Intertidal zone shrub feature extraction method and device based on point cloud, equipment and medium

Abstract

The invention discloses an intertidal zone shrub feature extraction method and device based on point cloud, equipment and a medium, and relates to the technical field of point cloud data processing. The method comprises the following steps: acquiring initial laser point cloud data and initial underwater topographic data of an intertidal zone in different time periods; performing data fusion processing on the initial laser point cloud data and the initial underwater terrain data to obtain all-terrain fusion point cloud data; extracting ground points from the fused point cloud data based on a progressive triangulation network encryption filtering mode to obtain ground points and non-ground points; and performing shrub feature extraction based on the ground points and the non-ground points to obtain shrub feature information. According to the method, the shrub feature information of the tropical intertidal zone can be automatically, efficiently and accurately extracted based on the point cloud data, the labor cost is reduced, the measurement efficiency of the intertidal zone is improved, and better calculation precision is kept.

Description

Technical Field

[0001] This application relates to the field of point cloud data processing technology, and in particular to a method, apparatus, equipment and medium for feature extraction of intertidal shrubs based on point clouds. Background Technology

[0002] The intertidal zone, being part of the coastal zone, is easily affected by tides, resulting in poor measurement conditions. Currently, intertidal zone mapping techniques mainly include manual field measurements, traditional optical image extraction methods, and airborne radar remote sensing tests. However, natural coastal zones often possess unique geographical environments such as reefs, mudflats, and wetlands, making traditional manual field measurements inefficient and unable to ensure rapid updates of basic geographic information for the intertidal zone; furthermore, there are operational risks that endanger personal safety, and labor costs are high. Additionally, when using traditional optical image extraction methods, shrubs, algae, and wetlands exhibit overlapping NDVI values ​​in multispectral images (e.g., shrubs NDVI 0.3-0.6 vs. algae NDVI 0.4-0.7), making effective differentiation impossible. Moreover, the complex topography of the intertidal zone, with micro-topographical undulations at the edges of tidal channels (such as shell ridges) causing abrupt elevation changes, can lead to misclassification as shrub clusters (increasing the false alarm rate by 15-20%). It is evident that due to the complex intertidal environment (e.g., tidal variations, sediment disturbance, and mixed vegetation), traditional optical image extraction methods are easily affected by lighting and shadows, resulting in low accuracy. Furthermore, traditional airborne radar data is affected by tides, resulting in significant point cloud noise, making ground point extraction difficult and thus hindering shrub extraction. Therefore, how to reduce labor costs and automatically, efficiently, and accurately extract shrub features from tropical intertidal regions is a pressing technical problem that needs to be solved. Summary of the Invention

[0003] This application aims to address at least one of the technical problems existing in the prior art. To this end, this application proposes a method, apparatus, device, and medium for extracting intertidal shrub features based on point cloud data. This method can automatically, efficiently, and accurately extract shrub feature information of tropical intertidal regions based on point cloud data, reducing labor costs, improving the measurement efficiency of intertidal regions, and maintaining good computational accuracy.

[0004] In a first aspect, embodiments of this application provide a method for feature extraction of intertidal shrubs based on point clouds, including:

[0005] Acquire initial laser point cloud data and initial underwater topographic data of the intertidal zone in different time periods;

[0006] The initial laser point cloud data and the initial underwater terrain data are fused to obtain fused point cloud data of the entire terrain.

[0007] Ground points are extracted from the fused point cloud data using a progressive triangulation encryption filtering method, resulting in ground points and non-ground points.

[0008] Shrub feature extraction is performed based on the ground points and non-ground points to obtain shrub feature information.

[0009] Secondly, embodiments of this application provide a point cloud-based intertidal shrub feature extraction device, comprising:

[0010] The data acquisition module is used to acquire initial laser point cloud data and initial underwater topographic data of the intertidal zone in different time periods;

[0011] The data fusion module is used to perform data fusion processing on the initial laser point cloud data and the initial underwater terrain data to obtain fused point cloud data of the entire terrain.

[0012] The ground point extraction module is used to extract ground points from the fused point cloud data based on a progressive triangulation encryption filtering method, thereby obtaining ground points and non-ground points.

[0013] The shrub feature extraction module is used to extract shrub features based on the ground points and the non-ground points to obtain shrub feature information.

[0014] Thirdly, embodiments of this application provide an 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 computer program to implement the point cloud-based intertidal shrub feature extraction method as described in the first aspect.

[0015] Fourthly, embodiments of this application provide a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the point cloud-based intertidal shrub feature extraction method as described in the first aspect.

[0016] This application embodiment includes the following steps in the process of extracting intertidal shrub features: First, initial laser point cloud data and initial underwater topographic data of the intertidal region are acquired in time intervals; automatic data acquisition improves the measurement efficiency of the intertidal region and reduces labor costs; second, the initial laser point cloud data and the initial underwater topographic data are fused to obtain fused point cloud data of the entire topography; more accurate and effective fused point cloud data is obtained based on multi-feature data fusion, providing a good data foundation for subsequent automatic and efficient extraction of ground points and non-ground points; next, ground points are extracted from the fused point cloud data based on a progressive triangulation filtering method to obtain ground points and non-ground points; automatic, efficient, and accurate extraction of ground points yields classified ground points and non-ground points, providing a good data foundation for subsequent shrub feature extraction and helping to maintain good computational accuracy; finally, shrub feature extraction is performed based on the ground points and non-ground points to obtain shrub feature information; thus, feature extraction can be completed automatically, efficiently, and accurately to obtain shrub feature information of the intertidal region. In other words, the solution of this application embodiment can automatically, efficiently and accurately extract shrub feature information of tropical intertidal regions based on point cloud data, reduce labor costs, improve the measurement efficiency of intertidal regions and maintain better calculation accuracy.

[0017] Other features and advantages of this application will be set forth in the following description and will be apparent in part from the description or may be learned by practicing the application. The objectives and other advantages of this application may be realized and obtained by means of the structures particularly pointed out in the description and the accompanying drawings. Attached Figure Description

[0018] The above and / or additional aspects and advantages of this application will become apparent and readily understood from the description of the embodiments taken in conjunction with the following drawings, in which:

[0019] Figure 1 This is a schematic diagram of the functional module structure of a point cloud-based intertidal shrub feature extraction device provided in one embodiment of this application;

[0020] Figure 2 This is a schematic flowchart of a point cloud-based intertidal shrub feature extraction method provided in one embodiment of this application;

[0021] Figure 3 This is a parameter diagram of a progressive triangular mesh filtering method provided in one embodiment of this application;

[0022] Figure 4 This is a schematic diagram illustrating the specific process of extracting ground points based on a progressive triangulation encryption filtering method according to an embodiment of this application;

[0023] Figure 5 This is provided in one embodiment of the present application. Figure 2 A flowchart illustrating the specific method of step S140;

[0024] Figure 6 This is a schematic diagram of the hardware structure of an electronic device provided in one embodiment of this application. Detailed Implementation

[0025] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments.

[0026] It should be understood that in the description of this application, the orientation descriptions, such as up, down, front, back, left, right, etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this application.

[0027] It should be noted that although a logical order is shown in the flowcharts in this application, in some cases, the steps shown or described may be performed in a different order than that shown in the flowcharts. In the description of this application, "several" means one or more, and "more" means two or more. The terms "first" and "second" are used only to distinguish technical features and should not be construed as indicating or implying relative importance, or implicitly indicating the number of technical features indicated, or implicitly indicating the order in which the technical features are indicated.

[0028] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.

[0029] This application provides a point cloud-based method, device, electronic device, and computer-readable storage medium for intertidal shrub feature extraction, relating to the field of point cloud data processing technology. The method includes: acquiring initial laser point cloud data and initial underwater topographic data of the intertidal region at different time intervals; performing data fusion processing on the initial laser point cloud data and initial underwater topographic data to obtain fused point cloud data of the entire topography; extracting ground points from the fused point cloud data using a progressive triangulation filtering method to obtain ground points and non-ground points; and extracting shrub features based on the ground points and non-ground points to obtain shrub feature information. This application can automatically, efficiently, and accurately extract shrub feature information of tropical intertidal regions based on point cloud data, reducing labor costs, improving the measurement efficiency of intertidal regions, and maintaining good computational accuracy.

[0030] The embodiments of this application will be further described below with reference to the accompanying drawings.

[0031] like Figure 1 As shown, Figure 1 This is a schematic diagram of the functional module structure of a point cloud-based intertidal shrub feature extraction device 100 provided in one embodiment of this application; the point cloud-based intertidal shrub feature extraction device 100 includes: a data acquisition module 110, a data fusion module 120, a ground point extraction module 130, and a shrub feature extraction module 140.

[0032] Specifically, the data acquisition module 110 is used to acquire initial laser point cloud data and initial underwater topographic data of the intertidal zone in different time periods.

[0033] Specifically, the data fusion module 120 is used to perform data fusion processing on the initial laser point cloud data and the initial underwater terrain data to obtain fused point cloud data of the entire terrain.

[0034] Specifically, the ground point extraction module 130 is used to extract ground points from the fused point cloud data based on the progressive triangulation encryption filtering method, so as to obtain ground points and non-ground points.

[0035] Specifically, the shrub feature extraction module 140 is used to extract shrub features based on ground points and non-ground points to obtain shrub feature information.

[0036] The point cloud-based intertidal shrub feature extraction device 100 of this application realizes the point cloud-based intertidal shrub feature extraction method of this application through the cooperation of the data acquisition module 110, the data fusion module 120, the ground point extraction module 130, and the shrub feature extraction module 140. It can automatically, efficiently, and accurately extract shrub feature information of tropical intertidal regions based on point cloud data, reduce labor costs, improve the measurement efficiency of intertidal regions, and maintain good calculation accuracy.

[0037] Those skilled in the art will understand that the device embodiments described above are merely illustrative, and the units described as separate components may or may not be physically separate; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0038] Those skilled in the art will understand that the device architecture and application scenarios described in the embodiments of this application are for the purpose of more clearly illustrating the technical solutions of the embodiments of this application, and do not constitute a limitation on the technical solutions provided in the embodiments of this application. Those skilled in the art will know that as device architectures evolve and new application scenarios emerge, the technical solutions provided in the embodiments of this application are also applicable to similar technical problems. Based on the above device structure, various embodiments of the point cloud-based intertidal shrub feature extraction method of this application are proposed below.

[0039] like Figure 2 As shown, this point cloud-based intertidal shrub feature extraction method can be applied to, for example... Figure 1 In the system framework shown, the point cloud-based intertidal shrub feature extraction method may include, but is not limited to, steps S110 to S140.

[0040] Step S110: Acquire initial laser point cloud data and initial underwater topographic data of the intertidal zone in different time periods.

[0041] Step S120: Perform data fusion processing on the initial laser point cloud data and the initial underwater terrain data to obtain fused point cloud data of the entire terrain.

[0042] Step S130: Extract ground points from the fused point cloud data using a progressive triangulation encryption filtering method to obtain ground points and non-ground points.

[0043] Step S140: Extract shrub features based on ground points and non-ground points to obtain shrub feature information.

[0044] Through steps S110 to S140, in the process of extracting intertidal shrub features, firstly, initial laser point cloud data and initial underwater topographic data of the intertidal region are acquired in different time periods; automatic data acquisition improves the measurement efficiency of the intertidal region and reduces labor costs; secondly, the initial laser point cloud data and initial underwater topographic data are fused to obtain fused point cloud data of the entire topography; based on multi-feature data fusion, more accurate and effective fused point cloud data is obtained, providing a good data foundation for subsequent automatic and efficient extraction of ground points and non-ground points; next, ground points are extracted from the fused point cloud data based on a progressive triangulation filtering method, resulting in ground points and non-ground points; automatic, efficient, and accurate extraction of ground points yields classified ground points and non-ground points, providing a good data foundation for subsequent shrub feature extraction and helping to maintain good computational accuracy; finally, shrub feature extraction is performed based on ground points and non-ground points to obtain shrub feature information; thus, feature extraction of shrub feature information of the intertidal region can be completed automatically, efficiently, and accurately. In other words, the solution of this application embodiment can automatically, efficiently and accurately extract shrub feature information of tropical intertidal regions based on point cloud data, reduce labor costs, improve the measurement efficiency of intertidal regions and maintain better calculation accuracy.

[0045] According to some embodiments of this application, step S110 is further described. Step S110: Acquire initial laser point cloud data and initial underwater topographic data of the intertidal zone in time periods, including but not limited to steps S111 to S112.

[0046] Step S111: During the low tide period, airborne lidar is used to collect data on the intertidal zone to obtain raw lidar point cloud data; the raw lidar point cloud data is filtered and denoised to obtain denoised initial lidar point cloud data.

[0047] Step S112: During high tide, a beam detector is used to collect data on the intertidal zone to obtain raw underwater topographic data; the raw underwater topographic data is filtered and denoised to obtain denoised initial underwater topographic data; wherein, the initial laser point cloud data overlaps with the initial underwater topographic data.

[0048] Specifically, airborne lidar is a new type of remote sensing data acquisition technology that integrates a laser rangefinder, a positioning and orientation system, and a digital camera. By measuring the round-trip time of the laser pulse and combining it with the positioning and orientation data provided by the POS system, it can directly obtain high-precision three-dimensional ground coordinates, i.e., three-dimensional laser point cloud data.

[0049] Specifically, the beam detector can be a single-wave or multi-wave unmanned surface vessel (USV). This application does not impose specific restrictions on the type of beam detector used.

[0050] It should be noted that the initial laser point cloud data overlaps with the initial underwater topographic data, meaning that the initial laser point cloud data and the initial underwater topographic data have a certain degree of overlap, in order to facilitate subsequent data fusion processing.

[0051] Understandably, traditional single airborne radar data is affected by tides, resulting in significant point cloud noise, which makes ground point extraction difficult and thus hinders shrub feature extraction. By using step S111, raw laser point cloud data is collected during low tide, reducing the impact of tides on the collected raw laser point cloud data and ensuring good quality of the collected raw laser point cloud data.

[0052] It is understandable that high tide may cause slight changes in the topography of the intertidal zone. By collecting raw underwater topographic data during high tide in step S112, highly reliable raw underwater topographic data can be collected efficiently, thereby improving the integrity of underwater topographic mapping.

[0053] Specifically, the collected raw laser point cloud data and raw underwater topography data still contain flying points and noise. Therefore, by filtering and denoising the raw laser point cloud data and raw underwater topography data respectively, obvious flying points and noise in the data can be removed, resulting in higher quality initial laser point cloud data; this is beneficial to improving the reliability of shrub feature extraction.

[0054] Through steps S111 to S112, different types of data from the intertidal zone can be automatically collected in different time periods, improving the measurement efficiency of the intertidal zone and reducing labor costs; thus laying a data foundation for subsequent processing.

[0055] According to some embodiments of this application, step S120 is further described. Step S120: Data fusion processing is performed on the initial laser point cloud data and the initial underwater terrain data to obtain fused point cloud data of the entire terrain, including but not limited to steps S121 to S123.

[0056] Step S121: Perform encryption processing on the initial underwater topographic data based on linear interpolation to obtain encrypted underwater topographic data; wherein, the data resolution of the encrypted underwater topographic data is the same as the data resolution of the initial laser point cloud data.

[0057] Step S122: Perform coordinate transformation and format conversion on the encrypted underwater terrain data to obtain target underwater terrain data; perform coordinate transformation and format conversion on the initial laser point cloud data to obtain target laser point cloud data; wherein, the target underwater terrain data and the target laser point cloud data have the same coordinates and format.

[0058] Step S123: Fuse the target underwater terrain data and the target laser point cloud data to obtain fused point cloud data.

[0059] Specifically, in step S121, linear interpolation refers to: in a set of known data, obtaining two adjacent known data points, determining a straight line and its equation based on the two known data points, calculating a data point to be inserted using the equation, and inserting the data point to be inserted between the two known data points to complete the data insertion.

[0060] Since different devices were used to collect the initial laser point cloud data and the initial underwater topographic data, the data resolutions of the two data are different, which is not conducive to data fusion. Therefore, this application uses the encryption processing based on linear interpolation in step S121 to make the data resolution of the initial underwater topographic data the same as that of the initial laser point cloud data, so as to facilitate data fusion.

[0061] In step S122, coordinate transformation and format conversion are performed on the initial laser point cloud data and the initial underwater terrain data respectively to obtain the target underwater terrain data and target laser point cloud data coordinates with the same format under the same coordinate system, so as to facilitate data fusion.

[0062] Step S123 yields integrated point cloud data that combines surface and underwater data, reducing errors caused by a single data source.

[0063] Through steps S121 to S123, more accurate and effective fused point cloud data is obtained based on multi-feature data fusion, providing a good data foundation for the subsequent automatic and efficient extraction of ground points and non-ground points.

[0064] Further explaining step S130, specifically, the progressive triangulation filtering method refers to the progressive encrypted triangulation filtering algorithm. This algorithm classifies points into ground points and non-ground points. It first generates a sparse initial triangulation using initial seed points, then iteratively encrypts the triangulation layer by layer until all ground points are classified.

[0065] According to some embodiments of this application, step S130 is further described. Step S130: extracting ground points from the fused point cloud data based on the progressive triangulation encryption filtering method to obtain ground points and non-ground points, including but not limited to steps S131 to S134.

[0066] Step S131: The fused point cloud data is gridded based on a preset grid size to obtain regular grid cells; wherein, the fused point cloud data includes data points to be classified; and each regular grid cell contains data points to be classified.

[0067] Step S132: Obtain the lowest terrain point in each regular grid cell and determine it as a set of initial seed points.

[0068] Step S133: Generate an initial triangular mesh based on a set of initial seed points.

[0069] Step S134: Perform iterative encryption processing based on the data points to be classified and the initial triangular network to divide the data points to be classified into ground points and non-ground points.

[0070] It is understandable that different grid sizes are used when performing gridding processing on different point cloud data. For example, in point cloud data containing buildings, the largest building size is measured as the preset grid size to perform gridding processing on the point cloud data, resulting in regular grid cells; while for point cloud data without buildings, the default value is used as the preset grid size to perform gridding processing, resulting in regular grid cells. Therefore, this application does not impose specific restrictions on the value of the preset grid size used in step S131.

[0071] This application completes the selection of the initial seed point through steps S131 to S132.

[0072] Step S133 generates a sparse initial triangular mesh based on the initial seed points, laying the foundation for subsequent iterative encryption processing.

[0073] Specifically, in combination Figure 3The iterative encryption process in step S134 is further explained as follows: First, for a discrete data point P to be classified, the target triangle into which the projection of the data point onto the horizontal plane falls is queried; wherein, the target triangle includes a first vertex V1, a second vertex V2, and a third vertex V3. Second, the vertical distance d between the data point and the plane determined by the target triangle is calculated, and the vertical distance d is determined as the iteration distance. Next, the first angle α1 between the line connecting the data point P and the first vertex V1 and the plane containing the target triangle is determined, the second angle α2 between the line connecting the data point P and the second vertex V2 and the plane containing the target triangle is determined, and the third angle α3 between the line connecting the data point P and the third vertex V3 and the plane containing the target triangle is determined. The largest angle among the first angle α1, the second angle α2, and the third angle α3 is determined as the iteration angle. Next, the iteration angle is compared with a preset angle threshold, and the iteration distance is compared with a preset distance threshold to determine if: the iteration angle is less than the preset angle threshold and the iteration distance is less than the preset distance threshold. If yes, the data point P to be classified is determined as a ground point; otherwise, the data point P to be classified is determined as a non-ground point. If it is a ground point, it is added to the current triangulation to obtain an updated triangulation. The next data point to be classified is visited, and the above steps are repeated based on the updated triangulation. This process is repeated iteratively, with the triangulation being encrypted in real time, until no new ground points are found. When all data points to be classified have been traversed, the classification of all data points is complete.

[0074] Through steps S131 to S134, classified ground points and non-ground points can be obtained automatically, efficiently and accurately, providing a good data foundation for subsequent shrub feature extraction and helping to maintain good computational accuracy.

[0075] Combination Figure 4 Here is an example illustrating the specific process of extracting ground points based on the progressive triangulation encryption filtering method.

[0076] Step S401: Select initial seed points based on the fused point cloud data.

[0077] Step S402: Construct a sparse initial triangular mesh based on the initial seed points.

[0078] Step S403: Traverse the current non-ground points and check whether the filtering distance and filtering angle between the data point and the triangle plane where its projection is located are both less than a preset threshold. If yes, proceed to step S404; otherwise, proceed to step S405.

[0079] Step S404: Mark as a new ground point.

[0080] Step S405: Maintain the marker as a non-ground point.

[0081] Step S406: Determine if the number of newly marked ground points is 0; if yes, proceed to step S408; if no, proceed to steps S407 and S403.

[0082] Step S407: Update the current triangulation based on the new ground points to obtain the updated triangulation.

[0083] Step S408: Output the classification results.

[0084] According to some embodiments of this application, in conjunction with Figure 5 Further explanation of step S140: Based on ground points and non-ground points, shrub features are extracted to obtain shrub feature information, including but not limited to steps S141 to S145.

[0085] Step S141: For each non-ground point, calculate the height difference between the non-ground point and the ground point, and calculate the roughness of the non-ground point based on the principal component analysis method.

[0086] Step S142: Perform spectral feature calculation on non-ground points to obtain the normalized green-red difference index of the point cloud.

[0087] Step S143: Calculate the vertical density distribution of the non-ground point cloud based on all non-ground points.

[0088] Step S144: Perform threshold segmentation based on height difference, roughness, point cloud normalized green-red difference index, vertical density distribution and preset threshold information to select shrub candidate points from non-ground points.

[0089] Step S145: Perform spatial clustering optimization on the shrub candidate points to obtain shrub feature information.

[0090] Through steps S141 to S145, feature extraction can be completed automatically, efficiently, and accurately to obtain shrub feature information in the intertidal zone.

[0091] Specifically, to further explain step S141, the formula for calculating the height difference between non-ground points and ground points is as follows:

[0092] ΔH = Z_point - Z_ground; where Z_point is the elevation value of the non-ground point; Z_ground is the elevation value of the ground point corresponding to the vertical projection of the non-ground point.

[0093] Specifically, Principal Component Analysis (PCA) is a statistical method primarily used for dimensionality reduction and data visualization. PCA effectively transforms high-dimensional datasets into low-dimensional spaces, reducing data complexity; it also identifies the main directions of change in the data, which (i.e., principal components) can be used as new features in subsequent analyses. In PCA, eigenvalue decomposition is used to identify the main directions of change in the data. Eigenvectors corresponding to larger eigenvalues ​​represent the main directions of data variation.

[0094] Specifically, step S141, the process of calculating the roughness of non-ground points based on principal component analysis (PCA) is as follows: A non-ground point is selected, and a neighborhood point cloud is selected based on a preset radius or the k-nearest neighbor method; the covariance matrix of the neighborhood point cloud is constructed, specifically including: calculating the mean of the points in the neighborhood point cloud; calculating the offset of each point relative to the mean; and constructing the covariance matrix based on the offsets. This covariance matrix has three eigenvalues ​​and three eigenvectors; the minimum eigenvalue of the PCA is determined as the roughness. In this embodiment, the roughness is the minimum eigenvalue of the local neighborhood PCA.

[0095] Step S141 calculates the height difference and roughness of the non-ground point cloud, laying the foundation for subsequent threshold segmentation.

[0096] According to some embodiments of this application, step S142 is further described. Step S142: spectral feature calculation processing is performed on non-ground points to obtain the normalized green-red difference index of the point cloud, including but not limited to steps S1421 to S1423.

[0097] Step S1421: Calculate the first average pixel value of the green channel and the second average pixel value of the red channel for non-ground points.

[0098] Step S1422: Subtract the second average pixel value from the first average pixel value to obtain the pixel difference value, and add the first average pixel value and the second average pixel value to obtain the pixel sum value.

[0099] Step S1423: Compare the pixel difference with the pixel sum to obtain the normalized green-red difference index of the point cloud.

[0100] The calculation formula used in steps S1421 to S1423 is: NGRDI=(GR) / (G+R); where NGRDI is the normalized green-red difference index of the point cloud; G is the first average pixel value of the green channel; and R is the second average pixel value of the red channel.

[0101] The normalized green-red difference index of the non-ground point cloud is calculated through steps S1421 to S1423, laying a foundation for subsequent threshold segmentation.

[0102] Further illustrate step S143. Step S143: Calculate the vertical density distribution of the non-ground point cloud based on all non-ground points. Specifically: In the three-dimensional space where the non-ground points are located, divide the space into several voxels (layers) along the Z-axis direction (vertical direction); count the ratio of the number of points contained in each voxel to the total number of non-ground points, and each ratio corresponds to the vertical density of each voxel; obtaining the vertical density of each voxel means obtaining the vertical density distribution. Through step S143 of this application, the vertical density distribution of the non-ground point cloud is calculated, laying a foundation for subsequent threshold segmentation.

[0103] Further illustrate step S144. Step S144: Perform threshold segmentation processing based on the height difference, roughness, normalized green-red difference index of the point cloud, vertical density distribution, and preset threshold information, and screen and determine shrub candidate points from the non-ground points, specifically including the following steps:

[0104] Obtain the first height threshold h1, the second height threshold h2 greater than the first height threshold h1, the roughness threshold w1, the index threshold w2, and the density distribution threshold interval (p1, p2) from the preset threshold information; configure each threshold segmentation condition. Among them, the first threshold segmentation condition is: h1 < ΔH < h2; the second threshold segmentation condition is: roughness > w1; the third threshold segmentation condition is: NGRDI > w2; the fourth threshold segmentation condition is: p1 < vertical density < p2; determine the non-ground points that simultaneously meet the first threshold segmentation condition, the second threshold segmentation condition, the third threshold segmentation condition, and the fourth threshold segmentation condition as shrub candidate points.

[0105] Specifically, the first height threshold h1 = 0.3m, the second height threshold h2 = 2m, the roughness threshold w1 = 0.1, and the index threshold w2 = 0.5. The first height threshold h1, the second height threshold h2 greater than the first height threshold h1, the roughness threshold w1, the index threshold w2, and the density distribution threshold interval (p1, p2) can be determined according to different seasons, seasons, and feature extraction requirements. This application does not limit the specific values of the preset threshold information.

[0106] According to some embodiments of this application, further illustrate step S145. Step S145: Perform spatial clustering optimization processing on the shrub candidate points to obtain shrub feature information, including but not limited to steps S1451 to S1453.

[0107] Step S1451: Perform Euclidean clustering on the shrub candidate points to obtain candidate clustering clusters and remove isolated points.

[0108] Step S1452: Count the number of cluster points for each candidate cluster.

[0109] Step S1453: The candidate clusters with more than the preset number of cluster points are identified as the target clusters, and the target clusters are identified as shrub feature information.

[0110] Specifically, Euclidean clustering is a clustering algorithm based on Euclidean distance. The process involves selecting one or more points as initial cluster centers, calculating the Euclidean distance between the initial cluster centers and other data points, and grouping data points whose Euclidean distance is less than a preset distance threshold into the same cluster. This process is repeated to gradually divide the data points into different clusters, and isolated points are removed. In this embodiment, step S1451 optimizes the clustering of shrub candidate points to obtain candidate clusters.

[0111] It is understandable that the preset point count threshold can be determined with reference to the point cloud density and the crown diameter of the shrubs. The value of the preset point count threshold can be configured according to actual needs. This application does not impose specific restrictions on the value of the preset point count threshold.

[0112] Steps S1451 to S1453 further optimize the shrub extraction results, yielding more accurate shrub feature information. This allows for automatic, efficient, and accurate feature extraction to obtain shrub feature information from the intertidal zone.

[0113] like Figure 6 As shown, the present invention also provides an electronic device, comprising:

[0114] The processor 601 can be implemented using a general-purpose central processing unit (CPU), microprocessor, application specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this application.

[0115] The memory 602 can be implemented as a read-only memory (ROM), static storage device, dynamic storage device, or random access memory (RAM). The memory 602 can store the operating system and other applications. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program code is stored in the memory 602 and is called and executed by the processor 601 to execute the point cloud-based intertidal shrub feature extraction method of this application embodiment.

[0116] The input / output interface 603 is used to implement information input and output;

[0117] The communication interface 604 is used to enable communication and interaction between this device and other devices. Communication can be achieved through wired means (such as USB, network cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.).

[0118] Bus 605 transmits information between various components of the device (e.g., processor 601, memory 602, input / output interface 603, and communication interface 604);

[0119] The processor 601, memory 602, input / output interface 603, and communication interface 604 are connected to each other within the device via bus 605.

[0120] This application embodiment also provides a storage medium, which is a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, it implements the above-described point cloud-based intertidal shrub feature extraction method.

[0121] Memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer-executable programs. Furthermore, memory may include high-speed random access memory, and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, memory may optionally include memory remotely located relative to the processor, and these remote memories can be connected to the processor via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof. The device embodiments described above are merely illustrative, and the units described as separate components may or may not be physically separate, and may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0122] It will be understood by those skilled in the art that all or some of the steps and systems in the methods disclosed above can be implemented as software, firmware, hardware, and suitable combinations thereof. Some or all of the physical components can be implemented as software executed by a processor, such as a central processing unit, digital signal processor, or microprocessor, or as hardware, or as an integrated circuit, such as an application-specific integrated circuit. Such software can be distributed on a computer-readable medium, which can include computer storage media (or non-transitory media) and communication media (or transient media). As is known to those skilled in the art, the term computer storage media includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, program modules, or other data). Computer storage media includes, but is not limited to, RAM, ROM, EEPROM, flash memory or other memory technologies, CD-ROM, digital versatile disc (DVD) or other optical disc storage, magnetic cartridges, magnetic tape, disk storage or other magnetic storage devices, or any other medium that can be used to store desired information and is accessible to a computer. Furthermore, as is known to those skilled in the art, communication media typically include computer-readable instructions, data structures, program modules, or other data in modulated data signals such as carrier waves or other transmission mechanisms, and may include any information delivery medium.

[0123] The above provides a detailed description of the preferred embodiments of this application. However, this application is not limited to the above-described embodiments. Those skilled in the art can make various equivalent modifications or substitutions without departing from the spirit of this application. All such equivalent modifications or substitutions are included within the scope defined by this application.

Claims

1. A method for feature extraction of intertidal shrubs based on point clouds, characterized in that, include: Acquire initial laser point cloud data and initial underwater topographic data of the intertidal zone in different time periods; The initial laser point cloud data and the initial underwater terrain data are fused to obtain fused point cloud data of the entire terrain. Ground points are extracted from the fused point cloud data using a progressive triangulation encryption filtering method, resulting in ground points and non-ground points. Shrub feature information is obtained by extracting shrub features based on the ground points and non-ground points.

2. The method for feature extraction of intertidal shrubs based on point clouds according to claim 1, characterized in that, The acquisition of initial laser point cloud data and initial underwater topographic data of the intertidal zone in time-segmented manner includes: During periods of low tide, airborne lidar is used to collect data on the intertidal zone to obtain raw lidar point cloud data; the raw lidar point cloud data is then filtered and denoised to obtain denoised initial lidar point cloud data. During high tide, a beam detector is used to collect data on the intertidal zone to obtain raw underwater topographic data; the raw underwater topographic data is then filtered and denoised to obtain denoised initial underwater topographic data; wherein the initial laser point cloud data overlaps with the initial underwater topographic data.

3. The method for feature extraction of intertidal shrubs based on point clouds according to claim 1, characterized in that, The process of fusing the initial laser point cloud data and the initial underwater topographic data to obtain fused point cloud data of the entire topography includes: The initial underwater topographic data is encrypted using linear interpolation to obtain encrypted underwater topographic data; wherein the data resolution of the encrypted underwater topographic data is the same as that of the initial laser point cloud data. The encrypted underwater terrain data is subjected to coordinate transformation and format conversion to obtain target underwater terrain data; the initial laser point cloud data is subjected to coordinate transformation and format conversion to obtain target laser point cloud data; wherein, the target underwater terrain data and the target laser point cloud data have the same coordinates and format. The underwater terrain data and laser point cloud data of the target are fused to obtain fused point cloud data.

4. The method for feature extraction of intertidal shrubs based on point clouds according to claim 1, characterized in that, The step of extracting ground points from the fused point cloud data using a progressive triangulation encryption filtering method to obtain ground points and non-ground points includes: The fused point cloud data is gridded based on a preset grid size to obtain regular grid cells; wherein, the fused point cloud data includes data points to be classified; and each regular grid cell contains data points to be classified. Obtain the lowest terrain point in each of the aforementioned regular grid cells and determine it as a set of initial seed points; Generate an initial triangular mesh based on a set of initial seed points; Based on the data points to be classified and the initial triangulation, iterative encryption processing is performed to divide the data points to be classified into ground points and non-ground points.

5. The method for feature extraction of intertidal shrubs based on point clouds according to claim 1, characterized in that, The step of extracting shrub features based on the ground points and the non-ground points to obtain shrub feature information includes: For each non-ground point, the height difference between the non-ground point and the ground point is calculated, and the roughness of the non-ground point is calculated based on principal component analysis. The spectral characteristics of the non-ground points are calculated to obtain the normalized green-red difference index of the point cloud. The vertical density distribution of the non-ground point cloud is calculated based on all the aforementioned non-ground points; Threshold segmentation is performed based on the height difference, the roughness, the normalized green-red difference index of the point cloud, the vertical density distribution, and the preset threshold information to screen and determine shrub candidate points from the non-ground points. Spatial clustering optimization is performed on the candidate shrub points to obtain the shrub feature information.

6. The method for feature extraction of intertidal shrubs based on point clouds according to claim 5, characterized in that, The step of performing spectral feature calculation on the non-ground points to obtain the normalized green-red difference index of the point cloud includes: Calculate the first average pixel value of the green channel and the second average pixel value of the red channel for the non-ground points; Subtracting the second average pixel value from the first average pixel value yields the pixel difference value; adding the first average pixel value to the second average pixel value yields the pixel sum value. The point cloud normalized green-red difference index is obtained by comparing the pixel difference with the pixel sum.

7. The method for feature extraction of intertidal shrubs based on point clouds according to claim 5, characterized in that, The spatial clustering optimization process performed on the candidate shrub points to obtain the shrub feature information includes: Euclidean clustering is performed on the candidate shrub points to obtain candidate clusters, and isolated points are removed; Count the number of cluster points in each of the candidate clusters; Candidate clusters with a number of cluster points greater than a preset threshold are identified as target clusters, and the target clusters are identified as the shrub feature information.

8. A device for extracting features of intertidal shrubs based on point clouds, characterized in that, include: The data acquisition module is used to acquire initial laser point cloud data and initial underwater topographic data of the intertidal zone in different time periods; The data fusion module is used to perform data fusion processing on the initial laser point cloud data and the initial underwater terrain data to obtain fused point cloud data of the entire terrain. The ground point extraction module is used to extract ground points from the fused point cloud data based on a progressive triangulation encryption filtering method, thereby obtaining ground points and non-ground points; The shrub feature extraction module is used to extract shrub features based on the ground points and the non-ground points to obtain shrub feature information.

9. An electronic device, characterized in that, It includes at least one processor and a memory for communicatively connecting to the at least one processor; the memory stores instructions executable by the at least one processor, which, when executed by the at least one processor, enable the at least one processor to perform the point cloud-based intertidal shrub feature extraction method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions for causing a computer to perform the point cloud-based intertidal shrub feature extraction method as described in any one of claims 1 to 7.

Images