Unmanned aerial vehicle land surveying and mapping system and method
By carrying a laser calibration camera and GPS system on the drone, it automatically identifies and calculates the boundaries and area of cultivated land, solving the problem of low land surveying and mapping efficiency in the existing technology, achieving efficient and accurate land surveying and mapping and reporting generation, supporting agricultural management and Policy formulation.
Patent Information
- Application Number
- CN202411012602.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-26
- Publication Date
- 2024-11-08
AI Technical Summary
The existing land surveying and mapping methods mainly rely on manual handheld measurement tools, resulting in low work efficiency and high labor intensity, making it difficult to achieve efficient and accurate farmland surveying and mapping.
The drone is equipped with a laser calibration camera, combined with the GPS positioning system, and automatically identify the boundaries and crop types of cultivated land through image acquisition and laser ranging technology, calculate the actual cultivated land area, and generate a detailed surveying and mapping report.
It realizes the full-chain automation from data acquisition to result output, significantly improves the efficiency and quality of land surveying and mapping, ensures high accuracy and reliability of surveying and mapping results, and provides detailed data support for agricultural management and policy formulation.
Smart Images

Figure CN118913230A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of unmanned aerial vehicle (UAV) surveying and mapping, and in particular to an unmanned aerial vehicle (UAV) land surveying and mapping system and method. Background Art
[0002] Land surveying is a work that comprehensively applies surveying principles and modern technical means to measure, analyze and express geographical information such as topography, location, shape, size, height, etc. of the surface and underground space. It involves collecting geospatial data, including terrain features such as boundaries, buildings, roads, water systems, vegetation, etc., obtaining data through ground surveying, aerial photogrammetry, satellite remote sensing and other methods, and then using high-tech means such as GIS (Geographic Information System), GPS (Global Positioning System) and RS (Remote Sensing Technology) for processing, mapping and management.
[0003] Cultivated land surveying and mapping refers to the process of accurately measuring and mapping cultivated land areas using surveying and mapping techniques and methods, aiming to obtain multi-dimensional information such as the spatial distribution, area, shape, soil type, land slope, irrigation conditions, etc. This process is of great significance for rationally planning land use, improving agricultural production efficiency, monitoring land degradation, implementing cultivated land protection policies, and evaluating agricultural ecological conditions.
[0004] The existing surveying and mapping methods mainly use manual handheld measuring tools for measurement, which has low work efficiency and high labor intensity. Summary of the invention
[0005] The purpose of the present invention is to provide a UAV land surveying and mapping system and method, which aims to realize the automation of the entire chain from data collection, processing to output of results, thereby greatly improving the efficiency and quality of land surveying and mapping.
[0006] To achieve the above-mentioned object, in a first aspect, the present invention provides a UAV land surveying and mapping system, comprising a UAV platform, a laser calibration camera, a cultivated land identification module, an area calculation module and a report generation module;
[0007] The UAV platform is used to carry a laser calibration camera for aerial photography;
[0008] The laser calibration camera is used to calibrate the distance of the cultivated land accurately and collect ground images using a laser-assisted camera;
[0009] The farmland identification module is used to identify farmland boundaries and crop types from captured images;
[0010] The area calculation module is used to convert the cultivated land area identified in the image into the actual cultivated land area based on the cultivated land boundary and the distance data measured by the laser in combination with the camera parameters;
[0011] The report generation module is used to generate a surveying and mapping report based on the positioning information and the actual cultivated land area.
[0012] Wherein, the UAV platform includes a UAV unit, a route setting unit and a GPS positioning unit;
[0013] The route setting unit is used to plan the flight route based on the surveyed cultivated land area;
[0014] The drone unit is used to perform flight actions based on the flight route;
[0015] The GPS positioning unit is used to record the flight trajectory and shooting position coordinates of the drone in real time.
[0016] Wherein, the laser calibration camera includes a calibration unit, a distance measurement unit, an image acquisition unit and a synchronization unit;
[0017] The calibration unit is used to perform time synchronization and optical alignment on the camera and the laser;
[0018] The distance measuring unit is used to start the laser to emit laser pulses to the ground to be measured and receive echo signals;
[0019] The image acquisition unit is used to continuously capture ground images;
[0020] The synchronization unit is used to synchronize the laser distance measurement data with the ground image so that each frame of the image has corresponding distance information.
[0021] The farmland identification module includes an image acquisition unit, an image correction unit, an image enhancement unit, a vegetation segmentation unit, a crop feature extraction unit, a model training unit and a farmland identification unit;
[0022] The image acquisition unit is used to collect ground images taken by the drone;
[0023] The image correction unit is used to perform radiation correction on the ground image to eliminate atmospheric effects, and to perform geometric correction to ensure the geographic accuracy of the image;
[0024] The image enhancement unit is used to improve image quality through histogram equalization and contrast enhancement;
[0025] The vegetation segmentation unit is used to set a threshold based on the normalized difference vegetation index to perform preliminary segmentation to obtain a vegetation area image;
[0026] The crop feature extraction unit is used to extract crop features from the vegetation area image, and the crop features include spectral features, texture features and spatial layout features;
[0027] The model training unit is used to train a support vector machine using crop features and a training set;
[0028] The model training unit is used to identify the cultivated land in the vegetation area image based on the support vector machine to obtain the cultivated land boundary and the corresponding crop type.
[0029] Wherein, the area calculation module includes a camera calibration unit, a distance acquisition unit, a measurement unit, a mapping unit and an area calculation unit;
[0030] The camera calibration unit is used to perform lens distortion correction and geometric calibration on the camera, and use a calibration object of known size to determine the intrinsic and extrinsic parameters of the camera;
[0031] The distance acquisition unit is used to acquire the distance data from multiple points provided by the laser to the camera, and synchronize with the moment when the camera captures the image;
[0032] The measuring unit is used to calculate the point cloud data of the cultivated land in three-dimensional space by combining the camera parameters and the distance data using the triangulation principle;
[0033] The mapping unit is used to project the three-dimensional point cloud data back to the two-dimensional image plane;
[0034] The area calculation unit is used to perform polygonal approximation on the mapped cultivated land area boundary and calculate the actual coverage area of the cultivated land using a polygonal area calculation formula.
[0035] Wherein, the area calculation module further includes a correction unit, and the correction unit is used to compensate the calculated actual coverage area.
[0036] Wherein, the report generation module includes an information summary unit, a classification unit and a visualization unit;
[0037] The summarizing unit is used to summarize the surveying and mapping information of all cultivated lands;
[0038] The classification unit is used to classify and count the total cultivated land area by region and crop type;
[0039] The visualization unit is used to perform visualization processing on the data using a data visualization tool.
[0040] Wherein, the vegetation segmentation unit includes an analysis subunit, an iterative optimization unit and a segmentation subunit;
[0041] The analysis subunit is used to find the peaks and valleys of the NDVI value distribution by using histogram analysis, and select the valley points as thresholds;
[0042] The iterative optimization subunit is used to maximize the variance between categories using the Otsu method;
[0043] The segmentation subunit is used to perform binarization processing on the NDVI image using a selected threshold value, marking pixels with NDVI values higher than the threshold value as vegetation pixels, and marking pixels with NDVI values lower than the threshold value as non-vegetation pixels.
[0044] In a second aspect, the present invention further provides a method for UAV land surveying and mapping, comprising: the UAV platform is equipped with a laser calibration camera for aerial photography;
[0045] Use laser-assisted cameras to accurately calibrate the distance of cultivated land and collect ground images;
[0046] Identify farmland boundaries and crop types from captured images;
[0047] Based on the cultivated land boundary and the distance data measured by the laser, the cultivated land area identified in the image is converted into the actual cultivated land area in combination with the camera parameters;
[0048] Generate surveying and mapping reports based on positioning information and actual cultivated land area.
[0049] The invention discloses a land surveying system and method for an unmanned aerial vehicle. The unmanned aerial vehicle platform flies autonomously or remotely according to a preset flight path. Its high flexibility ensures comprehensive coverage and detailed shooting of various landforms. Laser calibration camera: installed under the unmanned aerial vehicle, this is a composite sensor that integrates a high-definition camera and a precision laser ranging system. The camera continuously shoots ground images during the flight of the unmanned aerial vehicle, and the laser emits a light beam to accurately mark the ground feature points, thereby realizing the three-dimensional spatial precise positioning of the shooting area. The cultivated land recognition module uses an image segmentation algorithm to identify and outline the boundaries of the cultivated land and distinguish between cultivated and uncultivated areas. Subsequently, various crop types such as rice, wheat, cotton, etc. are automatically classified and identified through pattern recognition of color, texture and morphological features, providing refined information for agricultural production planning. The area calculation module uses the precise distance data obtained from the laser calibration camera, combined with the GPS positioning information of the unmanned aerial vehicle and the internal parameters of the camera (such as focal length and field of view angle). The module uses geometric calculation to map the cultivated land boundaries identified in the two-dimensional image to the real three-dimensional space, thereby accurately calculating the actual area of each piece of cultivated land. This process ensures the high accuracy and reliability of the surveying and mapping results. The report generation module is responsible for summarizing all surveying and mapping data and automatically generating structured and visual surveying and mapping reports. The report not only contains key information such as the location coordinates, precise area, and crop type distribution of each piece of cultivated land, but may also incorporate historical data comparative analysis, crop health assessment, and land suitability evaluation. The reports are in various forms, which are suitable for in-depth research by professional and technical personnel, and also facilitate government managers and farmers to quickly understand the status of land resources, and provide data support for the formulation of scientific land management and agricultural policies. In summary, this drone land surveying and mapping system has achieved full-chain automation from data collection, processing to output of results through integrated software and hardware solutions, greatly improving the efficiency and quality of land surveying and mapping. BRIEF DESCRIPTION OF THE DRAWINGS
[0050] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0051] Figure 1 It is a structural diagram of an unmanned aerial vehicle land surveying and mapping system according to the first embodiment of the present invention.
[0052] Figure 2 It is a structural diagram of the UAV platform of the second embodiment of the present invention.
[0053] Figure 3 It is a structural diagram of a laser calibration camera according to the second embodiment of the present invention.
[0054] Figure 4 It is a structural diagram of a cultivated land identification module according to the second embodiment of the present invention.
[0055] Figure 5 It is a structural diagram of the area calculation module of the second embodiment of the present invention.
[0056] Figure 6 It is a structural diagram of a report generation module according to a second embodiment of the present invention.
[0057] Figure 7 It is a structural diagram of a vegetation segmentation unit according to a second embodiment of the present invention.
[0058] Figure 8 It is a flow chart of a UAV land surveying method according to the third embodiment of the present invention.
[0059] UAV platform 101, laser calibration camera 102, cultivated land identification module 103, area calculation module 104, report generation module 105, UAV unit 201, route setting unit 202, GPS positioning unit 203, calibration unit 204, distance measurement unit 205, image acquisition unit 206, synchronization unit 207, image acquisition unit 208, image correction unit 209, image enhancement unit 210, vegetation segmentation unit 211, crop feature extraction unit 212, model training unit 213, cultivated land identification unit 214, camera calibration unit 215, distance acquisition unit 216, measurement unit 217, mapping unit 218, area calculation unit 219, correction unit 220, information summary unit 221, classification unit 222, visualization unit 223, analysis subunit 224, iterative optimization unit 225, segmentation subunit 226. DETAILED DESCRIPTION
[0060] Embodiments of the present invention are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present invention, and should not be construed as limiting the present invention.
[0061] First embodiment
[0062] See also Figure 1The present invention provides a UAV land surveying and mapping system, comprising a UAV platform 101, a laser calibration camera 102, a cultivated land identification module 103, an area calculation module 104 and a report generation module 105; the UAV platform 101 is used to carry the laser calibration camera 102 for aerial photography; the laser calibration camera 102 is used to use a laser to assist the camera to accurately calibrate the distance of cultivated land and collect ground images; the cultivated land identification module 103 is used to identify cultivated land boundaries and crop types from captured images; the area calculation module 104 is used to convert the cultivated land area identified in the image into the actual cultivated land area based on the cultivated land boundary and the distance data measured by the laser in combination with the camera parameters; the report generation module 105 is used to generate a surveying and mapping report based on positioning information and the actual cultivated land area.
[0063] In this embodiment, the drone platform 101 is used as the carrier of the system, and a high-performance, long-endurance professional drone is selected, which has stable flight and high-precision GPS positioning functions. It can not only adapt to a variety of complex terrain environments, but also can autonomously perform tasks according to the preset flight path to ensure that every inch of land in the measurement area is covered. The drone platform 101 is specially designed with a sturdy and stable load mounting system, which is specifically used to carry the laser calibration camera 102 to ensure the clarity and stability of image acquisition under high-speed flight and various climatic conditions. The laser calibration camera 102 integrates a high-resolution camera and laser ranging technology, which can emit a laser beam to the ground during high-altitude operations, and capture the reflected beam through the camera to achieve accurate distance measurement of the target area. This process not only improves the accuracy of the measurement, but also can collect high-quality ground images under different lighting conditions, providing rich visual information for subsequent image analysis. The application of laser calibration technology enables the camera to more accurately identify and lock the boundaries of cultivated land, avoiding the visual errors in traditional methods.
[0064] The cultivated land identification module 103 uses deep learning and computer vision technology. This module can quickly identify and segment the cultivated land boundaries from the massive image data collected by the drone, and can also identify the crop type according to the texture, color and other characteristics of the crop. This not only helps to accurately distinguish the types of crops, but also provides basic data for subsequent crop health monitoring and pest warning. The area calculation module 104 combines the actual distance data obtained by laser ranging with the pixel size of the camera image, and uses geometric principles and image processing algorithms. This module can accurately calculate the actual area of the cultivated land. This process takes into account the influence of factors such as terrain undulations and lens distortion, ensuring the high accuracy of the calculation results. In addition, the module can further refine the area estimation according to the planting density of different crops to meet more detailed agricultural management needs. The report generation module 105 is based on the real-time positioning information of the drone, the calculated actual cultivated land area and the crop identification results. This module automatically generates a detailed surveying and mapping report. The report not only includes key data such as total cultivated land area, the proportion of various crop planting areas, and the coordinates of plot locations, but may also integrate GIS map services to display surveying and mapping results in an intuitive graphical interface, making it easier for decision makers and agricultural workers to quickly understand and use this information to optimize resource allocation, adjust crop layout, and predict yields. In summary, the drone land surveying system has greatly improved the efficiency and accuracy of land surveying through highly automated and intelligent technical means, providing strong support for modern agricultural management and sustainable development.
[0065] Second embodiment
[0066] See also Figure 2 to Figure 7 On the basis of the first embodiment, the present invention further provides a UAV land surveying and mapping system, wherein the UAV platform 101 includes a UAV unit 201, a route setting unit 202 and a GPS positioning unit 203; the route setting unit 202 is used to plan a flight route based on the surveyed cultivated land area; the UAV unit 201 is used to perform flight actions based on the flight route; the GPS positioning unit 203 is used to record the UAV flight trajectory and shooting position coordinates in real time.
[0067] The drone unit 201 is equipped with a sensor array, including but not limited to a barometer, a gyroscope, an accelerometer, etc., to ensure stable flight performance under complex meteorological conditions. In addition, the drone unit 201 is also equipped with a high-capacity battery and an intelligent power management system to ensure long-term aerial operation capabilities, sufficient to cover a large area of surveying and mapping. After receiving the flight instructions planned by the route setting unit 202, the drone unit 201 can automatically adjust the flight attitude, speed and altitude, and fly accurately along the predetermined route. At the same time, the high-definition camera or other professional surveying and mapping equipment carried by it continuously shoots or scans the ground. The route setting unit 202 is responsible for planning the most optimized flight route before the surveying and mapping task begins, based on the geographical information, terrain characteristics and specific surveying and mapping needs of the cultivated area. Using GIS (geographic information system) technology and algorithms, the route setting unit 202 can avoid obstacles to ensure that the drone can fully and completely cover all target areas when performing tasks. The GPS positioning unit 203 is integrated into the core control system of the drone platform 101. The GPS positioning unit 203 uses the global satellite navigation system to achieve centimeter-level precise positioning. It not only tracks and records the three-dimensional spatial position of the drone in real time to ensure the accuracy of the flight trajectory, but also synchronously records the specific geographic coordinates of each image taken.
[0068] The laser calibration camera 102 includes a calibration unit 204, a distance measuring unit 205, an image acquisition unit 206 and a synchronization unit 207; the calibration unit 204 is used to perform time synchronization and optical alignment on the camera and the laser; the distance measuring unit 205 is used to start the laser to emit laser pulses to the ground to be measured and receive echo signals; the image acquisition unit 206 is used to continuously capture ground images; the synchronization unit 207 is used to synchronize the laser distance measurement data with the ground image so that each frame of the image has corresponding distance information.
[0069] The calibration unit 204 realizes the time synchronization between the camera and the laser, ensures the time consistency during data collection, and completes the optical alignment task, that is, to ensure that the laser beam path is well geometrically aligned with the camera's visual axis, which is the cornerstone of the subsequent data fusion and analysis accuracy. This process may include automatic or manual correction using specific reference marks to ensure that the measurement error is minimized. The ranging unit 205 is the core part of the system that directly participates in physical measurement. It is responsible for controlling the laser to emit short pulses of laser light to the surface of the ground to be measured. The echo signals formed by these laser pulses after reflection will be captured by a highly sensitive receiver. By accurately measuring the round-trip time of the laser pulse and combining it with the speed of light, the ranging unit 205 can calculate the precise distance from the laser to the ground. The image acquisition unit 206 continuously captures the visual information of the ground and generates a continuous sequence of images. These images not only provide rich texture information, but also can analyze the dynamic changes of the surface in combination with the time series. This unit is usually equipped with a high-resolution CMOS or CCD sensor, which can capture clear and detailed images under different lighting conditions, providing basic materials for subsequent image analysis and processing. The synchronization unit 207 ensures that the laser distance measurement data is seamlessly connected with the ground image at the corresponding moment. The synchronization unit 207 can assign the corresponding accurate distance information to each frame of the image, forming image data with depth information. This synchronization not only improves the efficiency of data processing, but also greatly enhances the practicality and accuracy of the final measurement results, allowing users to intuitively understand and analyze the spatial characteristics of the ground.
[0070] The cultivated land identification module 103 includes an image acquisition unit 208, an image correction unit 209, an image enhancement unit 210, a vegetation segmentation unit 211, a crop feature extraction unit 212, a model training unit 213 and a cultivated land identification unit 214; the image acquisition unit 208 is used to collect ground images taken by a drone; the image correction unit 209 is used to perform radiation correction on the ground image to eliminate atmospheric effects, and perform geometric correction to ensure the geographic accuracy of the image; the image enhancement unit 210 is used to improve the image quality through histogram equalization and contrast enhancement; the vegetation segmentation unit 211 is used to set a threshold based on the normalized difference vegetation index for preliminary segmentation to obtain a vegetation area image; the crop feature extraction unit 212 is used to extract crop features from the vegetation area image, and the crop features include spectral features, texture features and spatial layout features; the model training unit 213 is used to train a support vector machine using crop features and a training set; the model training unit 213 is used to identify the cultivated land in the vegetation area image based on the support vector machine to obtain the cultivated land boundary and the corresponding crop type.
[0071] The image acquisition unit 208 is directly connected to the image sensing system of the drone to automatically collect high-definition ground images taken on the preset flight route. These images contain rich ground information and are the basic data for subsequent analysis. The image correction unit 209 first performs radiation correction on the image to eliminate image distortion caused by factors such as atmospheric scattering and absorption, and restore the true reflection of the surface. Then, through geometric correction, the deformation caused by factors such as shooting angle and terrain undulation is corrected, so that the pixel points on the image can accurately correspond to the actual geographical location, thereby improving the geographical accuracy of subsequent analysis. Considering the diversity of actual shooting conditions, the image may be affected by problems such as uneven lighting and low contrast. The image enhancement unit 210 uses histogram equalization and contrast enhancement technology to effectively improve the overall brightness, contrast and detail performance of the image, making features such as vegetation and soil more clear and discernible, and providing better quality image materials for subsequent processing. The vegetation segmentation unit 211 uses the Normalized Difference Vegetation Index (NDVI), an indicator widely used in vegetation detection, to set a reasonable threshold, preliminarily distinguish the vegetation area from the non-vegetation area from the corrected and enhanced image, further refine the analysis scope, and focus on the area with agricultural significance. The crop feature extraction unit 212 deeply analyzes the vegetation area image and extracts a variety of key features of the crop. This includes the spectral characteristics of the crop (such as the reflectivity of different bands), which reflects the growth status of the crop; texture characteristics, which reveal information such as crop density and health; and spatial layout characteristics, which help understand the crop planting pattern. These features comprehensively reflect the biophysical characteristics and growth environment of the crop. The model training unit 213 uses machine learning algorithms such as support vector machines (SVM) for model training based on the previously extracted crop features and combined with the existing training data set. SVM is often used in complex pattern recognition tasks due to its excellent classification ability and generalization performance. Through a large number of iterations and optimizations, a prediction model that can accurately identify different types of crops is constructed.
[0072] The cultivated land identification unit 214 uses the trained support vector machine model to perform in-depth analysis on the previously segmented vegetation area, not only identifying the specific cultivated land boundaries, but also further determining the types of crops in the cultivated land. This process is the ultimate goal of the entire module, providing data support for agricultural resource management.
[0073] The area calculation module 104 includes a camera calibration unit 215, a distance acquisition unit 216, a measurement unit 217, a mapping unit 218 and an area calculation unit 219; the camera calibration unit 215 is used to perform lens distortion correction and geometric calibration on the camera, and use a calibration object of known size to determine the intrinsic and extrinsic parameters of the camera; the distance acquisition unit 216 is used to obtain the distance data from multiple points to the camera provided by the laser, and synchronize it with the moment when the camera captures the image; the measurement unit 217 is used to combine the camera parameters and the distance data, and use the triangulation principle to calculate the point cloud data of the cultivated land in the three-dimensional space; the mapping unit 218 is used to project the three-dimensional point cloud data back to the two-dimensional image plane; the area calculation unit 219 is used to perform polygonal approximation on the mapped cultivated land area boundary, and apply the polygonal area calculation formula to calculate the actual coverage area of the cultivated land.
[0074] The area calculation module 104 further includes a correction unit 220, and the correction unit 220 is used to compensate the calculated actual coverage area.
[0075] The camera calibration unit 215 ensures the geometric accuracy of the captured image by performing detailed lens distortion correction on the camera to eliminate common problems such as barrel or pincushion distortion. In addition, by using calibration objects of known size such as calibration plates, the unit can accurately measure the internal parameters (such as focal length, principal point coordinates) and external parameters (such as position and direction) of the camera, laying a solid foundation for subsequent three-dimensional reconstruction. The distance acquisition unit 216 is responsible for working closely with the laser ranging system to collect the exact distance information of different ground points pointed by the laser relative to the camera in real time. These data not only require high precision, but also need to be strictly synchronized with the moment when the camera captures the image, ensuring that each pixel point can be associated with an exact three-dimensional spatial position, thereby constructing an accurate distance model of the cultivated land surface. Based on the internal and external parameters of the camera and the data from the distance acquisition unit 216, the measurement unit 217 uses triangulation to convert the pixel points on the two-dimensional image into coordinate points in three-dimensional space, generating dense cultivated land cloud data. This process not only reveals the subtle ups and downs of the cultivated land surface, but also provides rich geometric information for subsequent area calculations. The function of the mapping unit 218 is to convert complex three-dimensional point cloud data into an easy-to-process two-dimensional image plane representation. This conversion process involves perspective projection or other mapping techniques, which ensures the faithful reproduction of ground features and also prepares suitable graphic data for area calculation. The area calculation unit 219 uses an image processing algorithm to identify and extract the boundary contour of the cultivated land area, and then simplifies the boundary shape through polygonal approximation technology, and then accurately calculates the actual coverage area of the cultivated land based on the polygonal area calculation formula. The correction unit 220, as a newly added key link, aims to further improve the reliability of area calculation. It can compensate for minor deviations that may occur during the calculation process or measurement errors caused by environmental factors (such as shadows, crop occlusions), and by introducing environmental models, historical data comparisons or other correction algorithms, the final area value is closer to the actual situation and meets the high-standard agricultural surveying and mapping needs.
[0076] The report generation module 105 includes an information aggregation unit 221, a classification unit 222 and a visualization unit 223; the aggregation unit is used to aggregate the surveying and mapping information of all cultivated land; the classification unit 222 is used to classify and count the total area of cultivated land by region and crop type; the visualization unit 223 is used to visualize the data using a data visualization tool.
[0077] The information aggregation unit 221 extracts all surveying and mapping information related to cultivated land from multi-dimensional and multi-source systems, including but not limited to cultivated land location coordinates, land quality assessment, soil composition analysis, historical farming records, etc. At this stage, the module will use data integration technology to ensure the integrity and consistency of all information, forming a comprehensive cultivated land information database, laying a solid foundation for subsequent analysis.
[0078] The classification unit 222 is responsible for logical classification and statistical analysis of the massive amount of information collected. The core of its work is to classify the cultivated land in detail according to preset standards or policy requirements, such as administrative divisions, crop types, etc. This unit will calculate the total area of cultivated land under each classification. Finally, the visualization unit 223 converts the above analysis results into charts, maps, dashboards, etc., and uses modern data visualization tools such as Tableau, Power BI or custom-developed graphical interfaces to intuitively display the various complex relationships and trends of cultivated land data.
[0079] The vegetation segmentation unit 211 includes an analysis subunit 224, an iterative optimization unit 225 and a segmentation subunit 226; the analysis subunit 224 is used to find the peaks and valleys of the NDVI value distribution by using histogram analysis, and select the valley points as thresholds; the iterative optimization subunit is used to use the Otsu method to maximize the variance between categories; the segmentation subunit 226 is used to binarize the NDVI image using the selected threshold, and mark pixels with NDVI values higher than the threshold as vegetation pixels, and pixels with NDVI values lower than the threshold as non-vegetation pixels.
[0080] The analysis subunit 224 further explores the distribution characteristics of the normalized difference vegetation index (NDVI) through histogram analysis. As an indicator widely used to evaluate the growth status of vegetation, the distribution of its value range can often intuitively reflect the degree and changes of vegetation coverage. The analysis subunit 224 cleverly identifies the natural boundary between vegetation and non-vegetation areas by conducting a comprehensive statistical analysis of the NDVI values, especially finding peaks and valleys in the constructed histogram. The valley point is selected as the preliminary segmentation threshold, which lays a precise foundation for subsequent processing. The iterative optimization subunit takes over the baton and introduces the classic Otsu method, which is an adaptive threshold selection technology that aims to iteratively calculate the inter-class variance of foreground and background pixels under different thresholds, and select the threshold that makes the variance reach the maximum value as the optimal segmentation condition. This dynamic optimization process significantly enhances the robustness of vegetation segmentation, and can maintain a high classification accuracy even in the face of complex lighting conditions or terrain changes, reducing the risk of misclassification. Finally, the segmentation subunit 226 performs an accurate binarization operation based on the optimal threshold generated by the collaboration of the first two subunits. In this stage, all pixels whose NDVI values exceed the threshold are clearly marked as pixels representing vegetation, and are assigned a specific color or value (usually white or high brightness value), while those pixels below the threshold are correspondingly marked as non-vegetation areas (expressed as black or low brightness value). Through this step, the original complex NDVI image is converted into a clearly identifiable binary image, which intuitively shows the vegetation distribution pattern in the study area.
[0081] Third embodiment
[0082] See also Figure 8 Based on the second embodiment, the present invention further provides a UAV land surveying and mapping method, comprising:
[0083] S301: The UAV platform 101 carries out aerial photography with a laser calibration camera 102;
[0084] A high-performance UAV is used as the flight platform, which is equipped with a high-resolution laser calibration camera 102. The UAV flies autonomously or remotely according to a preset route, and the camera continuously captures ground images during the flight. The laser calibration function ensures that high-quality and high-precision image data can be obtained even under complex lighting conditions, laying a solid foundation for subsequent analysis.
[0085] S302: Using a laser-assisted camera to accurately calibrate the distance of the cultivated land and collect ground images;
[0086] To further improve the accuracy of surveying and mapping, the laser on the drone can directly illuminate specific points on the ground and simultaneously record the positions of these beams in the image. This process not only helps correct image distortion caused by changes in flight altitude, but also provides accurate distance calibration for each frame of the image, so that each pixel in the image can correspond to an exact position in real space. At the same time, this link also completes the detailed image acquisition of the surface features of the cultivated land.
[0087] S303: Identify farmland boundaries and crop types from the captured images;
[0088] The image processing algorithms and machine learning techniques are used to conduct in-depth analysis of the collected image data. First, the specific boundaries of cultivated land are identified through edge detection and morphological operations, distinguishing cultivated land from other non-cultivated areas. Secondly, by learning and comparing features such as color, texture, and shape, different types of crops, such as wheat, corn, and soybeans, are automatically identified, achieving accurate identification of crop planting distribution.
[0089] S304: based on the cultivated land boundary and the distance data measured by the laser, the cultivated land area identified in the image is converted into the actual cultivated land area in combination with the camera parameters;
[0090] Combining the GPS positioning information, laser ranging data, and the focal length, field of view and other parameters of the camera, the cultivated land area in the two-dimensional image is accurately converted into the three-dimensional area in the real world through geometric perspective transformation and three-dimensional reconstruction technology. This conversion process ensures the high accuracy and reliability of the surveying and mapping results, providing an important basis for the scientific management of land resources.
[0091] S305: Generate a surveying and mapping report based on the positioning information and the actual cultivated land area.
[0092] Finally, all the collected and analyzed data are integrated into a detailed mapping report. The report not only contains basic information such as the precise geographical location, actual area, and crop type distribution of each piece of cultivated land, but may also include time series analysis, crop growth status assessment, land suitability evaluation, etc. This report provides agricultural decision makers, researchers, and local governments with a comprehensive and easy-to-understand overview of land use through intuitive charts, maps, and text descriptions, providing strong data support for the formulation of strategies to promote agricultural modernization and sustainable development.
[0093] What is disclosed above is only a preferred embodiment of the present invention, and it certainly cannot be used to limit the scope of rights of the present invention. Ordinary technicians in this field can understand that all or part of the processes of the above embodiment and equivalent changes made according to the claims of the present invention still fall within the scope of the invention.
Claims
1. An unmanned aerial vehicle land surveying and mapping system, characterized in that: It includes UAV platform, laser calibration camera, cultivated land identification module, area calculation module and report generation module; The UAV platform is used to carry a laser calibration camera for aerial photography; The laser calibration camera is used to calibrate the distance of the cultivated land accurately and collect ground images using a laser-assisted camera; The farmland identification module is used to identify farmland boundaries and crop types from captured images; The area calculation module is used to convert the cultivated land area identified in the image into the actual cultivated land area based on the cultivated land boundary and the distance data measured by the laser in combination with the camera parameters; The report generation module is used to generate a surveying and mapping report based on the positioning information and the actual cultivated land area.
2. The UAV land surveying and mapping system according to claim 1, characterized in that: The UAV platform includes a UAV unit, a route setting unit and a GPS positioning unit; The route setting unit is used to plan the flight route based on the surveyed cultivated land area; The drone unit is used to perform flight actions based on the flight route; The GPS positioning unit is used to record the flight trajectory and shooting position coordinates of the drone in real time.
3. The UAV land surveying and mapping system according to claim 2, characterized in that: The laser calibration camera includes a calibration unit, a distance measurement unit, an image acquisition unit and a synchronization unit; The calibration unit is used to perform time synchronization and optical alignment on the camera and the laser; The distance measuring unit is used to start the laser to emit laser pulses to the ground to be measured and receive echo signals; The image acquisition unit is used to continuously capture ground images; The synchronization unit is used to synchronize the laser distance measurement data with the ground image so that each frame of the image has corresponding distance information.
4. The UAV land surveying and mapping system according to claim 3, characterized in that: The farmland identification module includes an image acquisition unit, an image correction unit, an image enhancement unit, a vegetation segmentation unit, a crop feature extraction unit, a model training unit and a farmland identification unit; The image acquisition unit is used to collect ground images taken by the drone; The image correction unit is used to perform radiation correction on the ground image to eliminate atmospheric effects, and to perform geometric correction to ensure the geographic accuracy of the image; The image enhancement unit is used to improve image quality through histogram equalization and contrast enhancement; The vegetation segmentation unit is used to set a threshold based on the normalized difference vegetation index to perform preliminary segmentation to obtain a vegetation area image; The crop feature extraction unit is used to extract crop features from the vegetation area image, and the crop features include spectral features, texture features and spatial layout features; The model training unit is used to train a support vector machine using crop features and a training set; The model training unit is used to identify the cultivated land in the vegetation area image based on the support vector machine to obtain the cultivated land boundary and the corresponding crop type.
5. The UAV land surveying and mapping system according to claim 4, characterized in that: The area calculation module includes a camera calibration unit, a distance acquisition unit, a measurement unit, a mapping unit and an area calculation unit; The camera calibration unit is used to perform lens distortion correction and geometric calibration on the camera, and use a calibration object of known size to determine the intrinsic and extrinsic parameters of the camera; The distance acquisition unit is used to acquire the distance data from multiple points provided by the laser to the camera, and synchronize with the moment when the camera captures the image; The measuring unit is used to calculate the point cloud data of the cultivated land in three-dimensional space by combining the camera parameters and the distance data using the triangulation principle; The mapping unit is used to project the three-dimensional point cloud data back to the two-dimensional image plane; The area calculation unit is used to perform polygonal approximation on the mapped cultivated land area boundary and calculate the actual coverage area of the cultivated land using a polygonal area calculation formula.
6. The UAV land surveying and mapping system according to claim 5, characterized in that: The area calculation module further includes a correction unit, and the correction unit is used to compensate the calculated actual coverage area.
7. The UAV land surveying and mapping system according to claim 6, characterized in that: The report generation module includes an information summary unit, a classification unit and a visualization unit; The summarizing unit is used to summarize the surveying and mapping information of all cultivated lands; The classification unit is used to classify and count the total cultivated land area by region and crop type; The visualization unit is used to perform visualization processing on the data using a data visualization tool.
8. The UAV land surveying and mapping system according to claim 7, characterized in that: The vegetation segmentation unit includes an analysis subunit, an iterative optimization unit and a segmentation subunit; The analysis subunit is used to find the peaks and valleys of the NDVI value distribution by using histogram analysis, and select the valley points as thresholds; The iterative optimization subunit is used to maximize the variance between categories using the Otsu method; The segmentation subunit is used to perform binarization processing on the NDVI image using a selected threshold value, marking pixels with NDVI values higher than the threshold value as vegetation pixels, and marking pixels with NDVI values lower than the threshold value as non-vegetation pixels.
9. A method for land surveying using an unmanned aerial vehicle. It is characterized in that Includes: UAV platform equipped with laser calibration camera for aerial photography; Use laser-assisted cameras to accurately calibrate the distance of cultivated land and collect ground images; Identify farmland boundaries and crop types from captured images; Based on the cultivated land boundary and the distance data measured by the laser, the cultivated land area identified in the image is converted into the actual cultivated land area in combination with the camera parameters; Generate surveying and mapping reports based on positioning information and actual cultivated land area.
Citation Information
Patent Citations
Precise registration method of ground laser-point clouds and unmanned aerial vehicle image reconstruction point clouds
CN103426165A
Plane area measuring method based on digital image analysis
CN103644865A
High-accuracy target positioning method of unmanned plane on basis of multi-station measurement
CN104501779A
Land area rapid detection method based on binocular vision and detection device thereof
CN105277144A
A tunnel point cloud data analysis method and system
CN109191521A
Cited By
Geographic surveying and mapping data acquisition and processing system based on image analysis
CN119445386A
