A spaceborne single-photon lidar sounding signal extraction method, device and equipment

Abstract

The application provides a kind of space-borne single-photon lidar sounding signal extraction method, device and equipment, relating to photon extraction technical field, wherein the space-borne single-photon lidar sounding signal extraction method includes: obtaining original photons, and extracting water surface characteristic photons from the original photons according to the elevation distribution and density characteristics of water surface photons;Through photon elevation iterative clustering segmentation and local density filtering, water surface characteristic photons are separated and extracted, and water surface signal photons are obtained;Through multi-scale robust regression and multi-stage filtering strategy, water bottom signal photons are extracted;Determine the sounding signal based on the water surface signal photons and the water bottom signal photons, and tide correction is carried out on the sounding signal. Through the application, the problem of poor photon extraction quality in the prior art can be effectively solved, providing an effective new way for accurate, automatic and low-cost acquisition of deep-sea or difficult-to-approach shallow water area water depth information.

Description

Technical Field

[0001] This invention relates to the field of photon extraction technology, and in particular to a method, apparatus and equipment for extracting depth sounding signals from a spaceborne single-photon lidar. Background Technology

[0002] Traditional underwater topographic surveying methods mainly include shipborne sonar bathymetry and airborne lidar (Light Detection and Ranging, LiDAR) bathymetry. These methods offer high accuracy and a wide measurement range, but are expensive and limited by the accessibility of the platforms they are deployed on, preventing them from entering disputed sea areas or airspace. Spaceborne single-photon lidar remote sensing, with its advantages of wide spatial coverage, lack of geographical limitations, and low cost, is gradually becoming an important means of acquiring shallow water topography in distant or sensitive areas. However, due to the extremely high detection sensitivity of single-photon lidar, it is susceptible to combined interference from inherent system noise and environmental background noise during data acquisition. Therefore, establishing an effective lidar signal photon identification mechanism is a crucial step in data processing to obtain accurate and reliable lidar bathymetry data.

[0003] Based on the principle of photon extraction from lidar signals, existing methods can be mainly divided into two categories: density-based clustering methods and spatial distribution-based methods.

[0004] Density-based clustering methods, which focus on the density differences between signal photons and noise, have been widely applied to signal photon extraction in terrestrial and shallow water areas. Typical algorithms include Density-Based Spatial Clustering of Applications with Noise (DBSCAN) and Ordering Points To Identify the Clustering Structure (OPTICS). These methods offer advantages such as ease of operation and high computational efficiency, but their sensitivity to clustering parameters limits their applicability in complex marine environments. Because lidar photon signal intensity attenuates with water volume, the density of single-photon lidar bottom signal photons varies significantly, making it difficult to determine uniform and suitable density clustering parameters. This results in poor integrity and accuracy in signal photon extraction using density-based clustering methods. Methods based on spatial distribution characteristics analyze the spatial distribution features of photons to construct a hierarchical extraction method for surface-bottom signal photons, improving the performance of water body signal photon extraction. For surface signal photons, existing methods mainly rely on the consistency of the elevation distribution of surface photons for extraction. Due to factors such as wave undulation, the elevation distribution range of surface signal photons varies drastically. Methods based on global fitting of elevation statistical histograms are prone to incomplete or erroneous extraction of surface photons, especially in extremely shallow waters, which can easily lead to incorrect segmentation of surface and underwater photons. Compared to the relatively regular distribution of surface photons, underwater photon extraction faces a dual challenge: photon attenuation in the water medium causes a sharp decrease in density with depth; and underwater topographic undulations result in a discrete spatial distribution. Because lidar signals attenuate with water and underwater topography is complex, methods based on spatial distribution characteristics are difficult to accurately characterize the trend of underwater topographic changes, easily leading to serious missed detections of sparsely distributed and drastically fluctuating underwater signal photons.

[0005] There is currently no effective solution to the problem of poor photon extraction quality in existing related technologies. Summary of the Invention

[0006] This invention provides a method, apparatus, and equipment for extracting depth sounding signals from a spaceborne single-photon lidar, in order to overcome the shortcomings of poor photon extraction quality in existing related technologies.

[0007] In a first aspect, the present invention provides a method for extracting depth sounding signals from a spaceborne single-photon lidar, comprising: Used to acquire raw photons, and extract water surface feature photons from the raw photons based on the elevation distribution and density characteristics of the water surface photons; By using iterative clustering and segmentation based on photon elevation and local density filtering, feature photons on the water surface are separated and refined to obtain water surface signal photons. underwater signal photons are extracted using a multi-scale robust regression and multi-level filtering strategy. The depth sounding signal is determined based on the surface signal photons and the bottom signal photons, and the depth sounding signal is tidal corrected.

[0008] According to the present invention, a method for extracting depth sounding signals from a spaceborne single-photon lidar includes extracting water surface feature photons from the original photons based on the elevation distribution and density characteristics of water surface photons, comprising: The nuclear density of the original photons is determined, and the elevation range of the original photons is determined based on the nuclear density to initially extract the water surface photons; A kd tree is constructed based on the original photon, and the scale transformation coefficients of the original photon's orbital distance are determined. Based on the surface feature photons, the elevation range of global surface and underwater photons is determined, and the preliminary feature photons within the elevation range are used as feature photons. According to a method for extracting depth sounding signals from a spaceborne single-photon lidar provided by the present invention, the method determines the kernel density of the original photon and, based on the kernel density, determines the elevation range of the original photon, and initially extracts surface photons, including: Estimate the kernel density of the original photon elevation, and determine the extraction threshold based on the mean and standard deviation of the kernel density; Photons with a kernel density greater than the extraction threshold within the elevation range are retained, and the elevation range of the original photons is determined based on the extreme points of the kernel density estimation curve, thus initially extracting photons from the water surface.

[0009] According to the present invention, a method for extracting depth sounding signals from a spaceborne single-photon lidar is provided. This method separates and refines feature photons on the water surface through iterative clustering and segmentation of photon elevation and local density filtering to obtain water surface signal photons. The method includes: Based on the water surface feature photons, an initial lower bound for water surface feature photon segmentation is determined; The feature photons are iteratively clustered and segmented to determine the overall threshold for segmentation of the feature photons; The feature photons are segmented along the trajectory direction, and the feature photons in each segment are iteratively clustered and segmented to determine the elevation threshold for iterative clustering and segmentation under different conditions. Based on the elevation threshold, local surface characteristic photons, local underwater characteristic photons, and local underwater photons are obtained, and the surface photons that meet the range requirements are added to the local surface characteristic photons. The original photons that meet the neighborhood requirements are added to the local water surface feature photons to obtain local water surface signal photons; All the local surface signal photons are merged to obtain the surface signal photons, and the local underwater feature photons and the local underwater photons are merged to obtain the underwater feature photons and the underwater photons, respectively.

[0010] According to the present invention, a method for extracting depth sounding signals from a spaceborne single-photon lidar includes filtering water surface photons that meet the required range and adding them to local water surface feature photons, comprising: Determine the orbital distance and elevation range of the local water surface characteristic photons; The water surface photons distributed within the elevation range of the local water surface characteristic photons are added to the local water surface characteristic photons.

[0011] According to the present invention, a method for extracting depth sounding signals from a spaceborne single-photon lidar includes selecting original photons that meet neighborhood requirements and adding them to the local water surface feature photons to obtain local water surface signal photons, comprising: Determine the K-neighborhood average distance and standard deviation of the local water surface feature photons; The neighborhood requirement is obtained based on the K-neighbor average distance and standard deviation of the local water surface feature photons. The original photons that meet the neighborhood requirements are added to the local water surface feature photons to obtain local water surface signal photons.

[0012] The present invention provides a method for extracting depth sounding signals from a spaceborne single-photon lidar, which extracts underwater signal photons through a multi-scale robust regression and multi-level filtering strategy, including: Several sliding windows of different sizes are set up to detect underwater photons and identify candidate underwater signal photons; Using the largest window for sliding window detection, the range of the sliding window along the track direction is recorded to define the distribution range of underwater signal photons; Photons within each sliding window are sorted according to their distance along the track. The elevation difference between each photon and its adjacent photons is calculated sequentially, and noisy photons are removed based on the elevation difference. For photons sorted by distance along the track, the elevation difference between adjacent photons is recorded, and noisy photons with small-scale abrupt changes are iteratively identified and eliminated until there are no noisy photons. For photons sorted by orbital distance, the orbital distances of adjacent photons are recorded, and noisy photons are identified from the candidate underwater signal photons until no noise photons are found. A second multi-scale sliding window robust regression is performed on the denoised underwater photons to determine the underwater signal photons; the underwater signal photons and underwater feature photons in different sliding windows are combined to form the underwater signal photon set.

[0013] According to the present invention, a method for extracting depth sounding signals from a spaceborne single-photon lidar is provided, which involves setting up several sliding windows of different scales to detect underwater photons and determine candidate underwater signal photons, including: Several sliding windows of different sizes are set up to detect underwater photons in a specific step size; Robust regression is performed based on underwater photons within a sliding window to determine the overall mean absolute error and the single-point absolute error. The underwater photons are screened based on the overall average absolute error and the single-point absolute error to determine candidate underwater signal photons.

[0014] Secondly, the present invention also provides a spaceborne single-photon lidar depth sounding signal extraction device, comprising: The preliminary detection module is used to acquire raw photons and extract water surface feature photons from the raw photons based on the elevation distribution and density characteristics of the water surface photons. The water surface detection module is used to separate and extract water surface feature photons by using photon elevation iterative clustering and local density filtering to obtain water surface signal photons; The underwater detection module is used to extract underwater signal photons through multi-scale robust regression and multi-level filtering strategies; An integrated correction module is used to determine the depth sounding signal based on the surface signal photons and the bottom signal photons, and to perform tidal correction on the depth sounding signal.

[0015] Thirdly, the present invention also provides 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 program to implement the spaceborne single-photon lidar depth sounding signal extraction method as described in the first aspect above.

[0016] Fourthly, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the spaceborne single-photon lidar depth sounding signal extraction method as described in the first aspect above.

[0017] Fifthly, the present invention also provides a computer program product, including a computer program that, when executed by a processor, implements the satellite-borne single-photon lidar depth sounding signal extraction method as described in the first aspect above.

[0018] Compared with the prior art, the present invention has the following beneficial effects: The spaceborne single-photon lidar depth sounding signal extraction method provided by this invention comprehensively utilizes the elevation distribution and density characteristics of water surface photons, ensuring the accuracy and completeness of water surface photon extraction and overcoming the influence of factors such as water surface wave undulations. This method also utilizes the multi-scale characteristics of the orderly distribution of underwater photons with topography, employing a multi-level filtering underwater signal photon detection method to accurately identify underwater signal photons, overcoming the shortcomings of existing related technologies in underwater photon detection performance under conditions of significant photon signal attenuation and rugged underwater terrain. This method effectively solves the problem of poor photon extraction quality in existing related technologies, providing an effective new approach for accurately, automatically, and cost-effectively acquiring water depth information in remote or inaccessible shallow water areas. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0020] Figure 1 This is a flowchart of the method for extracting depth sounding signals from a spaceborne single-photon lidar provided by the present invention; Figure 2 This is a schematic diagram of the depth sounding signal extraction process in an embodiment of the present invention; Figure 3 This is a schematic diagram of the raw photon data of the spaceborne single-photon lidar in Example 1 of the present invention; Figure 4 This is a schematic diagram of the raw photon data of the spaceborne single-photon lidar in Example 2 of the present invention; Figure 5 This is a schematic diagram of the signal photon extraction results from the water surface and bottom in Example 1 of the present invention; Figure 6 This is a schematic diagram of the signal photon extraction results from the water surface and bottom in Example 2 of the present invention; Figure 7 This is a schematic diagram of the water depth information extraction results in Example 1 of the present invention; Figure 8 This is a schematic diagram of the water depth information extraction results in Example 2 of the present invention; Figure 9 This is a structural block diagram of the spaceborne single-photon lidar depth sounding signal extraction device provided by the present invention; Figure 10 This is a schematic diagram of the structure of the electronic device provided by the present invention. Detailed Implementation

[0021] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0022] This invention provides a method for extracting depth sounding signals from a spaceborne single-photon lidar. Figure 1 This is a flowchart of the method for extracting depth sounding signals from a spaceborne single-photon lidar provided by the present invention, as follows: Figure 1 As shown, the method includes the following steps: Step S101: Obtain the original photons, and extract the water surface feature photons from the original photons based on the elevation distribution and density characteristics of the water surface photons; Step S102: Through iterative clustering and segmentation of photon elevation and local density filtering, the feature photons of the water surface are separated and refined to obtain the water surface signal photons. Step S103: Extract underwater signal photons through multi-scale robust regression and multi-level filtering strategies; Step S104: Determine the depth sounding signal based on the surface signal photons and the bottom signal photons, and perform tidal correction on the depth sounding signal.

[0023] In this method, firstly, raw photon data is acquired. Based on the elevation distribution and density characteristics of surface photons, surface feature photons are extracted from the raw photon data. Then, through iterative clustering of photon elevation and local density filtering, surface-to-subsurface photon separation and refined extraction are achieved, yielding surface signal photons, subsurface feature photons, and subsurface photons. Next, through multi-scale robust regression and multi-level filtering strategies, complete and accurate extraction of bottom signal photons is achieved. After extracting surface and bottom signal photons, dynamic atmospheric correction is applied to the surface signal photons, and the average elevation of the surface signal photons is calculated as the surface elevation. The corrected elevation of the bottom signal photons is calculated based on the refraction correction method proposed by Parrish. Finally, tidal correction is applied to the above bathymetry signals according to a global tidal model, converting the depth reference to the mean sea level. Through the above process, the accuracy and completeness of water surface photon elevation distribution and density characteristics are comprehensively utilized to ensure the accuracy and completeness of water surface photon extraction, overcoming the influence of factors such as water surface wave undulation. This method also utilizes the multi-scale characteristics of underwater photon distribution ordered with topography, employing a multi-level filtering underwater signal photon detection method to accurately identify underwater signal photons, overcoming the shortcomings of existing related technologies in underwater photon detection performance under conditions of significant photon signal attenuation and rugged underwater terrain. This method effectively solves the problem of poor photon extraction quality in existing related technologies, providing an effective new approach for accurately, automatically, and cost-effectively acquiring water depth information in remote or inaccessible shallow water areas.

[0024] Figure 2 This is a schematic diagram of the depth sounding signal extraction process in an embodiment of the present invention, as shown below. Figure 2 As shown, in some embodiments, step S101, extracting surface feature photons from the original photons based on the elevation distribution and density characteristics of surface photons, includes: determining the kernel density of the original photons and determining the elevation range of the original photons based on the kernel density, and initially extracting surface photons; constructing a kd tree based on the original photons and determining the scale transformation coefficient of the orbital distance of the original photons; determining the judgment threshold of feature photons based on the mean and standard deviation of the average distance of the K nearest neighbors of the surface photons, and extracting surface feature photons from the surface photons based on the judgment threshold, and extracting preliminary feature photons from the original photons; determining the elevation range of global surface and underwater photons based on the surface feature photons, and using the preliminary feature photons within the elevation range as feature photons.

[0025] In this embodiment, the kernel density of the original photons is determined, and the elevation range of the original photons is determined based on the kernel density. The preliminary extraction of water surface photons includes: estimating the kernel density of the elevation of the original photons, determining the extraction threshold based on the mean and standard deviation of the kernel density; retaining photons with kernel densities greater than the extraction threshold within the elevation range, and determining the elevation range of the original photons based on the extreme points of the kernel density estimation curve, thus preliminarily extracting water surface photons.

[0026] For example, Figure 3 and Figure 4 These are schematic diagrams of the raw photon data from spaceborne single-photon lidar in Examples 1 and 2, respectively. First, the kernel density is estimated from the raw photon elevation, and the mean kernel density is calculated. and standard deviation According to statistical rules, retain kernel densities greater than or equal to 1. ( (where the standard deviation is a factor), this process is repeated, and the elevation range is determined based on the extreme points of the kernel density estimation curve to initially extract water surface photons. Then, a kd-tree is constructed based on the original photons, and the scaling factor of the original photon's orbital distance is calculated according to the following formula. To ensure scale similarity along both the track distance and elevation directions:

[0027] in, Represents the scaling coefficients. n Indicates the number of photons. This represents the distance between a photon and its nearest neighbor photon along the orbital direction. This represents the distance between a photon and its nearest neighbor photon in the elevation direction.

[0028] Next, calculate the mean of the K-nearest neighbor distances of the initially extracted water surface photons. and standard deviation ,by ( Using a standard deviation multiple as the threshold for determining characteristic photons, points whose average K-nearest neighbor distance is less than the threshold are retained as water surface characteristic photons in the initially extracted photons, and points whose average K-nearest neighbor distance is less than the threshold are retained as preliminary characteristic photons in the original photons. Finally, based on the water surface characteristic photons, the upper limit of the water surface elevation is calculated according to the following formula. and the lower limit of underwater photon elevation Acquire characteristic photons and original photons distributed within the elevation range:

[0029] in, Indicates the upper limit of water surface elevation. This indicates the lower limit of underwater photon elevation. This represents the function for calculating the upper quartiles. This represents the function for calculating the lower quartiles. Indicates multiples of the interquartile range. This represents the average elevation of photons. h Indicates the photon elevation.

[0030] In some embodiments, step S102, which involves separating and refining water surface feature photons through iterative clustering and local density filtering based on photon elevation to obtain water surface signal photons, includes: determining an initial lower bound for feature photon segmentation based on the water surface feature photons; performing iterative clustering and segmentation on the feature photons to determine an overall threshold for feature photon segmentation; segmenting the feature photons along the trajectory direction and performing iterative clustering and segmentation on the feature photons within each segment to determine elevation thresholds for iterative clustering and segmentation under different conditions; obtaining local water surface feature photons based on the elevation thresholds and adding water surface photons that meet the range requirements to the local water surface feature photons; adding original photons that meet the neighborhood requirements to the local water surface feature photons to obtain local water surface signal photons; and merging all local water surface signal photons to obtain the water surface signal photons.

[0031] In this embodiment, the process of selecting water surface photons that meet the range requirements to add them to the local water surface characteristic photons includes: determining the orbital distance and elevation range of the local water surface characteristic photons; and adding water surface photons distributed within the elevation range of the local water surface characteristic photons to the local water surface characteristic photons.

[0032] The process of selecting original photons that meet the neighborhood requirements and adding them to local water surface feature photons to obtain local water surface signal photons includes: determining the K-neighbor average distance and standard deviation of the local water surface feature photons; obtaining the neighborhood requirements based on the K-neighbor average distance and standard deviation of the local water surface feature photons; and adding the original photons that meet the neighborhood requirements to the local water surface feature photons to obtain local water surface signal photons.

[0033] For example, firstly, based on the characteristic photons of the water surface, the initial lower bound of the elevation for characteristic photon segmentation is calculated according to the following formula:

[0034] in, Indicates the lower bound of the initial elevation. This represents the function for calculating the upper quartiles. This represents the function for calculating the lower quartiles. Indicates multiples of the interquartile range. h This represents the photon elevation. Then, iterative clustering and segmentation are performed on the characteristic photons, and the maximum elevation of the underwater photon cluster is determined. If the elevation exceeds the initial lower bound, stop the iteration; find the two adjacent photons whose elevation changes most between the maximum elevation of the underwater photon cluster and the initial lower bound, and use the average elevation of these two photons as the overall threshold for separating the surface and underwater photons. .

[0035] The feature photons are segmented along the trajectory direction, and iterative clustering is performed on the feature photons within each segment. When the number of iterations is 1, the segmentation uses a global segmentation threshold; when the number of iterations is greater than 1 and the elevation is located at... and When the number of photons between them is less than 2, the elevation will be greater than The minimum elevation of the characteristic photon group is used as the segmentation threshold for this section; when the number of iterations is greater than 1 and the elevation is located at... and When the number of photons between two adjacent photons is not less than 2, the average elevation value of the two adjacent photons with the largest elevation change is used as the segmentation threshold for that segment.

[0036] Based on the determined photon segmentation elevation threshold, local water surface feature photons are obtained and their orbital distance and elevation range are determined. The initially extracted water surface photons distributed within this range are added to the local water surface feature photon set, and local original photons distributed within this range are extracted.

[0037] Finally, based on the average distance and standard deviation of the K-neighborhood of the local water surface feature photons, non-outlier photons are extracted from the original photons according to statistical principles and added to the local water surface feature photon set to obtain the local water surface signal photons. All local water surface signal photons are then merged to obtain the water surface signal photon extraction result.

[0038] In some embodiments, step S103, which extracts underwater signal photons through a multi-scale robust regression and multi-level filtering strategy, includes: setting several sliding windows of different scales to detect underwater photons and determine candidate underwater signal photons; performing sliding window detection with the largest window, recording the range of the sliding window along the track direction, and defining the distribution range of underwater signal photons; sorting the photons in each sliding window according to their track distance, calculating the elevation difference between each photon and its adjacent photons in turn, and eliminating noise photons based on the elevation difference; for the photons sorted by track distance, recording the elevation difference between adjacent photons, iteratively identifying and eliminating noise photons with small-scale abrupt changes until there are no noise photons; for the photons sorted by track distance, recording the track distance between adjacent photons, identifying noise photons from the candidate underwater signal photons until there are no noise photons; performing a second multi-scale sliding window robust regression on the denoised underwater photons to determine the underwater signal photons; and merging the underwater signal photons and underwater feature photons in different sliding windows to form a set of underwater signal photons.

[0039] In this embodiment, several sliding windows of different scales are set to detect underwater photons and determine candidate underwater signal photons. This includes: setting several sliding windows of different scales to detect underwater photons with a specific step size; performing robust regression based on the underwater photons within the sliding windows to determine the overall average absolute error and the single-point absolute error; and filtering underwater photons based on the overall average absolute error and the single-point absolute error to determine candidate underwater signal photons.

[0040] For example, several sliding windows of different sizes are set up to detect underwater photons. Robust regression is performed based on the photon data within each window, and the overall mean absolute error is calculated for each window. Compared with single-point absolute error The overall average absolute error and single-point absolute error Interior points that satisfy the tolerance limits are selected as candidate underwater signal photons. The formulas for calculating the overall average absolute error and the single-point absolute error are as follows:

[0041]

[0042] in, Indicates the overall mean absolute error. n Indicates the number of photons. Indicates the true elevation of an interior point. Indicates the predicted elevation of an interior point.

[0043] When using the largest window for sliding window detection, the range along the track direction of the window with a higher proportion of inner points is defined as the underwater signal photon distribution interval. Photons within each sliding window are sorted according to their distance along the track. The elevation difference between each photon and its adjacent photons is calculated sequentially. If a photon's elevation is both higher and lower than a threshold compared to its adjacent photons, it is considered a noise photon and discarded. This process is iterated until no photons are identified as noise. Based on the above photon points and the underwater signal photon distribution interval, photons within the distribution interval are extracted. For photons sorted by distance along the track, the photon set is divided into multiple photon segments using points where the elevation difference between adjacent photons exceeds a threshold as breakpoints. If the number of points within a photon segment is less than a set threshold or there is a sudden change in elevation relative to adjacent segments, all photons within that segment are identified as noise photons. This process is iterated until no photons are identified as noise.

[0044] For photons sorted by orbital distance, record that the orbital interval between adjacent photons exceeds a distance threshold (which can be adjusted according to the sliding window size), and form a photon sequence number set. Identifying noise photons from candidate underwater signal photons Iterate this process until no more photons are identified as noise. The formula for identifying noise photons is as follows:

[0045] in, Indicates noise photons, Indicates candidate signal photons, This indicates the number of photons used to determine whether a photon is noise. n Represents the photon sequence number set J Size, NThis indicates the number of photons in the window.

[0046] Finally, a second multi-scale sliding window robust regression is performed on the denoised underwater photons. Based on statistical analysis, the non-outliers in each identified interior point are automatically determined as the final underwater signal photons. The extraction results from different windows are then combined with the underwater feature photons to obtain the final underwater signal photon set, as shown below. Figure 5 and Figure 6 As shown, Figure 5 and Figure 6 These are schematic diagrams showing the results of photon extraction from the water surface and bottom in Examples 1 and 2, respectively.

[0047] Based on this, the depth sounding signal is determined using surface and bottom signal photons, and tidal correction is applied to the depth sounding signal, such as... Figure 7 and Figure 8 As shown, Figure 7 and Figure 8 These are schematic diagrams showing the water depth information extraction results in Example 1 and Example 2, respectively.

[0048] The present invention also provides a spaceborne single-photon lidar depth sounding signal extraction device. The spaceborne single-photon lidar depth sounding signal extraction device provided by the present invention will be described below. The spaceborne single-photon lidar depth sounding signal extraction device described below and the spaceborne single-photon lidar depth sounding signal extraction method described above can be referred to in correspondence with each other. Figure 9 This is a structural block diagram of the spaceborne single-photon lidar depth sounding signal extraction device provided by the present invention, as shown below. Figure 9 As shown, the device includes: The preliminary detection module 901 is used to acquire the original photons and extract the water surface feature photons from the original photons based on the elevation distribution and density characteristics of the water surface photons. The water surface detection module 902 is used to separate and extract water surface feature photons by using photon elevation iterative clustering and local density filtering to obtain water surface signal photons; The underwater detection module 903 is used to extract underwater signal photons through multi-scale robust regression and multi-level filtering strategies; The integrated correction module 904 is used to determine the depth sounding signal based on surface signal photons and bottom signal photons, and to perform tidal correction on the depth sounding signal.

[0049] In operation, this device first acquires raw photon data using a preliminary detection module 901. Based on the elevation distribution and density characteristics of surface photons, it extracts surface feature photons from the raw photon data. Then, the surface detection module 902 separates and refines surface-to-subsurface photons through iterative clustering of photon elevations and local density filtering, obtaining surface signal photons. The seabed detection module 903 then uses a multi-scale robust regression and multi-level filtering strategy to achieve complete and accurate extraction of seabed signal photons. After extracting both surface and seabed signal photons, the integration and correction module 904 performs dynamic atmospheric correction on the surface signal photons and calculates the average elevation of the surface signal photons as the surface elevation. Based on the refraction correction method proposed by Parrish, the corrected elevation of the seabed signal photons is calculated. Finally, tidal correction is applied to the aforementioned depth sounding signals according to a global tidal model, converting the depth reference to the mean sea level. Through the above process, the accuracy and completeness of water surface photon elevation distribution and density characteristics are comprehensively utilized to ensure the accuracy and completeness of water surface photon extraction, overcoming the influence of factors such as water surface wave undulation. This device also utilizes the multi-scale characteristics of underwater photons distributed in an orderly manner with the terrain, employing a multi-level filtering underwater signal photon detection method to accurately identify underwater signal photons, overcoming the shortcomings of existing related technologies in underwater photon detection performance under conditions of significant photon signal attenuation and rugged underwater terrain. This device effectively solves the problem of poor photon extraction quality in existing related technologies, providing an effective new approach for accurately, automatically, and cost-effectively acquiring water depth information in remote or inaccessible shallow water areas.

[0050] Figure 10 An example is a schematic diagram of the physical structure of an electronic device, such as... Figure 10 As shown, the electronic device may include: a processor 1001, a communication interface 1002, a memory 1003, and a communication bus 1004. The processor 1001, communication interface 1002, and memory 1003 communicate with each other via the communication bus 1004. The processor 1001 can call logical instructions from the memory 1003 to execute a spaceborne single-photon lidar depth sounding signal extraction method, which includes: Used to acquire raw photons, and extract water surface feature photons from the raw photons based on the elevation distribution and density characteristics of the water surface photons; By using iterative clustering and segmentation based on photon elevation and local density filtering, feature photons on the water surface are separated and refined to obtain water surface signal photons. underwater signal photons are extracted using a multi-scale robust regression and multi-level filtering strategy. The depth sounding signal is determined based on surface signal photons and bottom signal photons, and tidal correction is performed on the depth sounding signal.

[0051] Furthermore, the logical instructions in the aforementioned memory 1003 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, essentially, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0052] On the other hand, the present invention also provides a computer program product, which includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer is able to execute the spaceborne single-photon lidar depth sounding signal extraction method provided by the above methods, the method comprising: Used to acquire raw photons, and extract water surface feature photons from the raw photons based on the elevation distribution and density characteristics of the water surface photons; By using iterative clustering and segmentation based on photon elevation and local density filtering, feature photons on the water surface are separated and refined to obtain water surface signal photons. underwater signal photons are extracted using a multi-scale robust regression and multi-level filtering strategy. The depth sounding signal is determined based on surface signal photons and bottom signal photons, and tidal correction is performed on the depth sounding signal.

[0053] In another aspect, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the satellite-borne single-photon lidar depth sounding signal extraction method provided by the methods described above, the method comprising: Used to acquire raw photons, and extract water surface feature photons from the raw photons based on the elevation distribution and density characteristics of the water surface photons; By using iterative clustering and segmentation based on photon elevation and local density filtering, feature photons on the water surface are separated and refined to obtain water surface signal photons. underwater signal photons are extracted using a multi-scale robust regression and multi-level filtering strategy. The depth sounding signal is determined based on surface signal photons and bottom signal photons, and tidal correction is performed on the depth sounding signal.

[0054] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; 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. Those skilled in the art can understand and implement this without any creative effort.

[0055] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms. Based on this understanding, the above technical solutions, in essence or the parts that contribute to the prior art, can be embodied in the form of software products. These computer software products can be stored in computer-readable storage media, such as ROM / RAM, magnetic disks, optical disks, etc., and include several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0056] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for extracting depth sounding signals from a spaceborne single-photon lidar, characterized in that, include: The original photons are acquired, and water surface feature photons are extracted from the original photons based on the elevation distribution and density characteristics of the water surface photons. By using iterative clustering and segmentation based on photon elevation and local density filtering, feature photons on the water surface are separated and refined to obtain water surface signal photons. underwater signal photons are extracted using a multi-scale robust regression and multi-level filtering strategy. The depth sounding signal is determined based on the surface signal photons and the bottom signal photons, and the depth sounding signal is tidal corrected.

2. The method for extracting depth sounding signals from a spaceborne single-photon lidar according to claim 1, characterized in that, Based on the elevation distribution and density characteristics of water surface photons, water surface feature photons are extracted from the original photons, including: The nuclear density of the original photons is determined, and the elevation range of the original photons is determined based on the nuclear density to initially extract the water surface photons; A kd tree is constructed based on the original photon, and the scale transformation coefficients of the original photon's orbital distance are determined. The threshold for determining feature photons is determined based on the mean and standard deviation of the average K-nearest neighbor distance of the water surface photons, and water surface feature photons are extracted from the water surface photons based on the threshold for determining feature photons, and preliminary feature photons are extracted from the original photons. Based on the surface feature photons, the elevation range of global surface and underwater photons is determined, and the preliminary feature photons within the elevation range are used as feature photons.

3. The method for extracting depth sounding signals from a spaceborne single-photon lidar according to claim 2, characterized in that, Determine the nuclear density of the original photons, and based on the nuclear density, determine the elevation range of the original photons to initially extract water surface photons, including: Estimate the kernel density of the original photon elevation, and determine the extraction threshold based on the mean and standard deviation of the kernel density; Photons with a kernel density greater than the extraction threshold within the elevation range are retained, and the elevation range of the original photons is determined based on the extreme points of the kernel density estimation curve, thus initially extracting photons from the water surface.

4. The method for extracting depth sounding signals from a spaceborne single-photon lidar according to claim 1, characterized in that, By using iterative clustering and segmentation based on photon elevation and local density filtering, feature photons on the water surface are separated and refined to obtain surface signal photons. Simultaneously, underwater feature photons and underwater photons are acquired, including: Based on the water surface feature photons, an initial lower bound for feature photon segmentation is determined; The feature photons are iteratively clustered and segmented to determine the overall threshold for segmentation of the feature photons; The feature photons are segmented along the trajectory direction, and the feature photons in each segment are iteratively clustered and segmented to determine the elevation threshold for iterative clustering and segmentation under different conditions. Based on the elevation threshold, local surface characteristic photons, local underwater characteristic photons, and local underwater photons are obtained, and the surface photons that meet the range requirements are added to the local surface characteristic photons. The original photons that meet the neighborhood requirements are added to the local water surface feature photons to obtain local water surface signal photons; All the local surface signal photons are merged to obtain the surface signal photons, and the local underwater feature photons and the local underwater photons are merged to obtain the underwater feature photons and the underwater photons, respectively.

5. The method for extracting depth sounding signals from a spaceborne single-photon lidar according to claim 4, characterized in that, The selection of water surface photons that meet the range requirements and their addition to local water surface characteristic photons includes: Determine the orbital distance and elevation range of the local water surface characteristic photons; The water surface photons distributed within the elevation range of the local water surface characteristic photons are added to the local water surface characteristic photons.

6. The method for extracting depth sounding signals from a spaceborne single-photon lidar according to claim 4, characterized in that, The original photons that meet the neighborhood requirements are added to the local water surface feature photons to obtain local water surface signal photons, including: Determine the K-neighborhood average distance and standard deviation of the local water surface feature photons; The neighborhood requirement is obtained based on the K-neighbor average distance and standard deviation of the local water surface feature photons. The original photons that meet the neighborhood requirements are added to the local water surface feature photons to obtain local water surface signal photons.

7. The method for extracting depth sounding signals from a spaceborne single-photon lidar according to claim 1, characterized in that, By employing a multi-scale robust regression and multi-level filtering strategy, underwater signal photons are extracted, including: Several sliding windows of different sizes were set up to detect underwater photons and identify candidate underwater signal photons; Using the largest window for sliding window detection, the range of the sliding window along the track direction is recorded to define the distribution range of underwater signal photons; Photons within each sliding window are sorted according to their distance along the track. The elevation difference between each photon and its adjacent photons is calculated sequentially, and noisy photons are removed based on the elevation difference. For photons sorted by distance along the track, the elevation difference between adjacent photons is recorded, and noisy photons with small-scale abrupt changes are iteratively identified and eliminated until there are no noisy photons. For photons sorted by orbital distance, the orbital distances of adjacent photons are recorded, and noisy photons are identified from the candidate underwater signal photons until no noise photons are found. A second multi-scale sliding window robust regression is performed on the denoised underwater photons to determine the underwater signal photons; the underwater signal photons and underwater feature photons in different sliding windows are combined to form the underwater signal photon set.

8. The method for extracting depth sounding signals from a spaceborne single-photon lidar according to claim 7, characterized in that, Several sliding windows of different sizes are set up to detect underwater photons and identify candidate underwater signal photons, including: Several sliding windows of different sizes are set up to detect underwater photons in a specific step size; Robust regression is performed based on underwater photons within a sliding window to determine the overall mean absolute error and the single-point absolute error. The underwater photons are screened based on the overall average absolute error and the single-point absolute error to determine candidate underwater signal photons.

9. A spaceborne single-photon lidar depth sounding signal extraction device, characterized in that, include: The preliminary detection module is used to acquire raw photons and extract water surface feature photons from the raw photons based on the elevation distribution and density characteristics of the water surface photons. The water surface detection module is used to separate and extract water surface feature photons by using photon elevation iterative clustering and local density filtering to obtain water surface signal photons; The underwater detection module is used to extract underwater signal photons through multi-scale robust regression and multi-level filtering strategies; An integrated correction module is used to determine the depth sounding signal based on the surface signal photons and the bottom signal photons, and to perform tidal correction on the depth sounding signal.

10. 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 method for extracting depth signals from a spaceborne single-photon lidar as described in any one of claims 1 to 8.

Images