A method, program product, device, and storage medium for ice ridge identification

Abstract

Embodiments of the present application disclose an ice ridge identification method, a program product, an apparatus and a storage medium, comprising: acquiring three-dimensional point cloud data of an ice surface; performing topographic processing on the ice surface to obtain each grid corresponding to the ice surface, and determining an elevation reference value according to the three-dimensional point cloud data of all the grids; determining an elevation anomaly area in the ice surface according to the three-dimensional point cloud data of each grid and the elevation reference value; performing ice ridge area feature analysis on the elevation anomaly area based on the three-dimensional point cloud data of the elevation anomaly area, and determining an ice ridge area according to each ice ridge area feature analyzed; the ice ridge area features include spatial continuity, area size, area area, point cloud density, edge change or morphological extension direction; performing ice ridge feature identification on the ice ridge area, and determining an ice ridge identification result according to each ice ridge feature identified. The method improves the accuracy and stability of ice ridge identification during ship navigation, and ensures the navigation safety of the ship in a complex polar ice area.

Description

Technical Field

[0001] The embodiments of the present invention relate to the field of marine engineering technology, and in particular to an ice ridge identification method, program product, device and storage medium. Background Technology

[0002] During polar navigation, the sea ice environment is complex and ever-changing. In addition to relatively flat ice surfaces, there may be ice ridges formed by the compression and accumulation of sea ice. Ice ridges are usually higher than the surrounding ice surface and may have a large local thickness and a strong obstructive effect. For polar vessels, icebreakers, or vessels navigating in ice-covered areas, failure to identify ice ridges ahead in time may lead to a decrease in ship speed, an increase in navigation resistance, abnormal stress on the hull, or the need to temporarily adjust the course.

[0003] Current methods for ice condition identification in polar navigation typically combine camera, radar, remote sensing images, or ice condition data with manual assistance to identify ice ridges. For example, camera images are used to analyze the features of the ice surface to determine if there are ice ridges ahead of the ship that could pose a navigational hazard. However, image analysis methods often rely on texture, color, shadows, or outlines to identify ice ridges. When the ice surface colors are similar, shadows are not obvious, the distance is far, or the image contrast is low, the difference between the ice ridge and the surrounding ice surface may not be prominent enough, leading to inaccurate and unstable identification of ice ridges and reducing the safety of ship navigation. Summary of the Invention

[0004] This invention provides an ice ridge identification method, program product, device, and storage medium, which can improve the accuracy and stability of ice ridge identification during ship navigation and ensure the safety of ship navigation.

[0005] In a first aspect, embodiments of the present invention provide a method for identifying ice ridges, comprising: Acquire 3D point cloud data of the ice surface; The ice surface is subjected to topographic processing to obtain each grid corresponding to the ice surface, and the elevation reference value is determined based on the three-dimensional point cloud data of all grids. The elevation anomaly areas in the ice surface are determined based on the three-dimensional point cloud data of each grid and the elevation reference value. Based on the three-dimensional point cloud data of the elevation anomaly region, the ridge region features of the elevation anomaly region are analyzed, and the ridge region is determined according to the analyzed ridge region features. The ridge region features include spatial continuity, region size, region area, point cloud density, edge variation or morphological extension direction. The ice ridge region is subjected to ice ridge feature identification, and the ice ridge identification result is determined based on the identified ice ridge features; the ice ridge features include ice ridge size features, slope change features, elevation features, or positional relationship features.

[0006] Secondly, embodiments of the present invention provide an ice ridge identification device, the device comprising: The data acquisition module is used to acquire three-dimensional point cloud data of the ice surface; The reference value determination module is used to perform terrain processing on the ice surface to obtain each grid corresponding to the ice surface, and determine the elevation reference value based on the three-dimensional point cloud data of all grids. The first region determination module is used to determine the elevation anomaly region in the ice surface based on the three-dimensional point cloud data of each grid and the elevation reference value; The second region determination module is used to perform ridge region feature analysis on the elevation anomaly region based on the three-dimensional point cloud data of the elevation anomaly region, and determine the ridge region according to the analyzed ridge region features; the ridge region features include spatial continuity, region size, region area, point cloud density, edge variation or morphological extension direction; The ice ridge identification module is used to identify ice ridge features in the ice ridge region and determine the ice ridge identification result based on the identified ice ridge features; the ice ridge features include ice ridge size features, slope change features, elevation features, or positional relationship features.

[0007] Thirdly, embodiments of the present invention also provide an electronic device, the electronic device including a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the ice ridge identification method as described in any of the embodiments of the present invention.

[0008] Fourthly, embodiments of the present invention also provide a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the ice ridge identification method as described in any of the embodiments of the present invention.

[0009] Fifthly, embodiments of the present invention provide a computer program product, including a computer program that, when executed by a processor, implements an ice ridge identification method as described in any of the embodiments of the present invention.

[0010] The method of this invention dynamically determines elevation reference values ​​using real-time acquired 3D point cloud data. These dynamic elevation reference values ​​then accurately identify elevation anomaly areas. Furthermore, irrelevant areas are precisely excluded based on the spatial continuity, size, and edge morphology of these anomaly areas, thus identifying potential ice ridge regions. For these ice ridge regions, multi-dimensional ice ridge features are used for accurate ice ridge identification, achieving adaptive identification from raw point cloud data to target ice ridges. Ice ridges can be accurately identified even when ice surface colors are similar and shadows are indistinct, significantly improving the accuracy, objectivity, and stability of ice ridge identification during ship navigation and effectively ensuring the safety of ship navigation in complex polar ice regions. Attached Figure Description

[0011] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0012] Figure 1 A flowchart of an ice ridge identification method provided in an embodiment of the present invention; Figure 2 This is a schematic diagram of the structure of the ice ridge identification device provided in an embodiment of the present invention; Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0013] The present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and not intended to limit it. Furthermore, it should be noted that, for ease of description, the accompanying drawings show only the parts relevant to the invention and not all structures. The acquisition, storage, use, and processing of data in the technical solutions of this application comply with relevant national laws and regulations. It should be noted that, in the embodiments of this application, certain existing industry solutions such as software, components, and models may be mentioned. These should be considered exemplary, intended only to illustrate the feasibility of implementing the technical solutions of this application, and do not imply that the applicant has already used or necessarily used the relevant content of such solutions.

[0014] Figure 1This is a flowchart illustrating an ice ridge identification method provided by an embodiment of the present invention. The method of this embodiment can improve the accuracy and stability of ice ridge identification during ship navigation, ensuring the safety of ship navigation. This method can be executed by an ice ridge identification device provided by an embodiment of the present invention, which can be implemented in software and / or hardware. The following embodiments will illustrate this using the integration of the device into an electronic device as an example. The electronic device can be a server or computer device, etc., used to implement an ice ridge identification method. (Refer to...) Figure 1 The method may specifically include the following steps:

[0015] Step 101: Obtain the three-dimensional point cloud data of the ice surface.

[0016] The 3D point cloud data is collected by a pre-installed lidar system on the ship during navigation. The lidar can be installed in a high position forward of the bridge, above the bow, or on the forward side of the ship, thus covering the largest possible forward area. Furthermore, the lidar's installation height is above the ice surface in front of the ship, allowing its laser beam to cover a wider lateral area when scanning downwards. The lidar's scanning area must at least cover the forward area near the ship's intended path, and can also cover a certain lateral range on both sides of the ship to identify ice ridges that may affect the ship's path selection.

[0017] During ship navigation, lidar emits laser pulses forward continuously or periodically and receives the echoes to obtain three-dimensional point cloud data. Each point in the three-dimensional point cloud data corresponds to the location of a target surface in the space ahead. Targets include relatively flat ice surfaces, scattered ice blocks, protruding ice ridges, snow-covered areas, and other obstacles such as floating ice floes and iceberg fragments.

[0018] 3D point cloud data may contain various noises and irrelevant information, and directly using it for ice ridge identification can lead to false positives and false negatives. In this solution, after obtaining the 3D point cloud data, it can be preprocessed to remove various noises and data irrelevant to ice ridge identification, facilitating subsequent use of the preprocessed 3D point cloud data for ice ridge identification and improving the efficiency of ice ridge identification. Specifically, since the lidar is installed on the ship, its scanning field of view may cover the ship's own structure at certain angles. After obtaining the 3D point cloud data, the point cloud data of the ship's hull structure can be extracted based on the pre-determined spatial position and geometry of the hull structure in the lidar coordinate system. The 3D point cloud data may also include outliers, which are isolated points that are spatially inconsistent with the surrounding point cloud. Outliers are characterized by their small number, random location, and lack of spatial continuity with surrounding points. In the 3D point cloud data excluding the ship's hull structure, the average distance between each point and its nearest neighbors is calculated. If the average distance corresponding to a point is large (e.g., exceeding a preset distance), then that point is identified as an outlier, and all outliers in the 3D point cloud data are further removed. The effective ranging range of a lidar is limited, and point cloud data outside this range may be invalid. Therefore, point cloud data outside the effective ranging range can be filtered and discarded.

[0019] Specifically, polar environments may contain water mist, snow dust, and suspended particles at the air-sea interface. When a laser pulse encounters these suspended particles during propagation, it generates echoes, creating false point cloud data. The reflectivity of water mist and snow dust particles is much lower than that of ice surfaces, therefore the echo intensity corresponding to suspended particles is significantly weaker. Abnormal echoes caused by water mist or snow dust can be identified and suppressed based on preset echo intensity thresholds.

[0020] The point cloud data acquired by lidar is initially located in the lidar's own local coordinate system. To facilitate subsequent processing and application, the point cloud data acquired by each lidar can be transformed into a unified basic coordinate system. After completing the basic coordinate transformation, a forward observation coordinate system is further established with the ship's current position as the origin and the ship's course as the longitudinal axis. The longitudinal coordinate of the observation coordinate system represents the fore-and-aft distance along the course, the lateral coordinate represents the left-right offset perpendicular to the course, and the vertical coordinate represents the altitude. The purpose of establishing the observation coordinate system is to visually align the point cloud data with the ship's navigation direction.

[0021] Step 102: Perform terrain processing on the ice surface to obtain the corresponding grids, and determine the elevation reference value based on the three-dimensional point cloud data of all grids.

[0022] Topographic processing is the process of transforming continuously distributed 3D point cloud data into discrete, regularized spatial unit representations. In the original point cloud data, points are continuously distributed, each existing independently without a clear spatial organizational relationship. This disordered state is detrimental to subsequent processing. Topographic processing divides the observation area into several regular spatial grids, each grid becoming an independent data processing unit. Statistical analysis can be performed on the point cloud data falling within each grid, thus transforming the disordered point cloud into an ordered terrain representation. The elevation reference value is a representative elevation value calculated based on the point cloud data contained in each grid, used to characterize the typical elevation level of the surrounding ordinary ice surface at the grid's location. In this scheme, optionally, the elevation reference value is determined based on the 3D point cloud data of all grids, including: performing scene feature analysis on the ice surface based on the 3D point cloud data of all grids to determine the data scene in which the ice surface is located; and calculating the elevation reference value based on the elevation reference value calculation method corresponding to the data scene.

[0023] The scene features include ice surface undulation features, ice surface tilt features, ship attitude features, observation distance, and point cloud density features; the data scenes include sparse ice ridge scenes, moderate ice ridge scenes, dense first ice ridge scenes, and dense second ice ridge scenes. Ice surface undulation features describe the natural fluctuation of ice surface elevation, reflecting the roughness of the ice surface itself. Specifically, all grids are traversed, and the elevation data of the point cloud within each grid is extracted, and the standard deviation of all point cloud elevations is calculated. The larger the standard deviation, the more severe the ice surface undulation and the rougher the ice surface. Ice surface tilt features describe whether there is an overall tilting trend in the ice surface. Ice surface tilt may be caused by changes in ship attitude or large-area ice surface slopes. Specifically, the ratio of the elevation difference between adjacent grids to the grid spacing is calculated along the longitudinal and transverse directions to obtain the longitudinal gradient and transverse gradient. The gradient distribution of all grids is statistically analyzed, and the average gradient and gradient standard deviation are calculated. If the average gradient is large, it indicates that the ice surface has a significant overall tilt. If the gradient standard deviation is small, it indicates that the tilt is relatively uniform; if the gradient standard deviation is large, it indicates that the ice surface undulation is irregular.

[0024] Ship attitude characteristics describe the degree to which the ship's own motion state affects point cloud data. Specifically, data from the ship's inertial measurement unit (IMU) can be acquired, including pitch, roll, and heave displacement. The range and standard deviation of pitch and roll angles over a recent period are calculated. Large ranges and standard deviations indicate ship instability and continuous hull movement, leading to constant changes in the lidar scanning angle and affecting the consistency of point cloud elevation measurements. Observation distance characteristics describe the differences in point cloud density at different distances. Typically, point clouds are denser in closer areas and sparser in farther areas. Point cloud density characteristics describe the spatial distribution of point cloud data (determined by the number of points), reflecting the statistical applicability of the data.

[0025] Specifically, different data scenarios correspond to different methods for determining elevation reference values. Data scenarios include sparse ice ridge scenarios, moderate ice ridge scenarios, scenarios with a dense first ice ridge and scenarios with a dense second ice ridge. The dense first ice ridge scenario is characterized by dense ice ridges and significant ice surface undulations, while the dense second ice ridge scenario is characterized by dense ice ridges, significant ice surface tilt, and large changes in ship attitude. Based on ice surface undulation characteristics, ice surface tilt characteristics, ship attitude characteristics, observation distance, and point cloud density characteristics, the specific data scenario category can be determined. After determining the data scenario, the elevation reference value is calculated according to the corresponding elevation reference value calculation method. In one optional implementation, when the data scenario is a sparse ice ridge scenario, the average elevation of all grids is calculated based on the 3D point cloud data of all grids, and the average elevation is determined as the elevation reference value; when the data scenario is a moderate ice ridge scenario, the median elevation of all grids is calculated based on the 3D point cloud data of all grids, and the median elevation is determined as the elevation reference value; when the data scenario is a first dense ice ridge scenario, the lower quantile elevation of all grids is calculated based on the 3D point cloud data of all grids, and the lower quantile elevation is determined as the elevation reference value; when the data scenario is a second dense ice ridge scenario, the fitting plane of each grid is determined based on the 3D point cloud data of each grid, and the reference elevation of each grid is determined based on the fitting plane of each grid, and the reference elevation of each grid is determined as the elevation reference value.

[0026] A sparse ice ridge scenario indicates that the ice surface in front is generally flat, with few and scattered ice ridges. In this scenario, the elevation distribution in the 3D point cloud data is relatively concentrated, with the elevation values ​​of most points close to the normal ice surface elevation. The average elevation is then determined as the elevation reference value. Under the premise of sparse ice ridges, the average elevation can accurately reflect the typical elevation level of the surrounding ordinary ice surface and will not be significantly deviated by individual ice ridge points.

[0027] A moderate number of ice ridges indicates that the ice surface has some undulations, with a moderate number of ridges, but not yet densely distributed. In this scenario, some high-value points in the 3D point cloud data correspond to ice ridges, but most are still concentrated near the normal ice surface elevation. Continuing to use the average elevation would cause a slight overestimation, potentially leading to the missed detection of some small ice ridges. When the data scenario is a moderate number of ice ridges, all elevation values ​​corresponding to the point cloud data are sorted from smallest to largest, and the value in the middle position is determined as the median elevation. This median elevation is then used as the elevation reference value.

[0028] The first dense ice ridge scenario indicates a relatively high density of ice ridges and significant ice surface undulation, but without a clear overall tilt, and the ship's attitude is relatively stable. In this scenario, the median elevation may have been artificially inflated by the ice ridges and no longer represent a normal ice surface. After identifying the data scenario as the first dense ice ridge scenario, the elevation values ​​of all point cloud data are sorted from smallest to largest. The lower quantile elevation is determined according to a preset quantile ratio and used as the elevation reference value. The quantile ratio can be dynamically adjusted based on the density of the ice ridges; the denser the ice ridges, the lower the quantile ratio.

[0029] The second dense ice ridge scenario indicates a relatively dense distribution of ice ridges and a significant overall tilt of the ice surface, or drastic changes in ship attitude. In this scenario, no single elevation reference value can accurately describe the spatial variations of the ice surface. After identifying the current scenario as a second dense ice ridge scenario, 3D point cloud data is extracted for each grid cell and its neighboring grid cells (e.g., a 3x3 or 5x5 area around the current grid cell). Based on this 3D point cloud data, a local plane is fitted using the least squares method. The fitted plane equation is used as the local reference plane for the current grid cell. Substituting the horizontal coordinates of the current grid cell's center position into the fitted plane equation yields the reference elevation of the current grid cell.

[0030] By employing a scene-adaptive elevation reference calculation method, reliable elevation reference values ​​can be obtained under different ice conditions and environmental conditions, providing an accurate reference for subsequent elevation anomaly detection and thus improving the accuracy of ice ridge identification.

[0031] Step 103: Determine the elevation anomaly areas in the ice surface based on the 3D point cloud data and elevation reference values ​​of each grid.

[0032] An elevation anomaly region is a continuous spatial area on the ice surface that exhibits a significant elevation rise relative to the surrounding normal ice surface; that is, an area where ice ridges may exist. In one optional implementation, the elevation anomaly region on the ice surface is determined based on the 3D point cloud data and elevation reference values ​​of each grid, including: calculating the elevation difference of each grid based on the 3D point cloud data and elevation reference values ​​of each grid; determining adjustment terms based on the data scenario, and adjusting a pre-determined benchmark threshold based on the adjustment terms to obtain a target threshold; determining each elevation anomaly grid based on the elevation difference of each grid and the target threshold, and determining the elevation anomaly region based on all elevation anomaly grids.

[0033] The adjustment items include observation distance adjustment, point cloud quality adjustment, ship attitude adjustment, and ice surface roughness adjustment. The elevation difference directly reflects the degree of protrusion of each location on the ice surface relative to the local normal ice surface. For each grid, all 3D point cloud data within the current grid are traversed, and the elevation coordinates of each point are extracted. The elevation value of each point is subtracted from the elevation reference value of the current grid to obtain the elevation difference of each point. If the elevation difference is positive, it indicates that the point is higher than the surrounding normal ice surface; if the elevation difference is zero or close to zero, it indicates that the point is level with the surrounding ice surface; if the elevation difference is negative, it indicates that the point is lower than the surrounding ice surface, possibly a depression or melt pool area. After calculating the elevation difference of all points within the grid, the overall elevation difference characteristics of the grid can be further statistically analyzed. For example, the average elevation difference of all points within the grid is calculated to obtain the average elevation difference of the grid; the maximum elevation difference of all points within the grid is calculated to obtain the maximum elevation difference of the grid; and the proportion of points with positive elevation differences is statistically analyzed to obtain the proportion of elevation anomalies in the grid.

[0034] The baseline threshold is preset, while the target threshold is obtained by adjusting the baseline threshold based on actual conditions. It is used to determine whether a grid belongs to an abnormal region. The observation distance adjustment term compensates for the decrease in lidar measurement accuracy with increasing distance. The laser pulses emitted by the lidar experience energy attenuation and scattering during propagation; the farther the distance, the weaker the echo signal and the greater the measurement noise. Therefore, for distant grids, the threshold needs to be appropriately relaxed to avoid false anomaly detections caused by measurement noise. Specifically, the distance from each grid or point to the lidar can be calculated, and based on a preset distance-related error coefficient, the distance is converted into elevation measurement uncertainty, thus obtaining the observation distance adjustment term. The greater the observation distance, the larger the observation distance adjustment term, and the higher the target threshold.

[0035] Point cloud quality adjustment terms reflect the reliability of 3D point cloud data. Point cloud quality can be determined based on point cloud density, echo intensity, and measurement consistency. Specifically, for each grid, the point cloud density within the current grid is determined (i.e., the number of points divided by the grid area); the echo intensity of each point is statistically analyzed, and the average echo intensity or the proportion of strong echoes is calculated; the dispersion of point cloud elevation within the current grid is analyzed, and measurement consistency is determined based on the dispersion. Further, the point cloud quality level is determined based on point cloud density, echo intensity, and measurement consistency, and the point cloud quality adjustment term is determined based on the point cloud quality level. For grids with high quality levels, the point cloud quality adjustment term is smaller or zero; for grids with low quality levels, the point cloud quality adjustment term is larger, and the target threshold is correspondingly increased.

[0036] The ship attitude adjustment term is used to compensate for the impact of ship motion on elevation measurement and benchmark estimation. During navigation, ships experience attitude changes such as pitch, roll, and heave, which cause continuous changes in the scanning angle and position of the lidar, thus affecting the consistency of point cloud elevation measurements. Specifically, attitude data output from the ship's inertial measurement unit (IMU), including pitch, roll, and heave displacement, is acquired. The range and standard deviation of each attitude parameter over a recent period are calculated, and attitude stability is assessed based on these ranges and standard deviations. When attitude changes are drastic (poor attitude stability), the ship attitude adjustment term can be increased.

[0037] The ice surface roughness adjustment term is used to adapt to the natural undulations of the ice surface itself. Specifically, the distribution characteristics of elevation differences are statistically analyzed in normal ice surface areas beforehand. Grids that are clearly not ice ridges (with very stable elevation difference distributions) are selected, and the standard deviation or range of their elevation differences is calculated and used as a quantitative indicator of ice surface roughness. The greater the ice surface roughness, the larger the fluctuation range of elevation differences in normal undulations, requiring an increase in the target threshold to distinguish between normal undulations and true anomalies, i.e., increasing the ice surface roughness adjustment term. Furthermore, the baseline threshold is added to each adjustment term to obtain the final target threshold.

[0038] Furthermore, each elevation anomaly grid is determined based on the elevation difference of each grid and the target threshold, and the elevation anomaly region is determined based on all elevation anomaly grids.

[0039] For each grid cell, the elevation difference of the current grid cell is compared with the corresponding target threshold. If the elevation difference of the current grid cell is greater than the target threshold, it indicates that the point cloud elevation within the current grid cell is significantly higher than the surrounding normal ice surface, and the current grid cell is identified as an elevation anomaly grid. Conversely, if the elevation difference is less than or equal to the target threshold, the current grid cell is identified as a normal ice surface grid. After identifying all elevation anomaly grid cells, which may be spatially discrete, to facilitate the identification of elevation anomaly regions, all elevation anomaly grid cells can be traversed to determine the adjacency relationship between each grid cell and its surrounding grid cells. If two elevation anomaly grid cells are adjacent vertically, horizontally, or diagonally, they are identified as belonging to the same connected region. Starting from the first elevation anomaly grid cell, a breadth-first or depth-first search is used to identify all connected elevation anomaly grid cells and mark them as elevation anomaly regions. The next elevation anomaly grid cell is then selected, and the above process is repeated until all elevation anomaly grid cells are assigned to a certain region, thus obtaining the elevation anomaly region.

[0040] By introducing multiple adjustment factors such as observation distance, point cloud quality, ship attitude, and ice surface roughness, the target threshold can adapt to measurement uncertainties and environmental interference in different scenarios. This avoids missed detections and false detections of fixed thresholds under long distances, low-quality data, or complex ice conditions, and provides a reliable data foundation for subsequent ice ridge feature screening.

[0041] Step 104: Based on the 3D point cloud data of the elevation anomaly area, perform glacier ridge region feature analysis on the elevation anomaly area, and determine the glacier ridge region based on the analyzed glacier ridge region features.

[0042] Among them, the characteristics of ice ridge regions include spatial continuity, regional size, regional area, point cloud density, edge variation or morphological extension direction.

[0043] Spatial continuity describes the coherence of point cloud distribution within an elevation anomaly region. It reflects whether the region consists of interconnected points forming a continuous whole, rather than scattered isolated points or fragmented segments. For example, the distance between each point and its neighbors can be calculated; if the distance is less than a preset connectivity threshold, the two points are considered connected. Region size describes the scale of the elevation anomaly region in three-dimensional space, including width, length, and height. Region area describes the spatial extent of the elevation anomaly region on a horizontal plane. Point cloud density describes the density of LiDAR point cloud distribution within the elevation anomaly region, reflecting the reliability of the data and the accuracy of the measurement. Edge variation describes the topographic transition characteristics at the boundary between the elevation anomaly region and the surrounding normal ice surface, reflecting whether the region exhibits the steep edge morphology typical of ice ridges. Morphological extension direction describes the main spatial extension trend of the elevation anomaly region, indicating whether it possesses the linear or banded extension characteristics typical of ice ridges. For example, the aspect ratio of the elevation anomaly region can be calculated. Ice ridges typically have an aspect ratio greater than 3:1 or even higher, and are clearly banded or ridge-shaped. If the aspect ratio is close to 1:1, the elevation anomaly region is more likely to be an isolated protrusion that is circular or elliptical.

[0044] After obtaining the regional characteristics of each ice ridge in the elevation anomaly region, interference areas that do not conform to the basic morphological characteristics of ice ridges are eliminated based on these characteristics, retaining only ice ridge regions with typical ice ridge features. Specifically, elevation anomaly regions that meet the corresponding preset conditions for each feature analysis can be identified as ice ridge regions. For example, discontinuous elevation anomaly regions can be eliminated first, and then the remaining regions can be filtered by area to eliminate regions with too small an area; by horizontal width and vertical length to eliminate regions whose dimensions do not conform to the ice ridge characteristics; by point cloud density to eliminate sparse and false regions; by edge variation to eliminate regions with inconsistent edge shapes; and finally by shape extension direction to eliminate regions with non-elongated shapes (ice ridge shapes), thus obtaining the final ice ridge regions.

[0045] Step 105: Identify ice ridge features in the ice ridge region and determine the ice ridge identification results based on the identified ice ridge features.

[0046] Ice ridge features include size characteristics, slope variation characteristics, elevation characteristics, and positional characteristics. Size characteristics include relative height, width, and continuous length; elevation characteristics represent the elevation difference between the ice ridge region and the smooth ice surface; positional characteristics represent the positional relationship of the ice ridge region relative to ship tracks; and point cloud density characteristics are also included.

[0047] The relative height feature reflects the vertical protrusion of the ice ridge crest relative to the surrounding smooth ice surface. It can be obtained by calculating the statistical difference between the elevation of each point within the ice ridge region and the reference elevation. The width feature reflects the lateral scale of the ice ridge perpendicular to its main extension direction. It can be obtained by calculating the vertical span after determining the main direction through principal component analysis. The continuous length feature reflects the longitudinal scale of the ice ridge along its main extension direction. It can be obtained by calculating the span along the main direction. The slope variation feature describes the steepness of the ice ridge region's edges and the internal elevation variation pattern. It can be obtained by calculating the elevation gradient between the boundary points of the ice ridge region and the surrounding normal ice surface points, as well as the local slope statistics between adjacent points within the ice ridge region. The elevation feature indicates the degree of elevation difference between the ice ridge region and the surrounding smooth ice surface. The positional relationship feature describes the spatial positional relationship of the ice ridge region relative to the ship's current track. It can be obtained by calculating the lateral offset distance, longitudinal distance, and relative azimuth angle between the centroid of the ice ridge region and the centerline of the ship's track. The point cloud density feature reflects the density of the lidar point cloud distribution within the ice ridge region.

[0048] The ice ridge identification result includes the target ice ridge and its confidence level. In one optional embodiment, ice ridge feature identification is performed on the ice ridge region, and the ice ridge identification result is determined based on the identified ice ridge features, including the following steps A1-A2:

[0049] Step A1: Calculate the scoring results of each ice ridge feature based on each ice ridge feature and its corresponding scoring criteria; adjust the basic feature weights corresponding to each ice ridge feature based on the pre-determined observation distance, ice surface type, and positional relationship features to obtain the target feature weights corresponding to each ice ridge feature.

[0050] Because the physical units and numerical ranges of various ice ridge features differ, after obtaining each feature, it can be normalized to convert it to a uniform numerical range. Different ice ridge features correspond to different scoring standards, and the scoring results for each feature can be determined based on these standards. For example, for the relative altitude feature, the scoring standard can be set to five levels. A normalized value between 0 and 0.2 is scored as one point, indicating insufficient altitude and an indistinct ice ridge feature. A normalized value between 0.2 and 0.4 is scored as two points, indicating low altitude and possibly a small ice ridge. A normalized value between 0.4 and 0.6 is scored as three points, indicating moderate altitude and conforming to general ice ridge features. A normalized value between 0.6 and 0.8 is scored as four points, indicating high altitude and a significant ice ridge feature. A normalized value between 0.8 and 1 is scored as five points, indicating very high altitude and a highly significant ice ridge feature.

[0051] The preset feature weights for each ice ridge feature reflect the importance of each feature for ice ridge identification in general scenarios. However, in practical applications, the relative importance of each feature changes in different scenarios. Therefore, the preset weights can be dynamically adjusted according to real-time observation conditions. Weight adjustment based on observation distance: When the ice ridge area is far from the ship, the lidar point cloud density decreases, and the measurement accuracy declines. Therefore, the weights of the corresponding ice ridge features (features strongly correlated with observation distance, such as slope change features, width features, and elevation features) can be reduced. Weight adjustment based on ice surface type: In flat ice areas, the contrast between the ice ridge and the surrounding ice surface is significant, and the ability to distinguish elevation and relative height features is strong. Therefore, the weights of elevation and relative height features can be increased. In broken ice areas, the ice surface itself is broken and complex, with large elevation fluctuations. Therefore, the weights of elevation and relative height features can be reduced. Weight adjustment based on positional relationship features: When the ice ridge area is located within the ship's core navigation channel, it poses the greatest threat to navigation safety. Therefore, the weights of positional relationship features can be increased.

[0052] Step A2: Determine the confidence level of the ice ridge region based on the target feature weights corresponding to each ice ridge feature and the scoring results of each ice ridge feature; determine the target ice ridge and the confidence level of the target ice ridge based on the ice ridge confidence level and the preset confidence level.

[0053] Ice ridge confidence is a quantitative indicator that comprehensively reflects the probability that an ice ridge region is a real ice ridge. After determining the target feature weights and scores for each ice ridge feature, the score for each feature is multiplied by its corresponding target feature weight, and the sum of these multiplications yields the ice ridge confidence. In other words, ice ridge confidence equals the sum of the scores for each feature multiplied by their corresponding target feature weights.

[0054] The preset reliability includes multiple thresholds used to classify the confidence level of ice ridges into different levels, thereby determining the ice ridge identification result. For example, the preset reliability includes a first preset reliability, a second preset reliability, and a third preset reliability, wherein the first preset reliability is the highest and the third preset reliability is the lowest.

[0055] When the confidence level of an ice ridge region is greater than or equal to the first preset confidence level, the ice ridge region is identified as a target ice ridge, and its confidence level is determined as the confidence level of the target ice ridge. When the confidence level of an ice ridge is between the preset confidence level and the first preset confidence level, the ice ridge region is identified as a target ice ridge. When the confidence level of an ice ridge is between the third preset confidence level and the second preset confidence level, the ice ridge region is identified as a target ice ridge. When the confidence level of an ice ridge region is lower than the third preset confidence level, the ice ridge region is excluded. Further, all target ice ridges and their confidence levels are obtained.

[0056] The basic weights are dynamically adjusted based on real-time scene factors such as observation distance, ice surface type, and positional relationships, enabling feature fusion to adapt to different navigation environments and ice conditions, thus avoiding recognition biases caused by fixed weights in complex scenarios. By integrating multi-dimensional features into a single ice ridge confidence score through weighted summation, a quantitative assessment of whether an ice ridge region is a real ice ridge is achieved, improving the accuracy and stability of ice ridge identification during ship navigation.

[0057] The ice ridge identification results also include early warning information. The ice ridge identification results are determined based on the identified ice ridge features, including: obtaining the ship speed and determining the ship's safe range based on the ship speed; determining whether the target ice ridge is within the ship's safe range based on the various ice ridge features of the target ice ridge; and generating early warning information based on the various ice ridge features of the target ice ridge when the target ice ridge is within the ship's safe range.

[0058] The ship's safe zone is a dynamic spatial area determined based on the ship's current position and forward heading, used to determine whether a target ice ridge poses a real threat to navigation. The ship's braking distance can be determined based on its speed, and the longitudinal length of the safe zone can be determined based on the braking distance. After identifying the target ice ridge, its precise position in the ship's forward observation coordinate system is determined in real time based on the target ice ridge's 3D point cloud data. The precise position includes longitudinal distance, lateral offset, and relative azimuth. The longitudinal distance is the distance from the target ice ridge's centroid to the ship's current position along the heading, reflecting the distance of the ice ridge from the ship. The lateral offset is the perpendicular distance from the target ice ridge's centroid to the ship's centerline, reflecting the lateral deviation of the ice ridge from the predetermined route. The relative azimuth is the angle between the target ice ridge and the ship's bow direction, reflecting the ice ridge's orientation in the ship's field of vision. Furthermore, based on the precise position of the target ice ridge in the ship's forward observation coordinate system, it is determined whether the target ice ridge is within the ship's safe zone. When the target ice ridge is within the safe zone, a warning message is generated based on the various ice ridge characteristics and displayed to the user.

[0059] By combining the dynamic determination of the safe range based on ship speed and generating graded early warning information, the identification strategy can be automatically adjusted according to the actual ice conditions and environmental conditions, effectively ensuring the navigation safety of ships in complex polar ice areas.

[0060] The technical solution of this embodiment involves acquiring three-dimensional point cloud data of the ice surface; performing terrain processing on the ice surface to obtain corresponding grids, and determining elevation reference values ​​based on the three-dimensional point cloud data of all grids; identifying elevation anomaly regions in the ice surface based on the three-dimensional point cloud data and elevation reference values ​​of each grid; performing ice ridge region feature analysis on the elevation anomaly regions based on the three-dimensional point cloud data of the elevation anomaly regions, and determining ice ridge regions based on the analyzed ice ridge region features; ice ridge region features include spatial continuity, region size, region area, point cloud density, edge variation, or morphological extension direction; identifying ice ridge features in the ice ridge regions, and determining the ice ridge identification result based on the identified ice ridge features; ice ridge features include ice ridge size features, slope variation features, elevation features, or positional relationship features. The technical solution of this embodiment dynamically determines elevation reference values ​​through real-time acquired three-dimensional point cloud data, and then accurately identifies elevation anomaly regions through dynamic elevation reference values. Furthermore, it accurately excludes irrelevant regions based on the spatial continuity, region size, and edge morphology of the elevation anomaly regions, and filters out ice ridge regions that may contain ice ridges. For ice ridge regions, more accurate ice ridge identification is achieved based on multi-dimensional ice ridge features, realizing adaptive identification from raw point cloud data to target ice ridges. Ice ridges can be accurately identified even when ice surface colors are similar and shadows are not obvious, significantly improving the accuracy, objectivity, and stability of ice ridge identification during ship navigation and effectively ensuring the safety of ship navigation in complex polar ice areas.

[0061] Figure 2 This is a schematic diagram of the structure of an ice ridge identification device provided in an embodiment of the present invention. This device is suitable for executing the ice ridge identification method provided in this embodiment of the present invention. Figure 2 As shown, the device may specifically include:

[0062] Data acquisition module 201 is used to acquire three-dimensional point cloud data of the ice surface; The reference value determination module 202 is used to perform terrain processing on the ice surface to obtain each grid corresponding to the ice surface, and determine the elevation reference value based on the three-dimensional point cloud data of all grids. The first region determination module 203 is used to determine the elevation anomaly region in the ice surface based on the three-dimensional point cloud data of each grid and the elevation reference value. The second region determination module 204 is used to perform ridge region feature analysis on the elevation anomaly region based on the three-dimensional point cloud data of the elevation anomaly region, and determine the ridge region according to the analyzed ridge region features; the ridge region features include spatial continuity, region size, region area, point cloud density, edge variation or morphological extension direction; The ice ridge identification module 205 is used to identify ice ridge features in the ice ridge region and determine the ice ridge identification result based on the identified ice ridge features; the ice ridge features include ice ridge size features, slope change features, elevation features, or positional relationship features.

[0063] Optionally, the reference value determination module 202 is specifically used to: perform scene feature analysis on the ice surface based on the three-dimensional point cloud data of all grids, and determine the data scene in which the ice surface is located; the scene features include ice surface undulation features, ice surface tilt features, ship attitude features, observation distance and point cloud density features; the data scene includes sparse ice ridge scene, moderate ice ridge scene, first dense ice ridge scene and second dense ice ridge scene; The elevation reference value is calculated based on the elevation reference value calculation method corresponding to the data scenario.

[0064] Optionally, the reference value determination module 202 is further configured to: when the data scenario is the sparse ice ridge scenario, calculate the average elevation of all grids based on the three-dimensional point cloud data of all grids, and determine the average elevation as the elevation reference value; When the data scenario is the ice ridge suitable scenario, the median elevation of all grids is calculated based on the three-dimensional point cloud data of all grids, and the median elevation is determined as the elevation reference value. When the data scenario is the first dense ice ridge scenario, the low quantile elevation of all grids is calculated based on the three-dimensional point cloud data of all grids, and the low quantile elevation is determined as the elevation reference value. When the data scenario is the second dense ice ridge scenario, the fitting plane of each grid is determined based on the three-dimensional point cloud data of each grid, and the reference elevation of each grid is determined based on the fitting plane of each grid, and the reference elevation of each grid is determined as the elevation reference value.

[0065] Optionally, the first region determination module 203 is specifically used to: calculate the elevation difference of each grid based on the three-dimensional point cloud data of each grid and the elevation reference value; Adjustment items are determined based on the data scenario, and a pre-determined benchmark threshold is adjusted based on the adjustment items to obtain the target threshold; the adjustment items include observation distance adjustment item, point cloud quality adjustment item, ship attitude adjustment item, and ice surface roughness adjustment item; Each elevation anomaly grid is determined based on the elevation difference of each grid and the target threshold, and the elevation anomaly region is determined based on all elevation anomaly grids.

[0066] Optionally, the ice ridge size features include relative height features, width features, and continuous length features; the elevation features are used to represent the elevation difference between the ice ridge region and the flat ice surface; the positional relationship features are used to represent the positional relationship of the ice ridge region relative to the ship track; the ice ridge features also include point cloud density features; the ice ridge identification result includes the target ice ridge and the confidence level of the target ice ridge.

[0067] Optionally, the ice ridge recognition module 205 is specifically used to: calculate the scoring results of each ice ridge feature based on each ice ridge feature and its corresponding scoring criteria; Based on the predetermined observation distance, ice surface type, and the positional relationship features, the basic feature weights corresponding to each preset ice ridge feature are adjusted to obtain the target feature weights corresponding to each ice ridge feature. The confidence level of the ice ridge region is determined based on the target feature weights corresponding to each ice ridge feature and the scoring results of each ice ridge feature. The target ice ridge and its confidence level are determined based on the ice ridge confidence level and the preset confidence level.

[0068] Optionally, the ice ridge identification module 205, the ice ridge identification result further includes warning information, and is also used to: obtain the ship speed, and determine the ship's safe range based on the ship speed; Based on the various ice ridge characteristics of the target ice ridge, determine whether the target ice ridge is within the safe range of the ship; When the target ice ridge is within the safe range of the ship, the warning information is generated based on various ice ridge characteristics of the target ice ridge.

[0069] The ice ridge identification device provided in this embodiment of the invention can execute the ice ridge identification method provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the method. Contents not described in detail in this embodiment can be referred to the descriptions in any method embodiment of the invention.

[0070] Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention, with reference to... Figure 3 , Figure 3 The electronic device 12 shown is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of this application. Figure 3 As shown, the electronic device 12 is represented in the form of a general-purpose computing device. The components of the electronic device 12 may include, but are not limited to: one or more processors or processing units 16, system memory 28, and bus 18 connecting different system components (including system memory 28 and processing unit 16).

[0071] Bus 18 represents one or more of several bus architectures, including a memory bus or memory controller, a peripheral bus, a graphics acceleration port, a processor, or a local bus using any of the various bus architectures. For example, these architectures include, but are not limited to, the Industry Standard Architecture (ISA) bus, the Micro Channel Architecture (MAC) bus, the Enhanced ISA bus, the Video Electronics Standards Association (VESA) local bus, and the Peripheral Component Interconnect (PCI) bus.

[0072] Electronic device 12 typically includes a variety of computer system readable media. These media can be any available media that can be accessed by electronic device 12, including volatile and non-volatile media, removable and non-removable media.

[0073] System memory 28 may include computer system readable media in the form of volatile memory, such as random access memory (RAM) 30 and / or cache memory 32. Electronic device 12 may further include other removable / non-removable, volatile / non-volatile computer system storage media. By way of example only, storage system 34 may be used to read and write non-removable, non-volatile magnetic media (… Figure 3 Not shown; usually referred to as a "hard drive"). Although Figure 3 As not shown, a disk drive for reading and writing to a removable non-volatile disk (e.g., a "floppy disk") and an optical disk drive for reading and writing to a removable non-volatile optical disk (e.g., a CD-ROM, DVD-ROM, or other optical media) may be provided. In these cases, each drive may be connected to bus 18 via one or more data media interfaces. System memory 28 may include at least one program product having a set (e.g., at least one) of program modules configured to perform the functions of the embodiments of this application.

[0074] A program / utility 40 having a set (at least one) of program modules 46 may be stored, for example, in system memory 28. Such program modules 46 include, but are not limited to, an operating system, one or more application programs, other program modules, and program data. Each or some combination of these examples may include an implementation of a network environment. Program modules 46 typically perform the functions and / or methods described in the embodiments of this application.

[0075] Electronic device 12 can also communicate with one or more external devices 14 (e.g., keyboard, pointing device, display 24, etc.), and with one or more devices that enable a user to interact with electronic device 12, and / or with any device that enables electronic device 12 to communicate with one or more other computing devices (e.g., network card, modem, etc.). This communication can be performed via input / output (I / O) interface 22. Furthermore, electronic device 12 can also communicate with one or more networks (e.g., local area network (LAN), wide area network (WAN), and / or public networks, such as the Internet) via network adapter 20. As shown, network adapter 20 communicates with other modules of electronic device 12 via bus 18. It should be understood that, although... Figure 3 As not shown, other hardware and / or software modules may be used in conjunction with electronic device 12, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0076] The processing unit 16 executes various functional applications and data processing by running programs stored in the system memory 28, such as implementing an ice ridge identification method provided in this embodiment of the invention: acquiring three-dimensional point cloud data of the ice surface; performing terrain processing on the ice surface to obtain each grid corresponding to the ice surface, and determining elevation reference values ​​based on the three-dimensional point cloud data of all grids; determining elevation anomaly regions in the ice surface based on the three-dimensional point cloud data of each grid and the elevation reference values; performing ice ridge region feature analysis on the elevation anomaly regions based on the three-dimensional point cloud data of the elevation anomaly regions, and determining ice ridge regions based on the analyzed ice ridge region features; the ice ridge region features include spatial continuity, region size, region area, point cloud density, edge changes or morphological extension direction; performing ice ridge feature identification on the ice ridge regions, and determining the ice ridge identification result based on the identified ice ridge features; the ice ridge features include ice ridge size features, slope change features, elevation features or positional relationship features.

[0077] This invention provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements an ice ridge identification method as provided in all embodiments of this invention: acquiring three-dimensional point cloud data of an ice surface; performing terrain processing on the ice surface to obtain grids corresponding to the ice surface, and determining elevation reference values ​​based on the three-dimensional point cloud data of all grids; determining elevation anomaly regions in the ice surface based on the three-dimensional point cloud data of each grid and the elevation reference values; performing ice ridge region feature analysis on the elevation anomaly regions based on the three-dimensional point cloud data of the elevation anomaly regions, and determining ice ridge regions based on the analyzed ice ridge region features; the ice ridge region features include spatial continuity, region size, region area, point cloud density, edge variation, or morphological extension direction; performing ice ridge feature identification on the ice ridge regions, and determining the ice ridge identification result based on the identified ice ridge features; the ice ridge features include ice ridge size features, slope variation features, elevation features, or positional relationship features. The computer-readable medium can be a computer-readable signal medium or a computer-readable storage medium. Computer-readable storage media can be, for example, but not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor electronic devices, apparatuses, or any combination thereof. More specific examples (a non-exhaustive list) of computer-readable storage media include: electrical connections having one or more wires; portable computer disks; hard disks; random access memory (RAM); read-only memory (ROM); erasable programmable read-only memory (EPROM or flash memory); optical fiber; portable compact disk read-only memory (CD-ROM); optical storage devices; magnetic storage devices; or any suitable combination thereof. In this document, a computer-readable storage medium can be any tangible medium that contains or stores a program that can be used by or in conjunction with an electronic device, apparatus, or apparatus that can be executed.

[0078] Computer-readable signal media may include data signals propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. Computer-readable signal media may also be any computer-readable medium other than computer-readable storage media, capable of sending, propagating, or transmitting programs for use by or in conjunction with an electronic device, apparatus, or device that executes instructions.

[0079] Program code contained on a computer-readable medium may be transmitted using any suitable medium, including but not limited to wireless, wire, optical fiber, RF, etc., or any suitable combination thereof.

[0080] Computer program code for performing the operations of this invention can be written in one or more programming languages ​​or a combination thereof. Programming languages ​​include object-oriented programming languages ​​such as Java, Smalltalk, and C++, as well as conventional procedural programming languages ​​such as C or similar languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0081] Note that the above description is merely a preferred embodiment of the present invention and the technical principles employed. Those skilled in the art will understand that the present invention is not limited to the specific embodiments described herein, and various obvious changes, readjustments, and substitutions can be made without departing from the scope of protection of the present invention. Therefore, although the present invention has been described in detail through the above embodiments, the present invention is not limited to the above embodiments, and may include many other equivalent embodiments without departing from the concept of the present invention, the scope of which is determined by the scope of the appended claims.

Claims

1. A method for identifying ice ridges, characterized in that, The method includes: Acquire 3D point cloud data of the ice surface; The ice surface is subjected to topographic processing to obtain each grid corresponding to the ice surface, and the elevation reference value is determined based on the three-dimensional point cloud data of all grids. The elevation anomaly areas in the ice surface are determined based on the three-dimensional point cloud data of each grid and the elevation reference value. Based on the three-dimensional point cloud data of the elevation anomaly region, the ridge region features of the elevation anomaly region are analyzed, and the ridge region is determined according to the analyzed ridge region features. The ridge region features include spatial continuity, region size, region area, point cloud density, edge variation or morphological extension direction. The ice ridge region is subjected to ice ridge feature identification, and the ice ridge identification result is determined based on the identified ice ridge features; the ice ridge features include ice ridge size features, slope change features, elevation features, or positional relationship features.

2. The method according to claim 1, characterized in that, Elevation reference values ​​are determined based on the 3D point cloud data of all grids, including: Based on the 3D point cloud data of all grids, scene feature analysis is performed on the ice surface to determine the data scene in which the ice surface is located; the scene features include ice surface undulation features, ice surface tilt features, ship attitude features, observation distance and point cloud density features; the data scene includes sparse ice ridge scene, moderate ice ridge scene, first ice ridge dense scene and second ice ridge dense scene. The elevation reference value is calculated based on the elevation reference value calculation method corresponding to the data scenario.

3. The method according to claim 2, characterized in that, The elevation reference value is calculated based on the elevation reference value calculation method corresponding to the data scenario, including: When the data scenario is the sparse ice ridge scenario, the average elevation of all grids is calculated based on the three-dimensional point cloud data of all grids, and the average elevation is determined as the elevation reference value. When the data scenario is the ice ridge suitable scenario, the median elevation of all grids is calculated based on the three-dimensional point cloud data of all grids, and the median elevation is determined as the elevation reference value. When the data scenario is the first dense ice ridge scenario, the low quantile elevation of all grids is calculated based on the three-dimensional point cloud data of all grids, and the low quantile elevation is determined as the elevation reference value. When the data scenario is the second dense ice ridge scenario, the fitting plane of each grid is determined based on the three-dimensional point cloud data of each grid, and the reference elevation of each grid is determined based on the fitting plane of each grid, and the reference elevation of each grid is determined as the elevation reference value.

4. The method according to claim 3, characterized in that, Based on the 3D point cloud data of each grid and the elevation reference value, the elevation anomaly areas in the ice surface are determined, including: The elevation difference of each grid is calculated based on the three-dimensional point cloud data of each grid and the elevation reference value; Adjustment items are determined based on the data scenario, and a pre-determined benchmark threshold is adjusted based on the adjustment items to obtain the target threshold; the adjustment items include observation distance adjustment item, point cloud quality adjustment item, ship attitude adjustment item, and ice surface roughness adjustment item; Each elevation anomaly grid is determined based on the elevation difference of each grid and the target threshold, and the elevation anomaly region is determined based on all elevation anomaly grids.

5. The method according to claim 1, characterized in that, The ice ridge size features include relative height features, width features, and continuous length features; the elevation features are used to represent the elevation difference between the ice ridge region and the flat ice surface; the positional relationship features are used to represent the positional relationship of the ice ridge region relative to the ship track; the ice ridge features also include point cloud density features; the ice ridge identification result includes the target ice ridge and the confidence level of the target ice ridge.

6. The method according to claim 5, characterized in that, The ice ridge region is subjected to ice ridge feature identification, and the ice ridge identification result is determined based on the identified ice ridge features, including: The scoring results for each ice ridge feature are calculated based on the various ice ridge features and their corresponding scoring criteria. Based on the predetermined observation distance, ice surface type, and the positional relationship features, the basic feature weights corresponding to each preset ice ridge feature are adjusted to obtain the target feature weights corresponding to each ice ridge feature. The confidence level of the ice ridge region is determined based on the target feature weights corresponding to each ice ridge feature and the scoring results of each ice ridge feature. The target ice ridge and its confidence level are determined based on the ice ridge confidence level and the preset confidence level.

7. The method according to claim 6, characterized in that, The ice ridge identification results also include early warning information. The ice ridge identification results are determined based on the identified ice ridge features, including: Obtain the ship's speed and determine the ship's safe range based on the ship's speed; Based on the various ice ridge characteristics of the target ice ridge, determine whether the target ice ridge is within the safe range of the ship; When the target ice ridge is within the safe range of the ship, the warning information is generated based on various ice ridge characteristics of the target ice ridge.

8. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the ice ridge identification method as described in any one of claims 1-7.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the ice ridge identification method as described in any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the ice ridge identification method as described in any one of claims 1 to 7.

Images