Terraced field reclamation optimization method and device based on unmanned aerial vehicle

Abstract

The invention relates to the technical field of terrace reclamation, in particular to an unmanned aerial vehicle-based terrace reclamation optimization method and device, and the method comprises the steps: carrying out the image shooting of a to-be-reclaimed terrace through employing an unmanned aerial vehicle, and obtaining the terrace image data of the to-be-reclaimed terrace; dEM data extraction is carried out on the terrace image data, and reclamation range information of the to-be-reclaimed terrace is determined; acquiring special condition information of a to-be-reclaimed terrace, and delimiting the special condition information as a reclamation constraint condition of the to-be-reclaimed terrace; a terrace reclamation three-dimensional model is constructed and finished, and a terrace reclamation face is generated; generating a control line according to the topographic condition of the to-be-reclaimed terrace, setting the control line as a control element of the to-be-reclaimed terrace, and marking the control element on a field map; and according to the terrace reclamation surface and the control line of the terrace to be reclaimed, rubber planting pits are delimited by combining the plant spacing and the planting preset density regulated by the rubber planting technology. According to the method, the three-dimensional model of the terraced field of the rubber garden can be quickly built, and auxiliary decision support is provided for construction and management of a rubber garden reclamation project.

Description

Technical Field

[0001] The present application relates to the technical field of terrace reclamation, and in particular to a terrace reclamation optimization method and device based on unmanned aerial vehicle (UAV). Background Art

[0002] With the continuous improvement of rubber planting and cultivation technology, new requirements for rubber plantation reclamation and planting have been put forward. It is urgent to adopt advanced reclamation planning technology to ensure the scientificity and rationality of rubber plantation reclamation to meet the needs of efficient, precise and environmentally friendly modern agriculture.

[0003] As an emerging surveying and mapping method, drone measurement technology has the characteristics of high efficiency, high precision, and high flexibility. First, by carrying high-definition cameras, lidar and other sensors, it can quickly obtain high-resolution ground image data and three-dimensional spatial data, providing accurate and comprehensive basic data support for the planning and design of rubber plantation reclamation. Secondly, drones can complete large-scale ground surveys and measurements in a short period of time, greatly improving work efficiency. Finally, based on the data measured by drones, a digital model of the rubber plantation can be established to achieve intelligent planning, design and decision support.

[0004] At present, the reclamation design of rubber plantations is mainly based on CAD-assisted manual calculation and demarcation, relying on manual measurement and ground survey. The whole process is cumbersome and lengthy, with high manpower, material and time costs. Especially in areas with complex terrain and lush vegetation, traditional methods often have problems such as low measurement accuracy, low work efficiency and high cost. In addition, in the process of rubber plantation planning, it is necessary to comprehensively consider a variety of factors, such as terrain, soil, climate, planting methods, etc., and there are great limitations in relying solely on manual operation and judgment. Therefore, the development of a rubber plantation reclamation planning and design method and tool based on drone measurement has important practical significance and broad market prospects for the construction of ecological rubber plantations. Summary of the invention

[0005] To achieve the above objectives, this application provides the following technical solutions:

[0006] According to a first aspect of the present invention, the present invention claims protection for a terrace reclamation optimization method based on a drone, comprising:

[0007] Using a drone to shoot images of the terraced fields to be reclaimed, and obtaining terraced field image data of the terraced fields to be reclaimed;

[0008] Extracting DEM data from the terrace image data, and determining the reclamation range information of the terrace to be reclaimed based on the slope information of the terrace to be reclaimed in the DEM data;

[0009] The drone is used to obtain special situation information of the terraced field to be reclaimed, and the special situation information is defined as a reclamation constraint condition for the terraced field to be reclaimed;

[0010] Constructing and trimming a three-dimensional model of terrace reclamation according to the reclamation constraint conditions and reclamation range information to generate a terrace reclamation surface;

[0011] Generate a control line according to the terrain conditions of the terrace to be reclaimed, set the control line as a control element of the terrace to be reclaimed, and mark it on the field map;

[0012] Rubber planting pits are delineated based on the terrace reclamation surface and control line of the terrace to be reclaimed, combined with the plant spacing and planting preset density in the rubber planting technical regulations.

[0013] Furthermore, the method of using a drone to take images of the terraced fields to be reclaimed to obtain terraced field image data of the terraced fields to be reclaimed also includes:

[0014] Analyze all historical terrace feature vectors in the terrace feature database and analyze the feature vectors;

[0015] Analyze the co-occurrence feature standard values ​​of all normal terrace standard feature vectors in the terrace feature library, and perform analysis on the co-occurrence features of the standard feature vectors;

[0016] Analyze the relevant characteristic information of the standard characteristic vector of normal terraces;

[0017] Performing correlation analysis on the relevant contents of the feature vector to be detected and the standard feature vector, wherein the basis for the correlation analysis is the identification engine terrace returned by the to-be-detected test and the target terrace connected in the standard feature vector;

[0018] Through data learning of normal drone positioning images, the features of normal drone positioning images are analyzed, and positive sample learning is performed on the features of the normal drone positioning images; abnormal drone positioning images generated by learning drones are analyzed, relevant features are analyzed, and negative sample learning is performed on the abnormal drone positioning images;

[0019] Perform classification configuration on the normal UAV positioning image features through a standardized method, combined with relevant reclamation equipment and environmental complexity, through a normalized standardized method;

[0020] The terrace decision value data is stored in the drone cache through the learning results;

[0021] The calculated terrace decision value data is used to perform feature analysis and terrace value calculation on the relevant UAV positioning image data in the current reclamation equipment. When the result is a negative number, the relevant UAV positioning image is considered to be a normal UAV positioning image, otherwise the relevant UAV positioning image is considered to be a terrace UAV positioning image. The absolute value of the relevant UAV positioning image is given as the correlation to assist the relevant terrace detection results.

[0022] Furthermore, the step of extracting DEM data from the terrace image data and determining the reclamation range information of the terrace to be reclaimed based on the slope information of the terrace to be reclaimed in the DEM data further includes:

[0023] Extracting DEM data from the terraced field image data by using an oblique photography model;

[0024] In the DEM data, based on the terrace slope and aspect information of the terrace to be reclaimed, the effective reclamation range of the terrace to be reclaimed is extracted, and the terrace to be reclaimed is classified by slope aspect.

[0025] Furthermore, the extracting the effective reclamation range of the terrace to be reclaimed according to the terrace slope aspect information of the terrace to be reclaimed in the DEM data, and classifying the slope aspect of the terrace to be reclaimed, further comprises:

[0026] Matrix-transferring the DEM data and performing peripheral expansion;

[0027] Constructing a pixel window, and corresponding each position of the matrix of the DEM data after peripheral expansion to the middle pixel of the pixel window;

[0028] In each pixel, the terrace slope information is calculated based on the pixel change rate in different directions;

[0029] Move the pixel window until all cells are traversed;

[0030] The terrace slope information of the terrace to be reclaimed is calculated based on the terrace slope information, and the terrace to be reclaimed is classified into south-facing slope and north-facing slope.

[0031] Furthermore, the using the drone to obtain special situation information of the terrace to be reclaimed, and defining the special situation information as a reclamation constraint condition for the terrace to be reclaimed, further includes:

[0032] The special situation information of the terraced fields to be reclaimed includes at least:

[0033] Boulders, steep slopes, dry ditches, and farm facilities.

[0034] Furthermore, the method of generating a control line according to the topographic conditions of the terrace to be reclaimed, setting the control line as a control element of the terrace to be reclaimed, and marking it on a field map further includes:

[0035] According to the terrain conditions of the terraced fields to be reclaimed, special marking points and control lines are generated at the valley line, ridge line, start and end line, and end point positions as control elements for terraced field reclamation, and are marked on the ground for reference by construction personnel;

[0036] Among them, a fusion algorithm based on hydrological analysis and curvature discrimination method is used to extract ridge lines and valley lines.

[0037] Furthermore, the method of extracting ridge lines and valley lines by using a fusion algorithm based on a hydrological analysis method and a curvature discrimination method further includes:

[0038] Use the created contour curvature calculation model or script to obtain the contour curvature;

[0039] The upper part of the hydrological analysis model is intercepted to obtain the storage grid data flow and positive and negative terrain;

[0040] Based on the contour line curvature, the initial ridge line curvature and initial valley line curvature are obtained using the conditional function and threshold;

[0041] Obtain initial ridgeline flow and valleyline flow based on hydrological analysis;

[0042] Based on the initial ridge lines and valley lines, the initial data are divided into 8 levels using the quantile method. The higher the level, the greater the possibility of correctness.

[0043] Add the stretched initial ridgeline curvature to the initial ridgeline flow data, and stretch the result to 0-1. The closer it is to 1, the more likely it is a ridgeline.

[0044] According to a second aspect of the present invention, the present invention claims protection for a terrace reclamation optimization device based on a drone, comprising:

[0045] one or more processors;

[0046] A memory having one or more programs stored thereon, when the one or more programs are executed by the one or more processors, the one or more processors implement the described terrace reclamation optimization method based on drones.

[0047] The present application relates to the technical field of terrace reclamation, and in particular to a terrace reclamation optimization method and device based on unmanned aerial vehicles, wherein an unmanned aerial vehicle is used to take images of terraces to be reclaimed, and terrace image data of the terraces to be reclaimed is obtained; DEM data is extracted from the terrace image data to determine the reclamation range information of the terraces to be reclaimed; special situation information of the terraces to be reclaimed is obtained, and the special situation information is delineated as the reclamation constraint conditions of the terraces to be reclaimed; a three-dimensional model of terrace reclamation is constructed and trimmed to generate a terrace reclamation surface; a control line is generated according to the terrain conditions of the terraces to be reclaimed, which is set as the control element of the terraces to be reclaimed and marked on a field map; according to the terrace reclamation surface and the control line of the terraces to be reclaimed, rubber planting pits are delineated in combination with the spacing and planting preset density of the rubber planting technical regulations. The present invention can quickly build a three-dimensional model of rubber plantation terraces, and provide auxiliary decision support for the construction and management of rubber plantation reclamation projects. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 A workflow diagram of a terrace reclamation optimization method based on drones as claimed in an embodiment of the present application;

[0049] Figure 2 A schematic diagram of a geological body model before reclamation of a terrace reclamation optimization method based on drones as claimed in an embodiment of the present application;

[0050] Figure 3 A schematic diagram of a step model after reclamation of a terrace reclamation optimization method based on drones as claimed in an embodiment of the present application;

[0051] Figure 4 A schematic diagram of a pixel window for a drone-based terrace reclamation optimization method claimed in an embodiment of the present application. DETAILED DESCRIPTION

[0052] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0053] The terms "first", "second" and "third" in this application are only used for descriptive purposes, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of indicated technical features. Thus, the features defined as "first", "second" and "third" can explicitly or implicitly include at least one of the features. In the description of this application, the meaning of "multiple" is at least two, such as two, three, etc., unless otherwise clearly and specifically defined. All directional indications in the embodiments of the present application (such as up, down, left, right, front, back...) are only used to explain the relative position relationship, movement, etc. between the components under a certain specific posture (as shown in the accompanying drawings). If the specific posture changes, the directional indication also changes accordingly. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but optionally also includes steps or units that are not listed, or optionally also includes other steps or units inherent to these processes, methods, products or devices.

[0054] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0055] According to the first embodiment of the present invention, the present invention claims a method for optimizing terrace reclamation based on drones, referring to Figure 1 ,include:

[0056] Using a drone to shoot images of the terraced fields to be reclaimed, and obtaining terraced field image data of the terraced fields to be reclaimed;

[0057] Extracting DEM data from the terrace image data, and determining the reclamation range information of the terrace to be reclaimed based on the slope information of the terrace to be reclaimed in the DEM data;

[0058] The drone is used to obtain special situation information of the terraced field to be reclaimed, and the special situation information is defined as a reclamation constraint condition for the terraced field to be reclaimed;

[0059] Constructing and trimming a three-dimensional model of terrace reclamation according to the reclamation constraint conditions and reclamation range information to generate a terrace reclamation surface;

[0060] Generate a control line according to the terrain conditions of the terrace to be reclaimed, set the control line as a control element of the terrace to be reclaimed, and mark it on the field map;

[0061] Rubber planting pits are delineated based on the terrace reclamation surface and control line of the terrace to be reclaimed, combined with the plant spacing and planting preset density in the rubber planting technical regulations.

[0062] In this embodiment, the following operations are performed in the three-dimensional modeling software ZGIS:

[0063] DEM generates geological model:

[0064] Construct virtual boreholes: simulate the geological structure;

[0065] Parameters: Select the generated DEM data, set the modeling accuracy (consistent with the generated DEM data accuracy: 12.5m), set the modeling depth (bottom truncation parameter), import virtual boreholes, import the range vector data before reclamation (range constraints) and generate Figure 2 The geological model shown;

[0066] The geological model generates a ladder model:

[0067] According to the requirements of terrace reclamation and construction, reclamation is carried out according to the site factors within the scope of the plot. The width of the terraces reclaimed for different slopes is different. For slopes below 15°, the width of the reclaimed terraces is 2.2m and the inclination is 6°; for slopes of 15°-25°, the width of the terraces is 1.8m and the inclination is 8°; for slopes of 26°-35°, the width of the terraces is 1.4m and the inclination is 8°. Excavation is carried out according to the geological model before wasteland reclamation. The main manifestation is the change of the surface DEM. Adjustments are made according to the height and width requirements of the terrace surface to finally form a terrace model.

[0068] Process DEM data according to the stepped design distribution range;

[0069] Formation of steps: Extract the contour lines of the DEM before reclamation, superimpose them with the step design range, extract the contour lines within the same step according to the two-dimensional distribution of the range, the elevation values ​​of the same step should be consistent, modify the elevation values ​​of the contour lines within the same range, and so on, to generate new contour line data;

[0070] Creation of TIN: Based on the newly generated contour data, select the Kriging interpolation method to generate irregular triangulated grid data (TIN);

[0071] Data conversion: TIN to raster data (digital elevation model DEM);

[0072] Finally get Figure 3 The stepped DEM data shown.

[0073] Furthermore, the method of using a drone to take images of the terraced fields to be reclaimed to obtain terraced field image data of the terraced fields to be reclaimed also includes:

[0074] Analyze all historical terrace feature vectors in the terrace feature database and analyze the feature vectors;

[0075] Analyze the co-occurrence feature standard values ​​of all normal terrace standard feature vectors in the terrace feature library, and perform analysis on the co-occurrence features of the standard feature vectors;

[0076] Analyze the relevant characteristic information of the standard characteristic vector of normal terraces;

[0077] Performing correlation analysis on the relevant contents of the feature vector to be detected and the standard feature vector, wherein the basis for the correlation analysis is the identification engine terrace returned by the to-be-detected test and the target terrace connected in the standard feature vector;

[0078] Through data learning of normal drone positioning images, the features of normal drone positioning images are analyzed, and positive sample learning is performed on the features of the normal drone positioning images; abnormal drone positioning images generated by learning drones are analyzed, relevant features are analyzed, and negative sample learning is performed on the abnormal drone positioning images;

[0079] Perform classification configuration on the normal UAV positioning image features through a standardized method, combined with relevant reclamation equipment and environmental complexity, through a normalized standardized method;

[0080] The terrace decision value data is stored in the drone cache through the learning results;

[0081] The calculated terrace decision value data is used to perform feature analysis and terrace value calculation on the relevant UAV positioning image data in the current reclamation equipment. When the result is a negative number, the relevant UAV positioning image is considered to be a normal UAV positioning image, otherwise the relevant UAV positioning image is considered to be a terrace UAV positioning image. The absolute value of the relevant UAV positioning image is given as the correlation to assist the relevant terrace detection results.

[0082] Furthermore, the step of extracting DEM data from the terrace image data and determining the reclamation range information of the terrace to be reclaimed based on the slope information of the terrace to be reclaimed in the DEM data further includes:

[0083] Extracting DEM data from the terraced field image data by using an oblique photography model;

[0084] In the DEM data, based on the terrace slope and aspect information of the terrace to be reclaimed, the effective reclamation range of the terrace to be reclaimed is extracted, and the terrace to be reclaimed is classified by slope aspect.

[0085] Among them, in this embodiment, when the DEM data is extracted from the terrace image data through the oblique photography model, the oblique photography model data generated by the drone aerial photography will carry elevation value information, and the elevation information is extracted;

[0086] By parsing the geographic information file, the coordinate string and the elevation value of the coordinate string point are obtained;

[0087] Coordinate string points generate discrete elevation points;

[0088] Interpolation methods (Kriging, inverse distance, etc.) are used to perform interpolation calculations to generate a digital elevation model with a data accuracy of 1m.

[0089] Furthermore, the extracting the effective reclamation range of the terrace to be reclaimed according to the terrace slope aspect information of the terrace to be reclaimed in the DEM data, and classifying the slope aspect of the terrace to be reclaimed, further comprises:

[0090] Matrix-transferring the DEM data and performing peripheral expansion;

[0091] Constructing a pixel window, and corresponding each position of the matrix of the DEM data after peripheral expansion to the middle pixel of the pixel window;

[0092] In each pixel, the terrace slope information is calculated based on the pixel change rate in different directions;

[0093] Move the pixel window until all cells are traversed;

[0094] The terrace slope information of the terrace to be reclaimed is calculated based on the terrace slope information, and the terrace to be reclaimed is classified into south-facing slope and north-facing slope.

[0095] In this embodiment, based on the slope aspects calculated previously, the slope aspects are reclassified into north slope (0-22.5°, 337.5-360°) and south slope (157.5-202.5°) according to the geographical orientation.

[0096] Calculation of slope: Calculated by the maximum mean method (Burrough, 1998).

[0097] Expand the periphery of the DEM data matrix by a circle matrix, whose value is the average value of all elevations;

[0098] Construct a 3×3 window, and the middle pixel corresponding to the window corresponds to the pixel at each position of the DEM.

[0099] Within each window, the maximum average method is used for calculation, as shown in formula (1):

[0100] Slope=arctan(√([dz / dx]2+[dz / dy]2))×180 / π (1)

[0101] Among them, dz / dx is the rate of change of pixel e in the x direction:

[0102] [dz / dx]=((c+2f+i)-(a+2d+g)) / (8×x_cellsize) (2)

[0103] dz / dy is the rate of change of pixel e in the y direction:

[0104] [dz / dy]=((g+2h+i)-(a+2b+c)) / (8×y_cellsize) (3)

[0105] x_cellsize and y_cellsize are the cell resolution in the x direction and the cell resolution in the y direction respectively.

[0106] Move the window until all cells are traversed.

[0107] (2) Calculate the slope aspect:

[0108]

[0109] Reference Figure 4 , the gray area is the outer expansion circle, and the white area is the DEM data range. This step is to process the edge points.

[0110] Pixel, also known as pixel point or pixel point, the cell here is a pixel (after the image is enlarged, it is also a pixel).

[0111] The window refers to building a 3×3 window with each pixel of the DEM data as the center. Starting from the upper left pixel of the data (that is, cell e), move from left to right until all pixels are traversed.

[0112] This window is used to calculate the slope of each pixel. For example, the window composed of abcdefghi can calculate the slope of pixel e through the slope calculation formula.

[0113] The eight-neighborhood method is mainly used for slope calculation. For each grid cell in the DEM data, take it as the center, select eight grid cells in the upper, lower, left, right and four diagonal directions around it, calculate the elevation difference of these eight grid cells, and calculate the slope value based on the elevation difference and the distance between the grid cells.

[0114] Furthermore, the using the drone to obtain special situation information of the terrace to be reclaimed, and defining the special situation information as a reclamation constraint condition for the terrace to be reclaimed, further includes:

[0115] The special situation information of the terraced fields to be reclaimed includes at least:

[0116] Boulders, steep slopes, dry ditches, and farm facilities.

[0117] Among them, in this embodiment, boulder identification: through the texture characteristics and color contrast of drone images, combined with the elevation mutation information in the DEM data, the distribution of boulders in the area is identified.

[0118] Steep slope identification: According to the slope information obtained from the DEM data, the slope threshold is set. When the slope is greater than 35°, it is considered a steep slope, and the steep slope areas in the area are identified.

[0119] Identification of farm buildings and facilities: Identify farm buildings and facilities in the area by using the building outlines and color features in drone images, combined with land use data in GIS or remote sensing classification results.

[0120] The identified boulders, steep slopes and farm facilities are used as constraints for terrace reclamation design and superimposed on the DEM data in the form of GIS layers to form restricted areas for terrace reclamation design.

[0121] Furthermore, the method of generating a control line according to the topographic conditions of the terrace to be reclaimed, setting the control line as a control element of the terrace to be reclaimed, and marking it on a field map further includes:

[0122] According to the terrain conditions of the terraced fields to be reclaimed, special marking points and control lines are generated at the valley line, ridge line, start and end line, and end point positions as control elements for terraced field reclamation, and are marked on the ground for reference by construction personnel;

[0123] Among them, a fusion algorithm based on hydrological analysis and curvature discrimination method is used to extract ridge lines and valley lines.

[0124] Furthermore, the method of extracting ridge lines and valley lines by using a fusion algorithm based on a hydrological analysis method and a curvature discrimination method further includes:

[0125] Use the created contour curvature calculation model or script to obtain the contour curvature;

[0126] The upper part of the hydrological analysis model is intercepted to obtain the storage grid data flow and positive and negative terrain;

[0127] Among them, a 3×3 window was selected to perform neighborhood analysis on the DEM data and calculate the average value.

[0128] The original DEM data is subtracted from the results of neighborhood analysis to obtain positive and negative terrain data.

[0129]

[0130] Reclassify the positive and negative terrain data twice.

[0131] Assign a value of 1 to areas greater than 0 (positive terrain) and a value of 0 to areas less than 0.

[0132] Assign a value of 1 to areas less than 0 (negative terrain) and a value of 0 to areas greater than 0.

[0133] Based on the contour line curvature, the initial ridge line curvature and initial valley line curvature are obtained using the conditional function and threshold;

[0134] Obtain initial ridgeline flow and valleyline flow based on hydrological analysis;

[0135] Based on the initial ridge lines and valley lines, the initial data are divided into 8 levels using the quantile method. The higher the level, the greater the possibility of correctness.

[0136] Add the stretched initial ridgeline curvature to the initial ridgeline flow data, and stretch the result to 0-1. The closer it is to 1, the more likely it is a ridgeline.

[0137] Among them, in this embodiment, ridgeline flow: is processed in arcgis.

[0138] Depression filling → flow direction calculation → flow calculation → extraction of zero value of backflow accumulation → data smoothing → mountain shadow auxiliary judgment → reclassification (determine the demarcation threshold as 0.5815) → data binarization (the part close to 1 is assigned to 1, and the rest is 0) → multiply the reclassified data with the positive terrain data → reclassification (assign 0 to attribute values ​​not 1) → obtain the final ridge line.

[0139] Valley line traffic: processed in ArcGIS.

[0140] Obtain inverse terrain → calculate flow direction (no need to fill depressions) → calculate flow rate → extract zero value of backflow accumulation → smooth data → auxiliary judgment of mountain shadow → reclassify (determine the demarcation threshold as 0.6142) → binarize data (assign 1 to the part close to 1 and 0 to the rest) → multiply the reclassified data with the negative terrain data → reclassify (assign 0 to attribute values ​​not equal to 1) → obtain the final ridgeline.

[0141] The extraction of ridge lines and valley lines adopts a fusion algorithm based on hydrological analysis and curvature discrimination: the two algorithms are used for weighted evaluation. The higher the score, the more accurate the extraction results are recognized by both algorithms, and the lower the score, the opposite is true.

[0142] Use the created contour curvature calculation model or script to obtain the contour curvature. The formula is as follows:

[0143]

[0144] Where c_paln is the mean curvature:

[0145]

[0146] Cut off the upper part of the hydrological analysis model to obtain the storage grid data (flow) and positive and negative terrain;

[0147] Based on the contour curvature, the initial ridge line curvature and initial valley line curvature are obtained using conditional functions and thresholds.

[0148] Obtain initial ridgeline flow and valleyline flow based on hydrological analysis;

[0149] Based on the initial ridge lines and valley lines in the previous step, the quantile method is used to divide the above initial data into 8 levels. The higher the level, the greater the possibility of correctness.

[0150] Add the stretched initial ridgeline curvature to the initial ridgeline flow data, and stretch the result to 0-1. The closer it is to 1, the more likely it is a ridgeline. The same is true for valley lines.

[0151] In this embodiment, based on the design results of rubber plantation terrace reclamation, combined with the rubber planting technical regulations, the plant spacing is 2.5m-3m, the row spacing is 8m-10m, and the planting density is 400 plants / hm2 on average. 2 -500 plants / hm 2 Rubber planting pits are demarcated according to the requirements.

[0152] According to a second embodiment of the present invention, the present invention claims protection for a terrace reclamation optimization device based on a drone, comprising:

[0153] one or more processors;

[0154] A memory having one or more programs stored thereon, when the one or more programs are executed by the one or more processors, the one or more processors implement the described terrace reclamation optimization method based on drones.

[0155] Compared with the prior art, the present invention applies UAV measurement technology to rubber plantation reclamation design, which can fully perceive complex scenes in a large-scale, high-precision and high-definition manner, and efficiently and automatically generate three-dimensional spatial scenes of terraces in combination with terrace reclamation design constraints, thereby improving the accuracy of terrace modeling, standardizing rubber plantation reclamation design, and providing auxiliary decision-making support for the early planning and design of terrace reclamation.

[0156] Based on drone measurement technology, large-scale geospatial data can be quickly obtained, and its high-precision sensors and image processing technology can generate high-precision digital elevation models. Compared with traditional manual measurement methods, drone measurement has greatly improved the efficiency and quality of data acquisition. At the same time, drone measurement can obtain three-dimensional spatial information of the rubber plantation reclamation area, including terrain, landforms, vegetation coverage, etc., to provide data support for the comprehensive planning and evaluation of rubber plantation reclamation design.

[0157] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0158] In addition, each functional unit in each embodiment of the present application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The above integrated unit can be implemented in the form of hardware or in the form of software functional units. The above is only an implementation method of the present application, and does not limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made using the contents of the specification and drawings of this application, or directly or indirectly used in other related technical fields, is also included in the patent protection scope of the present application.

[0159] The specific implementation methods of the invention are described in detail above, but they are only examples, and the present application is not limited to the specific implementation methods described above. For those skilled in the art, any equivalent modification or substitution of the invention is also within the scope of the present application, and therefore, the equalization, modification, and improvement made without departing from the spirit and principle of the present application should be included in the scope of the present application.

Claims

1. A terrace reclamation optimization method based on drones, characterized in that: include: Using a drone to shoot images of the terraced fields to be reclaimed, and obtaining terraced field image data of the terraced fields to be reclaimed; Extracting DEM data from the terrace image data, and determining the reclamation range information of the terrace to be reclaimed based on the slope information of the terrace to be reclaimed in the DEM data; The drone is used to obtain special situation information of the terraced field to be reclaimed, and the special situation information is defined as a reclamation constraint condition for the terraced field to be reclaimed; Constructing and trimming a three-dimensional model of terrace reclamation according to the reclamation constraint conditions and reclamation range information to generate a terrace reclamation surface; Generate a control line according to the terrain conditions of the terrace to be reclaimed, set the control line as a control element of the terrace to be reclaimed, and mark it on the field map; Rubber planting pits are delineated based on the terrace reclamation surface and control line of the terrace to be reclaimed, combined with the plant spacing and planting preset density in the rubber planting technical regulations.

2. The method for optimizing terrace reclamation based on drone according to claim 1, characterized in that: The step of using a drone to take images of the terraced fields to be reclaimed to obtain image data of the terraced fields to be reclaimed also includes: Analyze all historical terrace feature vectors in the terrace feature database and analyze the feature vectors; Analyze the co-occurrence feature standard values ​​of all normal terrace standard feature vectors in the terrace feature library, and perform analysis on the co-occurrence features of the standard feature vectors; Analyze the relevant characteristic information of the standard characteristic vector of normal terraces; Performing correlation analysis on the relevant contents of the feature vector to be detected and the standard feature vector, wherein the basis for the correlation analysis is the identification engine terrace returned by the to-be-detected test and the target terrace connected in the standard feature vector; Through data learning of normal drone positioning images, the features of normal drone positioning images are analyzed, and positive sample learning is performed on the features of the normal drone positioning images; abnormal drone positioning images generated by learning drones are analyzed, relevant features are analyzed, and negative sample learning is performed on the abnormal drone positioning images; Perform classification configuration on the normal UAV positioning image features through a standardized method, combined with relevant reclamation equipment and environmental complexity, through a normalized standardized method; The terrace decision value data is stored in the drone cache through the learning results; The calculated terrace decision value data is used to perform feature analysis and terrace value calculation on the relevant UAV positioning image data in the current reclamation equipment. When the result is a negative number, the relevant UAV positioning image is considered to be a normal UAV positioning image, otherwise the relevant UAV positioning image is considered to be a terrace UAV positioning image. The absolute value of the relevant UAV positioning image is given as the correlation to assist the relevant terrace detection results.

3. The method for optimizing terrace reclamation based on drone according to claim 1, characterized in that: Extracting DEM data from the terrace image data, and determining the reclamation range information of the terrace to be reclaimed according to the slope information of the terrace to be reclaimed in the DEM data, further comprising: Extracting DEM data from the terraced field image data by using an oblique photography model; The effective reclamation range of the terrace to be reclaimed is extracted based on the terrace slope and aspect information of the terrace to be reclaimed in the DEM data, and the terrace to be reclaimed is classified by slope aspect.

4. The method for optimizing terrace reclamation based on drone according to claim 3 is characterized in that: The step of extracting the effective reclamation range of the terrace to be reclaimed according to the terrace slope and aspect information of the terrace to be reclaimed in the DEM data, and classifying the slope aspects of the terrace to be reclaimed, further comprises: Matrix-transferring the DEM data and performing peripheral expansion; Constructing a pixel window, and corresponding each position of the matrix of the DEM data after peripheral expansion to the middle pixel of the pixel window; In each pixel, the terrace slope information is calculated based on the pixel change rate in different directions; Move the pixel window until all cells are traversed; The terrace slope information of the terrace to be reclaimed is calculated according to the terrace slope information, and the terrace to be reclaimed is classified into south-facing slope and north-facing slope.

5. The method for optimizing terrace reclamation based on drone according to claim 3, characterized in that: The method of using the drone to obtain special situation information of the terraced field to be reclaimed, and defining the special situation information as a reclamation constraint condition of the terraced field to be reclaimed, further includes: The special situation information of the terraced fields to be reclaimed includes at least: Boulders, steep slopes, and farm facilities.

6. The method for optimizing terrace reclamation based on drone according to claim 3, characterized in that: The step of generating a control line according to the terrain conditions of the terrace to be reclaimed, setting the control line as a control element of the terrace to be reclaimed, and marking the control line on a field map further includes: According to the terrain conditions of the terraced fields to be reclaimed, special marking points and control lines are generated at the valley line, ridge line, start and end line, and end point positions as control elements for terraced field reclamation, and are marked on the ground for reference by construction personnel; Among them, a fusion algorithm based on hydrological analysis and curvature discrimination method is used to extract ridge lines and valley lines.

7. The method for optimizing terrace reclamation based on drone according to claim 6, characterized in that: The method of extracting ridge lines and valley lines by using a fusion algorithm based on a hydrological analysis method and a curvature discrimination method also includes: Use the created contour curvature calculation model or script to obtain the contour curvature; The upper part of the hydrological analysis model is intercepted to obtain the storage grid data flow and positive and negative terrain; Based on the contour line curvature, the initial ridge line curvature and initial valley line curvature are obtained using the conditional function and threshold; Obtain initial ridgeline flow and valleyline flow based on hydrological analysis; Based on the initial ridge lines and valley lines, the initial data are divided into 8 levels using the quantile method. The higher the level, the greater the possibility of correctness. Add the stretched initial ridgeline curvature to the initial ridgeline flow data, and stretch the result to 0-1. The closer it is to 1, the more likely it is a ridgeline.

8. A terrace reclamation optimization device based on drones, characterized in that: include: one or more processors; A memory having one or more programs stored thereon, wherein when the one or more programs are executed by the one or more processors, the one or more processors implement a terrace reclamation optimization method based on a drone according to any one of claims 1 to 7.

Images