Multi-ship multi-beam point cloud data matching method based on elevation correction, storage medium and equipment

Abstract

The invention discloses a multi-ship multi-beam point cloud data matching method based on elevation correction, a storage medium and equipment. The method comprises the following steps: a plurality of unmanned ships jointly collect point cloud data of underwater topography, read trajectory data of each unmanned ship, divide segments according to the trajectory data, and store trajectory point data in the segments; performing block division on all the point cloud data, and recording GPS time of the point cloud data and a leg to which the point cloud data belongs; performing section division on the point cloud data in each block, and performing leg correction matching on the point cloud data in each section; after leg correction matching of the point cloud data in each section is completed, leg correction matching of the point cloud data between adjacent sections is carried out; and after finishing the leg correction matching of the point cloud data among the sections, performing leg correction matching of the point cloud data among the adjacent sections to realize the matching of the multi-beam point cloud data of the multiple unmanned ships. According to the invention, the problem of mismatching of point cloud data acquired by multiple ships is solved.

Description

Technical Field

[0001] The present invention belongs to the field of surveying and mapping science and technology, and in particular, relates to a multi-vessel multi-beam point cloud data matching method based on elevation correction, a storage medium and a device. Background Art

[0002] In recent years, unmanned ships equipped with multi-beam systems have become an important technical means of obtaining underwater topographic measurement data, gradually replacing traditional manual measurement methods. With the continuous innovation and development of technology, my country's unmanned ship underwater topographic measurement technology has become quite mature, and the navigation performance of unmanned ships has become more stable and intelligent. The main advantages of unmanned ship technology in underwater topographic measurement include high efficiency, safety, lightness and compactness. In addition, the operating cost of unmanned ship measurement technology is low, which makes it widely welcomed by surveying and mapping units.

[0003] Multi-beam systems are used for underwater terrain scanning and can generate strip-shaped point cloud data that reflects the terrain undulations. The width of the scanning strip depends on the water depth and the scanning angle. In order to completely cover the waterway, it is generally necessary to scan multiple strips along the navigation direction and conduct inspection scanning strips perpendicular to the navigation direction at appropriate intervals to ensure the integrity and accuracy of the data. In addition, with the development of multi-beam bathymetric technology, it has become a common operating method to use multiple ships or multiple devices to scan strips simultaneously. The different sampling frequencies, intervals, and opening angles of different devices result in inconsistent density of point cloud data, which brings challenges to data fusion. There are also problems with the synchronization of time reference and space reference, which makes data processing complicated and critical.

[0004] During the data collection process, the multi-beam system is affected by GNSS positioning errors, equipment attitude parameter errors, inertial navigation system errors, etc., especially in the case of multiple ships or multiple devices, the sensor accuracy of each device is different, and the correction of system parameters is different, which can easily lead to the problem of mismatch in the overlapping area of ​​the scanned flight path point cloud. This error not only seriously reduces the accuracy of the measurement results, affects the classification and processing of point cloud data and underwater terrain modeling, but also may cause the final results to fail to meet the specification requirements.

[0005] In order to solve these problems, the prior art has proposed a variety of correction methods and data processing technologies to improve the matching degree and accuracy of point cloud data, but there are the following problems and shortcomings: it can only process the multi-beam scanning data of a single ship and a single device, and cannot process the data of multiple ships or multiple devices; it is inevitable that there will be misjudgment and omission of noise after filtering out the noise and then matching, resulting in the loss of original point cloud information in the matching result, and increasing the operation steps and labor costs; the method of uniformly correcting the overall fitting surface of all navigation strips is difficult to describe the waterway with changeable shapes, and the details are not matched in place; the method of local correction in a small range only focuses on gross errors and random errors, but ignores systematic errors. Although the result ensures the smoothness of the overlapping area, the overall shape is distorted.

[0006] Therefore, it is particularly important to develop a correction method that can process the scanning data of multiple ships or multiple devices and adapt to more complex measurement environments and higher accuracy requirements. Summary of the invention

[0007] In view of the problems existing in the prior art, the present invention provides a multi-ship multi-beam point cloud data matching method, storage medium and device based on elevation correction, aiming to solve the point cloud data mismatch problem caused by various errors in the prior art. The multi-ship multi-beam point cloud data matching method of the present invention analyzes the errors in the overlapping areas, fits the error quadratic surface in segments, and corrects the elevation data of the point cloud as a whole, so that the point clouds of different flight strips are matched consistently, thereby improving the accuracy and consistency of the measurement data.

[0008] To achieve the above technical objectives, the present invention adopts the following technical solution: a multi-vessel multi-beam point cloud data matching method based on elevation correction, specifically comprising the following steps:

[0009] Step 1: Multiple unmanned ships jointly collect point cloud data of underwater terrain, read the trajectory data of each unmanned ship, divide the route into sections according to the trajectory data, and save the trajectory point data within the route;

[0010] Step 2: Divide all point cloud data into blocks and record the GPS time and flight segment of the point cloud data;

[0011] Step 3: Divide the point cloud data in each block into sections, and perform segment correction and matching on the point cloud data in each section;

[0012] Step 4: After completing the segment correction and matching of the point cloud data in each section, perform segment correction and matching of the point cloud data between adjacent sections;

[0013] Step 5: After completing the segment correction and matching of the point cloud data between each section, perform segment correction and matching of the point cloud data between adjacent blocks to achieve the matching of multi-beam point cloud data of multiple unmanned ships.

[0014] Furthermore, step 1 includes the following sub-steps:

[0015] Step 1.1, read the trajectory data of each unmanned ship, obtain the GPS time, position coordinates and azimuth of each point in each trajectory data, and calculate the change between adjacent azimuths;

[0016] Step 1.2, taking the continuous trajectory data whose change between adjacent azimuths is less than a threshold as the candidate route of the unmanned ship;

[0017] Step 1.3, the alternative segments of each unmanned ship are sorted in ascending order according to the minimum value of the GPS time in each alternative segment, and the alternative segments connected head to tail are merged to obtain the segment of each unmanned ship;

[0018] Step 1.4: Number all the unmanned ship's segments, record the GPS time and location coordinates of each point in the segment, and save the track point data in the segment.

[0019] Furthermore, step 3 includes the following sub-steps:

[0020] Step 3.1, divide the point cloud data in each block into sections, split the point cloud data in each section according to the flight segments they belong to, sort the flight segments to which the point cloud data belong in the order of overlapping point clouds from large to small in the split point cloud data, and use the horizontal and vertical coordinates of the split point cloud data to generate the XY plane projection triangulation of the flight segments;

[0021] Step 3.2, find the trajectory point of the unmanned ship corresponding to each point cloud data in the section, calculate the weight of the point cloud data according to the distance and angle between the point cloud data and the trajectory point of the unmanned ship, and fit the weights of all the point cloud data in the section into a weighted quadratic surface;

[0022] Step 3.3: According to the flight segment to which the sorted point cloud data belongs, the correction and matching of the flight segments within the segment is performed in combination with the triangulated network and weighted quadratic surface of the flight segment.

[0023] Furthermore, in step 3.2, the process of calculating the weight w of the point cloud data according to the distance and angle between the point cloud data and the trajectory point of the unmanned ship is:

[0024]

[0025] Wherein, a represents the zenith angle between the point cloud and the trajectory point of the unmanned ship, and d represents the distance between the point cloud data and the trajectory point of the unmanned ship.

[0026] Furthermore, step 3.3 includes the following sub-steps:

[0027] Step 3.3.1. Take out the point cloud data C of the two corresponding flight segments in order A and C B , extract point cloud data C A Overlapping point cloud C in AO , calculate the overlapping point cloud C AO Each point cloud data in to point cloud data C B The distance of the XY plane projection triangulation network of the flight segment, the fitting error quadratic surface Q A ;

[0028] Step 3.3.2, obtain point cloud data C A Each point in the error quadratic surface Q A The error D on the point cloud data C is obtained A Each point in the point cloud data C A The weight P on the quadratic surface A , get point cloud data C A Each point in the point cloud data C B The weight P on the quadratic surface B , calculate the point cloud data C A Corrected Z value of each point in The point cloud data C A Overlapping point cloud C AO The Z value of each point in is corrected by the corresponding Z value d z,A Make corrections;

[0029] Step 3.3.3, obtain overlapping point cloud C AO Calculate the horizontal convex hull of point cloud data C A The shortest horizontal distance s from each non-overlapping point to the horizontal convex hull A , calculate the point cloud data C A Correction value of Z value of each non-overlapping point in The point cloud data C A The Z value of each non-overlapping point in is corrected by the corresponding Z value d f,A Make corrections;

[0030] Step 3.3.4: Get point cloud data C B Each point in the error quadratic surface Q A The error D on the point cloud data C is obtained B Each point in the point cloud data C A The weight P on the quadratic surface A , get point cloud data C B Each point in the point cloud data C B The weight P on the quadratic surface B , calculate the point cloud data C B Corrected Z value of each point in The point cloud data C B Overlapping point cloud C AO The Z value of each point in is corrected by the corresponding Z value d z,B Make corrections;

[0031] Step 3.3.5: Calculate point cloud data C B The shortest horizontal distance s from each non-overlapping point to the horizontal convex hull B , calculate the point cloud data C B Correction value of Z value of each non-overlapping point in The point cloud data C B The Z value of each non-overlapping point in is corrected by the corresponding Z value d f,B Make corrections;

[0032] Step 3.3.6: Corrected point cloud data C A and C B Merge, the weight of each point after merging is P AB =P A +P B ;

[0033] Step 3.3.7: Correct and match the merged point cloud data with the point cloud data of the next flight segment according to the method of steps 3.3.1 to 3.3.6, until all the point cloud data in each segment are corrected and matched to form a point cloud.

[0034] Furthermore, step 4 includes the following sub-steps:

[0035] Step 4.1: Extract adjacent segment S A , S B The common side MN of the two sections is vertically extended to both sides of the common side MN according to the set section buffer length L, so as to obtain the buffer zone R of the two sections;

[0036] Step 4.2: Extract the pre-correction segment S A Point cloud data C in buffer R AR , Segment S B Point cloud data C in buffer R BR , extract the corrected segment S A Point cloud data C' in buffer R AR , Segment S B Point cloud data C' in buffer R BR ;

[0037] Step 4.3: Calculate the point cloud data C before correction AR And the corrected point cloud data C' AR , point cloud data before correction C BR And the corrected point cloud data C' BRCorrection surface for the same flight segment k and

[0038] Step 4.4: Obtain corrected point cloud data C' AR Each point in the corrected surface Correction value d Az1 , in the correction surface Correction value d Bz1 , the horizontal distance d to the common edge MN MN,AR , calculate the corrected point cloud data C' AR Corrected Z value of each point in The corrected point cloud data C' AR The Z value of each point in is corrected by the corresponding Z value d z,AR Make corrections;

[0039] Step 4.5: Obtain corrected point cloud data C' BR Each point in the corrected surface Correction value d Az2 , in the correction surface Correction value d Bz2 , obtain the corrected point cloud data C' BR The horizontal distance d from each point to the common edge MN MN,BR , calculate the corrected point cloud data C' BR Corrected Z value of each point in The corrected point cloud data C' BR The Z value of each point in is corrected by the corresponding Z value d z,BR Make corrections;

[0040] Step 4.6: Corrected C' AR and C' BR Merge the segments and repeat steps 4.1-4.5 until the point cloud data of all segments in each block are corrected and matched to form a point cloud.

[0041] Furthermore, in step 4.3, the surface is corrected The calculation process is as follows: Get the point cloud data C before correction AR Point cloud data in flight segment k Get the corrected point cloud data C' AR Point cloud data in flight segment k Based on point cloud data Generate point cloud data by using the horizontal and vertical coordinates of each point XY plane projection triangulation Calculate point cloud data Projecting triangulated network from each point to XY plane The distance is fitted into the correction surface

[0042] Furthermore, step 5 includes the following sub-steps:

[0043] Step 5.1: Extract adjacent block H A , H B The common edge PQ of the block is vertically extended to both sides of the common edge PQ according to the set block buffer length T to obtain the buffer zone F of the two segments;

[0044] Step 5.2: Extract the pre-correction block H A Point cloud data C in buffer F AF 、Block H B Point cloud data C in buffer F BF , extract the corrected block H A Point cloud data C' in buffer F AF 、Block H B Point cloud data C' in buffer F BF ;

[0045] Step 5.3: Calculate the point cloud data C before correction AF And the corrected point cloud data C' AF , point cloud data before correction C BF And the corrected point cloud data C' BF Correction surface for the same segment l and

[0046] Step 5.4: Obtain corrected point cloud data C' AF Each point in the corrected surface Correction value d Al1 , obtain the corrected point cloud data C' BF In correcting the surface Correction value d Bl1 , obtain the corrected point cloud data C' AF The horizontal distance d from each point to the common edge PQ PQ,AF , calculate the corrected point cloud data C' AF Corrected Z value of each point in The corrected point cloud data C' AF The Z value of each point in is corrected by the corresponding Z value d z,AF Make corrections;

[0047] Step 5.5: Obtain corrected point cloud data C' BF Each point in the corrected surface Correction value d Al2 , in the correction surface Correction value d Bl2 , obtain the corrected point cloud data C'BF The horizontal distance d from each point to the common edge PQ PQ,BF , calculate the corrected point cloud data C' BF Corrected Z value of each point in The corrected point cloud data C' BF The Z value of each point in is corrected by the corresponding Z value d z,BF Make corrections;

[0048] Step 5.6: Corrected C' AF and C' BF Merge the blocks and repeat steps 5.1-5.5 until the point cloud data of all blocks are corrected and matched to form a point cloud.

[0049] Furthermore, the present invention also provides a computer-readable storage medium storing a computer program, wherein the computer program enables a computer to execute the multi-vessel multi-beam point cloud data matching method based on elevation correction.

[0050] Furthermore, the present invention also provides an electronic device, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the multi-vessel multi-beam point cloud data matching method based on elevation correction is implemented.

[0051] Compared with the prior art, the present invention has the following beneficial effects: the multi-vessel multi-beam point cloud data matching method based on elevation correction of the present invention divides the point cloud data collected by multiple ships into blocks and sections, and performs elevation correction within the sections, between sections and between blocks respectively. It is suitable for various complex waterway, lake and ocean measurement environments, especially in underwater topographic measurement with large terrain undulations. It improves the matching degree and accuracy of multi-beam point cloud data, ensures the consistency of point cloud data between flight strips, and effectively avoids the problems of breakage and mutation in subsequent processing; the method of the present invention can process the point cloud data of multiple unmanned ships, and eliminates the relative system error by fitting the error quadratic surface between overlapping flight segments, thereby avoiding the problem of respective error correction. The fitting of the error quadratic surface has good noise resistance, and no pre-processing such as denoising is required, so that the method does not require manual intervention, is simple to operate and is efficient. The multi-vessel multi-beam point cloud data matching method based on elevation correction of the present invention significantly improves the quality of the results of unmanned ship multi-beam measurement, and provides strong technical support for underwater topographic survey, engineering measurement and environmental monitoring and other fields. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 It is a flow chart of the multi-vessel multi-beam point cloud data matching method based on elevation correction of the present invention;

[0053] Figure 2Multi-beam point cloud data of the channel scanned by the two ships;

[0054] Figure 3 It is a partial schematic diagram of the flight segment extraction results;

[0055] Figure 4 A local schematic diagram after marking the flight segments for the point cloud;

[0056] Figure 5 It is a local schematic diagram before and after correction in the point cloud section;

[0057] Figure 6 It is a local schematic diagram before and after the correction between point cloud segments;

[0058] Figure 7 The figure is a local schematic diagram of generating a DEM by correcting the channel point clouds before and after using the multi-ship multi-beam point cloud data matching method based on elevation correction of the present invention. DETAILED DESCRIPTION

[0059] The technical solution of the present invention is further explained below in conjunction with the accompanying drawings.

[0060] like Figure 1 The flowchart of the multi-vessel multi-beam point cloud data matching method based on elevation correction of the present invention is as follows:

[0061] Step 1: Figure 2 , multiple unmanned ships jointly collect point cloud data of underwater terrain, read the trajectory data of each unmanned ship, divide the route into sections according to the trajectory data, and save the trajectory point data within the route; specifically, it includes the following sub-steps:

[0062] Step 1.1, read the trajectory data of each unmanned ship, obtain the GPS time, position coordinates and azimuth of each point in each trajectory data, and calculate the change between adjacent azimuths;

[0063] Step 1.2: The continuous trajectory data with the change between adjacent azimuth angles less than the threshold value is used as the candidate segment of the unmanned ship. In one technical solution of the present invention, the change threshold between adjacent azimuth angles is 30°, which can eliminate the U-turn and abnormal parts in the trajectory data of the unmanned ship. Figure 3-4 ;

[0064] Step 1.3, the alternative segments of each unmanned ship are sorted in ascending order according to the minimum value of the GPS time in each alternative segment, and the alternative segments connected head to tail are merged to obtain the segment of each unmanned ship;

[0065] Step 1.4: Number all the unmanned ship's segments, record the GPS time and location coordinates of each point in the segment, and save the track point data in the segment.

[0066] Step 2: To prevent a single point cloud file from being too large, which would cause memory burden and efficiency problems, all point cloud data are divided into blocks, such as Figure 2 As shown, the GPS time and flight segment of the point cloud data are recorded.

[0067] Step 3, since the error of point cloud data in the overlapping area usually shows a systematic change trend, this is closely related to its causes, such as GNSS positioning error, equipment attitude parameter error, inertial navigation system error, etc. For this kind of systematic error, it is inappropriate to process each point independently, because this will ignore the overall error distribution and be too affected by random errors. Therefore, by fitting the quadratic surface of the error of each point, these systematic errors can be better captured and corrected, and the error quadratic surface, as a smooth and continuous mathematical model, can provide stable results when data is interpolated. In addition, when the area is too large, the change of the error becomes complicated, and a single error quadratic surface cannot well express this change trend. Therefore, a segmented fitting method is adopted, that is, the area is divided into several sections, and each section is fitted with an error quadratic surface, so that the errors in each section can be more accurately described and corrected, thereby achieving overall accurate correction. Therefore, the present invention divides the point cloud data in each block into sections, and performs segment correction matching on the point cloud data in each section; specifically includes the following sub-steps:

[0068] Step 3.1, divide the point cloud data in each block into sections, split the point cloud data in each section according to the flight segments they belong to, extract the number of overlapping point clouds in the split point cloud data, and use it to determine the connectivity between the flight segments to which the split point cloud data belongs. If they are not connected, skip the section, otherwise, sort the flight segments to which the point cloud data belongs in the order of the overlapping point clouds in the split point cloud data from large to small, and use the horizontal and vertical coordinates of the split point cloud data to generate the XY plane projection triangulation of the flight segments;

[0069] Step 3.2: Find the trajectory point of the unmanned ship corresponding to each point cloud data in the segment, and calculate the weight of the point cloud data according to the distance and angle between the point cloud data and the trajectory point of the unmanned ship to which it belongs. Wherein, a represents the zenith angle between the point cloud and the trajectory point of the unmanned ship, d represents the distance between the point cloud data and the trajectory point of the unmanned ship, the horizontal coordinates of the current point are taken as the X and Y values, the weight is taken as the Z value, and the weighted point cloud is obtained. The weighted quadratic surface is fitted according to the weighted point cloud to show the systematic change trend of the point cloud weight in the section, and provide stable results when interpolating data;

[0070] Step 3.3: According to the flight segment to which the sorted point cloud data belongs, combined with the triangulated network and weighted quadratic surface of the flight segment, the flight segment within the segment is corrected and matched; specifically, the following sub-steps are included:

[0071] Step 3.3.1. Take out the point cloud data C of the two corresponding flight segments in order A and C B , extract point cloud data C A Overlapping point cloud C in AO , calculate the overlapping point cloud C AO Each point cloud data in to point cloud data C B The distance of the XY plane projection triangulation of the flight segment, indicating the distance of the point from the point cloud data C A to C B The error, C AO The horizontal coordinates of each point are X and Y values, and the error distance is Z value, so C A to C B The error point cloud is used to fit the error quadratic surface Q A ;

[0072] Step 3.3.2, obtain point cloud data C A Each point in the error quadratic surface Q A The error D on the point cloud data C is obtained A Each point in the point cloud data C A The weight P on the quadratic surface A , get point cloud data C A Each point in the point cloud data C B The weight P on the quadratic surface B , calculate the point cloud data C A Corrected Z value of each point in In order to avoid the break between the overlapping area and the non-overlapping area after correction, the point cloud data C A Overlapping point cloud C AO The Z value of each point in is corrected by the corresponding Z value d z,A Make corrections;

[0073] Step 3.3.3, obtain overlapping point cloud C AO Calculate the horizontal convex hull of point cloud data C A The shortest horizontal distance s from each non-overlapping point to the horizontal convex hull A , calculate the point cloud data C A Correction value of Z value of each non-overlapping point in The point cloud data C A The Z value of each non-overlapping point in is corrected by the corresponding Z value d f,A Make corrections;

[0074] Step 3.3.4: Get point cloud data C B Each point in the error quadratic surface Q A The error D on the point cloud data C is obtained B Each point in the point cloud data CA The weight P on the quadratic surface A , get point cloud data C B Each point in the point cloud data C B The weight P on the quadratic surface B , calculate the point cloud data C B Corrected Z value of each point in The point cloud data C B Overlapping point cloud C AO The Z value of each point in is corrected by the corresponding Z value d z,B Make corrections;

[0075] Step 3.3.5: Calculate point cloud data C B The shortest horizontal distance s from each non-overlapping point to the horizontal convex hull B , calculate the point cloud data C B Correction value of Z value of each non-overlapping point in The point cloud data S B The Z value of each non-overlapping point in is corrected by the corresponding Z value d f,B Correction; point cloud data C of the two flight segments A and C B The correction effect is as follows Figure 5 As shown, Figure 5 The left picture in the figure is a schematic diagram of the overlapping point clouds in two segments before correction. Figure 5 The right picture in the figure is a schematic diagram of the overlapping point clouds in the corresponding segment after correction. It can be seen that before correction, the overlapping point clouds of the two segments are layered. After correction, the overlapping point clouds are matched into one layer of point clouds and maintain the same change trend, thereby improving the accuracy and consistency of the measurement data.

[0076] Step 3.3.6: Corrected point cloud data C A and C B Merge, the weight of each point after merging is P AB =P A +P B ;

[0077] Step 3.3.7: Correct and match the merged point cloud data with the point cloud data of the next flight segment according to the method of steps 3.3.1 to 3.3.6, until all the point cloud data in each segment are corrected and matched to form a point cloud.

[0078] Step 4: After completing the segment correction and matching of the point cloud data in each segment, perform segment correction and matching of the point cloud data between adjacent segments. In the correction and matching process of the segment point cloud data, a correction surface is fitted in each segment to more accurately describe and correct the errors between the segments, thereby achieving overall accurate correction. Specifically, the following sub-steps are included:

[0079] Step 4.1: Determine the adjacent segment S A , S B Is there a common edge? If not, there is no need to perform adjacent segment S A , S B The point cloud data within the same area does not need to be stitched together, and there is no need to stitch adjacent segments S A , S B Correction of the point cloud data within, otherwise, extract the adjacent segment S A , S B The common side MN of the two sections is vertically extended to both sides of the common side MN according to the set section buffer length L, so as to obtain the buffer zone R of the two sections;

[0080] Step 4.2: Extract the pre-correction segment S A Point cloud data C in buffer R AR , Segment S B Point cloud data C in buffer R BR , if C AR or C BR The midpoint number is zero, then segment S A , S B The point clouds of C are not connected in the buffer R, and no correction is required; AR , C BR If the midpoints are all greater than zero, then segment S A , S B The point clouds of are connected in the buffer R, and need to be corrected, and the corrected segment S is extracted A Point cloud data C' in buffer R AR , Segment S B Point cloud data C' in buffer R BR ;

[0081] Step 4.3: Calculate the point cloud data C before correction AR And the corrected point cloud data C' AR , point cloud data before correction C BR And the corrected point cloud data C' BR Correction surface for the same flight segment k and Specifically, correcting the surface The calculation process is as follows: Get the point cloud data C before correction AR Point cloud data in flight segment k Get the corrected point cloud data C' AR Point cloud data in flight segment k Based on point cloud data Generate point cloud data by using the horizontal and vertical coordinates of each point XY plane projection triangulation Calculate point cloud data Projecting triangulated network from each point to XY plane The distance represents the point cloud data The correction distance formed in the intra-segment correction is The horizontal coordinates of each point are X and Y values, and the corrected distance is Z value, and the point cloud is obtained. Corrected point cloud corrected within the segment, fitting the quadratic surface to obtain the corrected surface Fitting to Corrected Surface

[0082] Correction Surface The calculation process is as follows: Get the point cloud data C before correction BR Point cloud data in flight segment k Get the corrected point cloud data C' BR Point cloud data in flight segment k Based on point cloud data Generate point cloud data by using the horizontal and vertical coordinates of each point XY plane projection triangulation Calculate point cloud data Projecting triangulated network from each point to XY plane The distance represents the point cloud data The correction distance formed in the intra-segment correction is The horizontal coordinates of each point are X and Y values, and the corrected distance is Z value, and the point cloud is obtained. Corrected point cloud corrected within the segment, fitting the quadratic surface to obtain the corrected surface Fitting to the corrected surface.

[0083] Step 4.4: C' AR and C' BR Perform inter-segment correction to ensure smooth transition of the correction differences within the two segments, and obtain the corrected point cloud data C' AR Each point in the corrected surface Correction value d Az1 , in the correction surface Correction value d Bz1 , the horizontal distance d to the common edge MN MN,AR , weight the difference in correction within the segment by distance, make a smooth transition, and calculate the corrected point cloud data C' AR Corrected Z value of each point in The corrected point cloud data C' AR The Z value of each point in is corrected by the corresponding Z value d z,AR Make corrections;

[0084] Step 4.5: Obtain corrected point cloud data C' BR Each point in the corrected surface Correction value dAz2 , in the correction surface Correction value d Bz2 , obtain the corrected point cloud data C' BR The horizontal distance d from each point to the common edge MN MN,BR , calculate the corrected point cloud data C' BR Corrected Z value of each point in The corrected point cloud data C' BR The Z value of each point in is corrected by the corresponding Z value d z,BR Make corrections; Figure 6 For adjacent segment S A , S B The effect diagram of the correction and matching of the midpoint cloud data. Figure 6 The left picture in the figure is the point cloud data before correction of adjacent segments. Figure 6 The right picture in the figure is the point cloud data after correction of adjacent segments. It can be seen that before correction, there are two segments with highly sudden changes at the junction of two adjacent segments. After correction, the point cloud data at the junction of adjacent segments are highly consistent and the transition is smoother.

[0085] Step 4.6: Corrected C' AR and C' BR Merge the segments and repeat steps 4.1-4.5 until the point cloud data of all segments in each block are corrected and matched to form a point cloud.

[0086] Step 5: The connected point clouds may have correction differences due to the different corrections in their respective blocks, causing mutations or breaks. Therefore, after completing the segment correction matching of the point cloud data between each section, the segment correction matching of the point cloud data between adjacent blocks is performed to achieve the matching of multi-beam point cloud data of multiple unmanned ships; specifically, the following sub-steps are included:

[0087] Step 5.1: Extract adjacent block H A , H B The common edge PQ of the block is vertically extended to both sides of the common edge PQ according to the set block buffer length T to obtain the buffer zone F of the two segments;

[0088] Step 5.2: Extract the pre-correction block H A Point cloud data C in buffer F AF 、Block H B Point cloud data C in buffer F BF , extract the corrected block H A Point cloud data C' in buffer F AF 、Block H B Point cloud data C' in buffer F BF ;

[0089] Step 5.3: Calculate the point cloud data C before correction AF And the corrected point cloud data C' AF , point cloud data before correction C BF And the corrected point cloud data C' BF Correction surface for the same segment l and

[0090] Step 5.4: Obtain corrected point cloud data C' AF Each point in the corrected surface Correction value d Al1 , obtain the corrected point cloud data C' BF In correcting the surface Correction value d Bl1 , obtain the corrected point cloud data C' AF The horizontal distance d from each point to the common edge PQ PQ,AF , calculate the corrected point cloud data C' AF Corrected Z value of each point in The corrected point cloud data C' AF The Z value of each point in is corrected by the corresponding Z value d z,AF Make corrections;

[0091] Step 5.5: Obtain corrected point cloud data C' BF Each point in the corrected surface Correction value d Al2 , in the correction surface Correction value d Bl2 , obtain the corrected point cloud data C' BF The horizontal distance d from each point to the common edge PQ PQ,BF , calculate the corrected point cloud data C' BF Corrected Z value of each point in The corrected point cloud data C' BF The Z value of each point in is corrected by the corresponding Z value d z,BF Make corrections;

[0092] Step 5.6: Corrected C' AF and C' BF Merge the blocks and repeat steps 5.1-5.5 until the point cloud data of all blocks are corrected and matched to form a point cloud.

[0093] like Figure 7 A local schematic diagram of a digital elevation model (DEM) is generated by correcting the channel point cloud before and after the multi-ship multi-beam point cloud data matching method based on elevation correction of the present invention. Figure 7The left picture in the figure is a partial schematic diagram of the DEM generated from the channel point cloud before correction. Figure 7 The right figure in the figure is a local schematic diagram of the DEM generated by the corrected channel point cloud. Due to the inconsistency of the elevations of the point clouds of different sections in the overlapping area, the DEM generated by the point cloud before correction has spike-like fluctuations in the overlapping area, and the transition from the overlapping area to the non-overlapping area is obviously abrupt; while the DEM generated by the corrected point cloud has no obvious fluctuations in the overlapping area, and the transition with the non-overlapping area is natural and smooth, which shows that the method of the present invention can correct the elevation data of the point cloud as a whole, so that the point clouds of different flight strips are matched consistently, thereby improving the accuracy and consistency of the measurement data.

[0094] In a technical solution of the present invention, a computer-readable storage medium is further provided, storing a computer program, wherein the computer program enables a computer to execute the multi-vessel multi-beam point cloud data matching method based on elevation correction.

[0095] In a technical solution of the present invention, an electronic device is also provided, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the multi-vessel multi-beam point cloud data matching method based on elevation correction is implemented.

[0096] In the embodiments disclosed in the present application, the computer storage medium can be a tangible medium that can contain or store a program for use by or in conjunction with an instruction execution system, device or equipment. The computer storage medium can include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices or equipment, or any suitable combination of the above. More specific examples of computer storage media can include electrical connections based on one or more lines, portable computer disks, hard disks, random access memories (RAM), read-only memories (ROM), erasable programmable read-only memories (EPROM or flash memory), optical fibers, portable compact disk read-only memories (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the above.

[0097] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed in this application can be implemented in electronic hardware or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0098] The above are only preferred embodiments of the present invention, and the protection scope of the present invention is not limited to the above embodiments. All technical solutions under the concept of the present invention belong to the protection scope of the present invention. It should be pointed out that for ordinary technicians in this technical field, some improvements and modifications without departing from the principle of the present invention should be regarded as the protection scope of the present invention.

Claims

1. A multi-vessel multi-beam point cloud data matching method based on elevation correction, characterized in that: The specific steps include: Step 1: Multiple unmanned ships jointly collect point cloud data of underwater terrain, read the trajectory data of each unmanned ship, divide the route into sections according to the trajectory data, and save the trajectory point data within the route; Step 2: Divide all point cloud data into blocks and record the GPS time and flight segment of the point cloud data; Step 3: Divide the point cloud data in each block into sections, and perform segment correction and matching on the point cloud data in each section; Step 4: After completing the segment correction and matching of the point cloud data in each section, perform segment correction and matching of the point cloud data between adjacent sections; Step 5: After completing the segment correction and matching of the point cloud data between each section, perform segment correction and matching of the point cloud data between adjacent blocks to achieve the matching of multi-beam point cloud data of multiple unmanned ships.

2. The multi-vessel multi-beam point cloud data matching method based on elevation correction according to claim 1 is characterized in that: Step 1 includes the following sub-steps: Step 1.1, read the trajectory data of each unmanned ship, obtain the GPS time, position coordinates and azimuth of each point in each trajectory data, and calculate the change between adjacent azimuths; Step 1.2, taking the continuous trajectory data whose change between adjacent azimuths is less than a threshold as the candidate route of the unmanned ship; Step 1.3, the alternative segments of each unmanned ship are sorted in ascending order according to the minimum value of the GPS time in each alternative segment, and the alternative segments connected head to tail are merged to obtain the segment of each unmanned ship; Step 1.4: Number all the unmanned ship's segments, record the GPS time and location coordinates of each point in the segment, and save the track point data in the segment.

3. The multi-vessel multi-beam point cloud data matching method based on elevation correction according to claim 1 is characterized in that: Step 3 includes the following sub-steps: Step 3.1, divide the point cloud data in each block into sections, split the point cloud data in each section according to the flight segments they belong to, sort the flight segments to which the point cloud data belong in the order of overlapping point clouds from large to small in the split point cloud data, and use the horizontal and vertical coordinates of the split point cloud data to generate the XY plane projection triangulation of the flight segments; Step 3.2, find the trajectory point of the unmanned ship corresponding to each point cloud data in the section, calculate the weight of the point cloud data according to the distance and angle between the point cloud data and the trajectory point of the unmanned ship, and fit the weights of all the point cloud data in the section into a weighted quadratic surface; Step 3.3: According to the flight segment to which the sorted point cloud data belongs, the correction and matching of the flight segments within the segment is performed in combination with the triangulated network and weighted quadratic surface of the flight segment.

4. The multi-vessel multi-beam point cloud data matching method based on elevation correction according to claim 3 is characterized in that: The process of calculating the weight w of the point cloud data according to the distance and angle between the point cloud data and the trajectory point of the unmanned ship in step 3.2 is: Wherein, a represents the zenith angle between the point cloud and the trajectory point of the unmanned ship, and d represents the distance between the point cloud data and the trajectory point of the unmanned ship.

5. The multi-vessel multi-beam point cloud data matching method based on elevation correction according to claim 3 is characterized in that: Step 3.3 includes the following sub-steps: Step 3.3.

1. Take out the point cloud data C of the two corresponding flight segments in order A and C B , extract point cloud data C A Overlapping point cloud C in AO , calculate the overlapping point cloud C AO Each point cloud data in to point cloud data C B The distance of the XY plane projection triangulation network of the flight segment, the fitting error quadratic surface Q A ; Step 3.3.2, obtain point cloud data C A Each point in the error quadratic surface Q A The error D on the point cloud data C is obtained A Each point in the point cloud data C A The weight P on the quadratic surface A , get point cloud data C A Each point in the point cloud data C B The weight P on the quadratic surface B , calculate the point cloud data C A Corrected Z value of each point in The point cloud data C A Overlapping point cloud C AO The Z value of each point in is corrected by the corresponding Z value d z,A Make corrections; Step 3.3.3, obtain overlapping point cloud C AO Calculate the horizontal convex hull of point cloud data C A The shortest horizontal distance s from each non-overlapping point to the horizontal convex hull A , calculate the point cloud data C A Correction value of Z value of each non-overlapping point in The point cloud data C A The Z value of each non-overlapping point in is corrected by the corresponding Z value d f,A Make corrections; Step 3.3.4: Get point cloud data C B Each point in the error quadratic surface Q A The error D on the point cloud data C is obtained B Each point in the point cloud data C A The weight P on the quadratic surface A , get point cloud data C B Each point in the point cloud data C B The weight P on the quadratic surface B , calculate the point cloud data C A Corrected Z value of each point in The point cloud data C B Overlapping point cloud C AO The Z value of each point in is corrected by the corresponding Z value d z,B Make corrections; Step 3.3.5: Calculate point cloud data C B The shortest horizontal distance s from each non-overlapping point to the horizontal convex hull B , calculate the point cloud data C B Correction value of Z value of each non-overlapping point in The point cloud data C B The Z value of each non-overlapping point in is corrected by the corresponding Z value d f,B Make corrections; Step 3.3.6: Corrected point cloud data C A and C B Merge, the weight of each point after merging is P AB =P A +P B ; Step 3.3.7: Correct and match the merged point cloud data with the point cloud data of the next flight segment according to the method of steps 3.3.1 to 3.3.6, until all the point cloud data in each segment are corrected and matched to form a point cloud.

6. The multi-vessel multi-beam point cloud data matching method based on elevation correction according to claim 1 is characterized in that: Step 4 includes the following sub-steps: Step 4.1: Extract adjacent segment S A , S B The common side MN of the two sections is vertically extended to both sides of the common side MN according to the set section buffer length L, so as to obtain the buffer zone R of the two sections; Step 4.2: Extract the pre-correction segment S A Point cloud data C in buffer R AR , Segment S B Point cloud data C in buffer R BR , extract the corrected segment S A Point cloud data C' in buffer R AR , Segment S B Point cloud data C' in buffer R BR ; Step 4.3: Calculate the point cloud data C before correction AR And the corrected point cloud data C' AR , point cloud data before correction C BR And the corrected point cloud data C' BR Correction surface for the same flight segment k and Step 4.4: Obtain corrected point cloud data C' AR Each point in the corrected surface Correction value d Az1 , in the correction surface Correction value d Bz1 , the horizontal distance d to the common edge MN MN,AR , calculate the corrected point cloud data C' AR Corrected Z value of each point in The corrected point cloud data C' AR The Z value of each point in is corrected by the corresponding Z value d z,AR Make corrections; Step 4.5: Obtain corrected point cloud data C' BR Each point in the corrected surface Correction value d Az2 , in the corrected surface Correction value d Bz2 , obtain the corrected point cloud data C' BR The horizontal distance d from each point to the common edge MN MN,BR , calculate the corrected point cloud data C' BR Corrected Z value of each point in The corrected point cloud data C' BR The Z value of each point in is corrected by the corresponding Z value d z,BR Make corrections; Step 4.6: Corrected C' AR and C' BR Merge the segments and repeat steps 4.1-4.5 until the point cloud data of all segments in each block are corrected and matched to form a point cloud.

7. The multi-vessel multi-beam point cloud data matching method based on elevation correction according to claim 6 is characterized in that: Correct the surface in step 4.3 The calculation process is as follows: Get the point cloud data C before correction AR Point cloud data in flight segment k Get the corrected point cloud data C' AR Point cloud data in flight segment k Based on point cloud data Generate point cloud data by using the horizontal and vertical coordinates of each point XY plane projection triangulation Calculate point cloud data Projecting triangulated network from each point to XY plane The distance is fitted into the correction surface 8. The multi-vessel multi-beam point cloud data matching method based on elevation correction according to claim 1 is characterized in that: Step 5 includes the following sub-steps: Step 5.1: Extract adjacent block H A , H B The common edge PQ of the block is vertically extended to both sides of the common edge PQ according to the set block buffer length T to obtain the buffer zone F of the two segments; Step 5.2: Extract the pre-correction block H A Point cloud data C in buffer F AF 、Block H B Point cloud data C in buffer F BF , extract the corrected block H A Point cloud data C' in buffer F AF 、Block H B Point cloud data C' in buffer F BF ; Step 5.3: Calculate the point cloud data C before correction AF And the corrected point cloud data C' AF , point cloud data before correction C BF And the corrected point cloud data C' BF Correction surface for the same segment l and Step 5.4: Obtain corrected point cloud data C' AF Each point in the corrected surface Correction value d Al1 , obtain the corrected point cloud data C' BF In correcting the surface Correction value d Bl1 , obtain the corrected point cloud data C' AF The horizontal distance d from each point to the common edge PQ PQ,AF , calculate the corrected point cloud data C' AF Corrected Z value of each point in The corrected point cloud data C' AF The Z value of each point in is corrected by the corresponding Z value d z,AF Make corrections; Step 5.5: Obtain corrected point cloud data C' BF Each point in the corrected surface Correction value d Al2 , in the correction surface Correction value d Bl2 , obtain the corrected point cloud data C' BF The horizontal distance d from each point to the common edge PQ PQ,BF , calculate the corrected point cloud data C' BF Corrected Z value of each point in The corrected point cloud data C' BF The Z value of each point in is corrected by the corresponding Z value d z,BF Make corrections; Step 5.6: Corrected C' AF and C' BF Merge the blocks and repeat steps 5.1-5.5 until the point cloud data of all blocks are corrected and matched to form a point cloud.

9. A computer-readable storage medium storing a computer program, characterized in that: The computer program enables the computer to execute the multi-vessel multi-beam point cloud data matching method based on elevation correction as described in any one of claims 1 to 8.

10. An electronic device, characterized in that: include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the multi-vessel multi-beam point cloud data matching method based on elevation correction as described in any one of claims 1 to 8 is implemented.

Images