for ship arrivals and departures. First, areas with significantly uneven distribution of arrival and departure events are identified, with darker points representing arrivals and lighter points representing departures (Step 1; if the number of arrivals within the same Geohash is more than ten times the number of departures, or if there are only arrivals, the channel is considered an arrival channel. Similarly, if the number of departures within the same Geohash is more than ten times the number of arrivals, or if there are only departures, the channel is considered an departure channel). These areas are then encoded using a level-6 Geohash grid (Step 2). Subsequently, the vessel trajectory segments from the arrival point to the first anchorage are extracted and displayed as a heat map, revealing the movement patterns of ships to areas such as anchorages (the brighter grids in Steps 3 and 4). Next, a trajectory grid is constructed, and outlier grids are removed to reduce noise (Step 5). Finally, the main channel is delineated to ensure that the extracted channel can accurately reflect the main flow path of the ship and has strong anti-anomaly capabilities (Step 6: The process of delineating the main channel is: first remove the grids in the Geohash grid where the number of trajectory points is lower than the grid trajectory point mean, and then divide the grid into connected sub-regions. The largest connected region is the main channel).

[0033] The functional area extraction algorithm was tested using real AIS data published by the National Oceanic and Atmospheric Administration (https: / / coast.noaa.gov / htdata / CMSP / AISDataHandler / 2021 / ).

[0034] Anchor area extraction: In order to improve the accuracy of anchorage extraction, this application selects the anchorage track points of ships with a length greater than 50 meters, and their Geohash distribution is as follows: Figure 4 As shown on the left, the visualization reveals that some areas contain outlier anchorages due to temporary vessel avoidance. By removing low-frequency and isolated anchorages, the main anchorages were ultimately identified and further divided into six sub-areas based on spatial connectivity (see the middle panel). Given the high traffic density within the harbor, a finer level 7 Geohash grid was used within the harbor, while a level 6 grid was used outside the harbor. Comparison with the official nautical chart (https: / / fishing-app.gpsnauticalcharts.com / i-boating-fishing-web-app / fishing-marine-charts-navigation.html?title=LOS+ANGELES+AND+LONG+BEACH+HARBORS+boating+app#11.59 / 33.6582 / -118.2340) shows that all annotated anchorages were successfully extracted, validating the integrity of the method. Furthermore, some previously unannotated anchorages were discovered (boxed areas in the middle sub-image). This suggests that due to heavy port traffic and limited throughput capacity, some ships are choosing to anchor outside the designated area while waiting to enter the port. This phenomenon may also be caused by discrepancies between planned and actual arrival times, further affecting port scheduling and exacerbating the waiting period outside the port.

[0035] Figure 4 The anchorage extraction results show the grid distribution of anchorages for ships longer than 50 meters. By removing isolated and low-frequency grids and clustering connected grids, the main anchorage areas were identified. A level 7 geohash grid was used inside the harbor, while a level 6 grid was used outside the harbor.

[0036] Extraction of main channel in and out of the port: By extracting the grids with significant differences in the number of arrival and departure trajectory points, four key areas were identified: two arrival areas and two departure areas. The berthing trajectory points were connected to these arrival and departure areas to obtain the following: Figure 5The four heat maps of ship trajectories shown above. The bright grids represent cells that are not connected to the main channel. Compared to the lower half of the image, the other unextracted grids represent Geohash cells with a lower-than-average trajectory point density. Comparing the main entrance and exit channels, it can be seen that the port has only one main entrance channel, while the exit is divided into two distinct branches. This phenomenon is due to the fact that some ships cannot receive immediate service after entering the port and must first wait at a designated anchorage. Since the right-hand entrance channel is closer to the port anchorage, oil tankers often refuel in this area, resulting in a higher density of trajectories directly into the port on this channel. In contrast, ships leaving the port do not require additional services and can leave the port directly, thus forming two exit branches. The above results reflect the differences in the operational processes of ships entering and leaving the port.

[0037] Figure 5 The main channel for ships entering and leaving the port is derived from ship trajectory data. The grids in the lower half of the image represent the identified main channels. Bright grids are outliers, and the remaining grids are low-frequency areas. Since ships entering the port typically anchor and await service, and the right entrance is closer to the port's anchorage, high-frequency entry channels are concentrated along the right entrance. In contrast, departing ships can leave the port directly, forming two main exit channels.

[0038] Figure 6 The main entry and exit channels extracted by the proposed method are demonstrated. The AIS data used in this application covers a wider area than the port boundaries defined in the nautical chart. The port entry channel on the left side of the nautical chart shows a downward sloping trend. This feature is accurately captured in our results of the main channel on the left side, which first slopes downward and then gradually becomes parallel. Similarly, the main channel at the bottom of the result also well reflects the upward trend of the channel in the nautical chart. The proposed method can further explore the main navigation paths for ships to enter the port area directly from the entrance channel, providing valuable route reference for port entry operations. In addition, the channels extracted by the proposed method near the two entrances to the port are highly consistent with the positions of the buoys on the nautical chart, further verifying the accuracy of the method. These results show that data-driven methods not only supplement the information of traditional nautical charts, but also expand their content. By revealing detailed ship navigation patterns in the port, this community method provides practical guidance for ship path planning, enhances the situational awareness of port management departments, and contributes to more efficient and safer port operations.

[0039] Figure 6 This is the extraction result of the main channel for entering and exiting the port. Due to the port's limited throughput capacity, ships often need to anchor and wait after entering the port. Therefore, nautical charts only mark the entrance channel and do not provide a detailed approach path. Compared with the entrance channel shown in the nautical chart, the method proposed in this application not only accurately captures the inclination trend of the entrance channel, but also reveals the complete structure of the main channel for entering and exiting the port.

[0040] In summary, the experimental results fully demonstrate the effectiveness and practical value of this framework for the automatic extraction of port functional areas. Further comparison with official nautical charts shows that the framework not only fully restores known functional areas but also discovers unlabeled active areas, demonstrating its data-driven innovation advantage. Overall, this method provides a solid data foundation and technical support for intelligent port management and maritime safety, and is expected to promote the continuous development of port management towards intelligent, refined, and efficient management.

[0041] The present application also provides a port functional area mining system based on geographic hash coding, which is implemented based on the above method and includes: A module for determining ship berthing functional areas is used to identify and classify trajectory points. Different Geohash levels are selected for the port and outer port areas, and a connectivity-based clustering method is used to segment the berthing area into multiple sub-areas. The berthing points are screened using Geohash grids, outliers and abnormal values ​​are removed, and functional areas with practical operational significance are formed. The module for determining the main channel for ships entering and leaving the port is used to identify areas with significantly uneven distribution of entry and exit points; the trajectory segments of ships from the entry point to the first berthing point are extracted and displayed in the form of a heat map; a connectivity-based clustering method is used to divide the area into multiple sub-areas, and outlier grids and outliers are removed to finally divide the main channel.

[0042] The present invention introduces multi-level geo-hash coding to achieve efficient spatial indexing and aggregation of ship trajectory data. Combining unsupervised behavior recognition with adaptive parameters and regional clustering based on connectivity, it can automatically and accurately extract various types of functional areas such as anchorages and waterways in ports and their adjacent waters. This method does not require manual labeling, can adapt to different port environments and diverse ship operation modes, and significantly improves the automation and intelligence level of functional area extraction. By deeply mining the spatial distribution characteristics of AIS trajectory data, the present invention not only improves the accuracy and robustness of functional area boundary determination, but also supports dynamic updating and multi-scale management of functional areas.

[0043] The implementation of this invention helps improve the efficiency of port space resource utilization, optimize ship scheduling and traffic organization, enhance port safety supervision and emergency response capabilities, and provide a solid data foundation and technical support for application scenarios such as smart port construction, shipping network optimization, and port planning and management. Furthermore, this method possesses excellent versatility and scalability, adapting to the actual needs of ports of different sizes and types, and promoting the development of intelligent, refined, and efficient port management.

[0044] The present application may also provide a computer device comprising: at least one processor, memory, at least one network interface, and a user interface. The various components in the device are coupled together via a bus system. It will be understood that the bus system is used to enable communication between these components. In addition to a data bus, the bus system also includes a power bus, a control bus, and a status signal bus.

[0045] The user interface may include a display, a keyboard, or a pointing device, such as a mouse, a trackball, a touchpad, or a touch screen.

[0046] It is understood that the memory in the embodiments disclosed in the present application may be a volatile memory or a non-volatile memory, or may include both volatile and non-volatile memories. Among them, the non-volatile memory may be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. The volatile memory may be a random access memory (RAM), which is used as an external cache. By way of example and not limitation, many forms of RAM are available, such as static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDRSDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link dynamic random access memory (SLDRAM), and direct RAM bus random access memory (DRRAM). The memories described herein are intended to include, but are not limited to, these and any other suitable types of memory.

[0047] In some embodiments, the memory stores the following elements, executable modules or data structures, or a subset or an extension thereof: an operating system and applications.

[0048] The operating system includes various system programs, such as the framework layer, core library layer, and driver layer, which are used to implement various basic services and handle hardware-based tasks. Application programs include various application programs, such as media players and browsers, which are used to implement various application services. The program that implements the method of the embodiment of the present disclosure can be included in the application program.

[0049] In the above embodiment, the processor may also call a program or instruction stored in the memory, specifically, a program or instruction stored in the application program, to: Perform the steps of the above method.

[0050] The above method can be applied to or implemented by a processor. The processor may be an integrated circuit chip with signal processing capabilities. During implementation, each step of the above method can be completed by hardware integrated logic circuits in the processor or by software instructions. The above processor may be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. The above-disclosed methods, steps, and logic block diagrams can be implemented or executed. The general-purpose processor may be a microprocessor or any conventional processor. The steps of the above-disclosed method can be directly implemented and executed by a hardware decoding processor, or by a combination of hardware and software modules in the decoding processor. The software module can be located in a storage medium well-known in the art, such as random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, etc. The storage medium is located in the memory, and the processor reads the information in the memory and, in conjunction with its hardware, completes the steps of the above method.

[0051] It is understood that the embodiments described herein may be implemented using hardware, software, firmware, middleware, microcode, or a combination thereof. For hardware implementation, the processing unit may be implemented in one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), general-purpose processors, controllers, microcontrollers, microprocessors, or other electronic units or combinations thereof for performing the functions described herein.

[0052] For software implementation, the technology of the present application can be implemented by executing the functional modules (e.g., procedures, functions, etc.) of the present application. The software code can be stored in a memory and executed by a processor. The memory can be implemented in the processor or external to the processor.

[0053] The present application may also provide a non-volatile storage medium for storing a computer program. When the computer program is executed by a processor, each step in the above method embodiment can be implemented.

[0054] Finally, it should be noted that the above embodiments are intended only to illustrate the technical solutions of this application and are not intended to limit the scope of the present invention. Although this application has been described in detail with reference to the embodiments, it should be understood by those skilled in the art that modifications or equivalent substitutions to the technical solutions of this application do not depart from the spirit and scope of the technical solutions of this application and should be encompassed by the claims of this application.

Claims

1. A method for mining port functional areas based on geo-hash coding, comprising: Determine the ship berthing functional area: identify and classify the berthing points; For the port and outer areas, different Geohash levels were selected and a connectivity-based clustering method was used to segment the berthing area into multiple sub-areas. The berthing points were screened using Geohash grids, outliers and abnormal values ​​were removed, and functional areas with practical operational significance were formed. Determine the main channel for ships entering and leaving the port: Identify areas with significantly uneven distribution of entry and exit points; extract the trajectory segments of ships from the port entry point to the first berth point and display them in the form of a heat map; use a connectivity-based clustering method to split the area into multiple sub-regions, remove outlier grids, and finally divide the main channel.

2. The method for mining port functional areas based on geographic hash coding according to claim 1 is characterized in that: The use of Geohash grids to screen the anchor points and remove outliers and abnormal values ​​includes: Calculate the mean number of trajectory points in each sub-region and remove sub-regions with fewer trajectory points than the mean. Remove sub-regions with fewer than a set threshold number of track points.

3. The method for mining port functional areas based on geographic hash coding according to claim 1 is characterized in that: The method of segmenting the parking area by using a connectivity-based clustering method includes: For each unvisited grid, a new sub-region is initialized, and all grids adjacent to the grid that have not been visited are recursively searched and added to the sub-region. During the traversal process, the visited grids are marked to avoid repeated searches. When all reachable neighbors are traversed, the current sub-region is added to the sub-region list. This process is repeated until all grids are assigned to a sub-region.

4. The method for mining port functional areas based on geographic hash coding according to claim 1 is characterized in that: The areas identified as having a significant imbalance in the distribution of entry and exit points include: If the number of port entry points in the same grid is a multiple of the number of port exit points or there are only port entry points, the waterway is an entry waterway; if the number of port exit points in the same grid is a multiple of the number of port entry points or there are only port exit points, the waterway is an exit waterway.

5. The method for mining port functional areas based on geographic hash coding according to claim 4 is characterized in that: The method for determining the port entry point and the port exit point includes: First, the trajectory points within the boundary grid are marked as target points. For each target point, the time difference between it and the previous and next trajectory points is calculated. If the time interval before the target point exceeds the set time threshold, or it is the first point and the subsequent interval is within the set time threshold, it is determined to be an entry point. If the time interval after the target point exceeds the set time threshold, or it is the last point and the previous interval is within the set time threshold, it is determined to be a departure point.

6. The method for mining port functional areas based on geographic hash coding according to claim 1, characterized in that: The method for dividing the main channel includes: The sub-region with the most connected grids is selected as the main channel.

7. A port functional area mining system based on geographic hash coding, implemented based on the method according to any one of claims 1 to 6, characterized in that: The system comprises: Determine the ship berthing functional area module, which is used to identify and classify the berthing points; select different Geohash levels for the port and the port area, and use a connectivity-based clustering method to divide the berthing area into multiple sub-areas; use Geohash grids to screen the berthing points, remove outliers and outliers, and form functional areas with practical operational significance; and The module for determining the main channel for ships entering and leaving the port is used to identify areas with significantly uneven distribution of entry and exit points; the trajectory segments of ships from the entry point to the first berthing point are extracted and displayed in the form of a heat map; a connectivity-based clustering method is used to divide the area into multiple sub-areas, and outlier grids and outliers are removed to finally divide the main channel.

Images