Method and system for automatically segmenting multi-time-sequence and multi-source data of field test plots

Abstract

The invention relates to a field test plot multi-time-sequence and multi-source data automatic segmentation method and system, and the method comprises the steps: obtaining original image data of different crop varieties of a field test plot, and carrying out the image processing, and obtaining registered multi-time-sequence and multi-source image data; wherein the original image data is image data with latitude and longitude information; extracting the latitude, longitude and altitude of each test cell from the image data, constructing a latitude and longitude coordinate matrix, and matching cell test information for each test cell; constructing a mask based on the longitude and latitude coordinate matrix of each test cell, wherein the mask represents the spatial range of the test cell; the mask of each test cell comprises a cell boundary, coordinate information and test information of the test cell; and applying the constructed masks to the registered multi-time-sequence and multi-source image data in batches, and extracting the image data of each cell through the space overlapping part of the masks and the image data. The method can effectively relieve the phenomena of poor effectiveness and misclassification of a traditional method.

Description

Technical Field

[0001] The present invention relates to the technical field of batch extraction of designated areas from drone images, and in particular to a method and system for automatically segmenting multi-time series and multi-source data of field test plots based on geospatial latitude and longitude data. Background Art

[0002] With the development of modern agriculture, the demand for precise management and monitoring of field plots is increasing. In traditional field plot boundary extraction methods, manual cutting is mainly relied on, which has the following problems: 1. Low efficiency: Manual cutting requires a lot of time and manpower. 2. Poor accuracy: It is easily affected by factors such as shadows and weeds inside the plot, resulting in misclassification. 3. Poor effectiveness: It is difficult to meet the real-time needs of modern agricultural monitoring. 4. The accuracy of boundary extraction is not high: Existing methods are difficult to accurately screen high-dimensional data in the target area, especially in complex terrain. For example, the coordinate system of the boundary does not match the coordinate system of the point cloud data, which will lead to errors. 5. Inconvenient data storage and sharing: The extracted data format is complex and not easy to use in subsequent analysis and data exchange.

[0003] Therefore, how to achieve automated and efficient field plot boundary extraction has important practical significance. Summary of the Invention

[0004] In response to the above problems, the purpose of the present invention is to provide a method and system for automatic segmentation of multi-time series and multi-source data in field experimental plots, which can effectively alleviate the poor effectiveness of traditional methods such as manual cutting and misclassification caused by shadows and weeds inside the plot.

[0005] To achieve the above-mentioned purpose, in a first aspect, the technical solution adopted by the present invention is: a method for automatic segmentation of multi-time series and multi-source data of a field test plot, which comprises: obtaining original image data of different crop varieties in the field test plot, and performing image processing to obtain aligned multi-time series and multi-source image data; wherein the original image data is image data with longitude and latitude information; extracting the longitude and latitude and altitude of each test plot from the image data, constructing a longitude and latitude coordinate matrix and matching the plot test information for each test plot; constructing a mask based on the longitude and latitude coordinate matrix of each test plot, the mask representing the spatial range of the test plot; the mask of each test plot includes the plot boundary and coordinate information and matches the test information of the test plot; applying the constructed mask in batches to the aligned multi-time series and multi-source image data, and extracting the image data of each plot through the spatial overlap between the mask and the image data.

[0006] Furthermore, original image data of different crop varieties in the field trial plot is obtained, including: obtaining image data of the field trial plot by an acquisition system, where the acquisition system is configured with a high-precision GPS module to obtain image data with latitude and longitude information.

[0007] Furthermore, image processing is performed, including: obtaining data with high-precision geographic coordinate information, processing the data using the coordinate information according to its coordinate range and coordinate system, and converting the longitude and latitude data in the geographic coordinate system into a projection coordinate system to match the spatial reference system of the image data; wherein, the data with high-precision geographic coordinate information includes orthophotos, point cloud data and spectral data.

[0008] Furthermore, the process of extracting the longitude and latitude of each experimental plot from the image data includes:

[0009] Through RTK positioning technology, ground measurement or point selection in the registered image data, the boundary inflection points and longitude and latitude data of each cell are extracted to determine the location and shape of each cell.

[0010] Furthermore, a latitude and longitude coordinate matrix is ​​constructed and cell test information is matched for each test cell, including:

[0011] The latitude and longitude data of all plots in the experimental site were constructed into a coordinate matrix, with the starting point and end point in the matrix being the same;

[0012] The plot corresponding to each coordinate matrix is ​​assigned set test information, which includes the plot number, test group, crop type and test parameters.

[0013] Furthermore, a mask is constructed based on the latitude and longitude coordinate matrix of each experimental plot. Specifically, a polygon algorithm or a rasterization-based mask generation method is used to ensure that the extracted image data accurately matches the actual geographic coordinates.

[0014] Furthermore, after extracting the image data of each cell through the spatial overlap of the mask and the image data, a data storage step is also included, specifically: classifying and saving the cell image data of each time series and data source, and saving the information of the test cell and the time series information as the file name.

[0015] In the second aspect, the technical solution adopted by the present invention is: a system for automatic segmentation of multi-time series and multi-source data of field test plots, which includes: an image acquisition module, which acquires the original image data of different crop varieties in the field test plot, and performs image processing to obtain aligned multi-time series, multi-source image data; wherein the original image data is image data with longitude and latitude information; a longitude and latitude extraction and matching module, which extracts the longitude and latitude and altitude of each test plot from the image data, constructs a longitude and latitude coordinate matrix and matches the plot test information for each test plot; a mask construction module, which constructs a mask based on the longitude and latitude coordinate matrix of each test plot, and the mask represents the spatial range of the test plot; the mask of each test plot includes the plot boundary, coordinate information and the test information of the test plot; a multi-time series and multi-source image data cutting module, which applies the constructed mask to the aligned multi-time series and multi-source image data in batches, and extracts the image data of each plot through the spatial overlap between the mask and the image data.

[0016] In a third aspect, the technical solution adopted by the present invention is: a computer-readable storage medium storing one or more programs, wherein the one or more programs include instructions, and when the instructions are executed by a computing device, the computing device executes any one of the above methods.

[0017] In a fourth aspect, the technical solution adopted by the present invention is: a computing device, comprising: one or more processors, a memory and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, and the one or more programs include instructions for executing any of the above methods.

[0018] The present invention has the following advantages due to the adoption of the above technical solution:

[0019] This invention significantly improves the efficiency and accuracy of field plot boundary extraction. It uses drones to acquire two-dimensional and three-dimensional data, combined with algorithms to automatically process the images and accurately segment experimental plots. This reduces manual labor and costs while improving data utilization and the accuracy of decision support. This invention provides powerful technical support for agricultural monitoring and research. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 This is a flow chart of a method for automatically segmenting multi-time series and multi-source data in a field test plot according to an embodiment of the present invention;

[0021] Figure 2 This is a three-dimensional visualization interface effect diagram of the segmented point cloud data in an embodiment of the present invention;

[0022] Figure 3a This is an orthophoto effect diagram after cutting in an embodiment of the present invention;

[0023] Figure 3b This is an example diagram of point cloud data after cutting in an embodiment of the present invention. DETAILED DESCRIPTION

[0024] UAV remote sensing technology uses drones as platforms equipped with remote sensing sensors to obtain surface information. It has a wide range of applications in areas such as agricultural monitoring and environmental assessment. The advantages of UAV remote sensing technology lie in its high temporal and spatial resolution, flexibility, and cost-effectiveness. UAV technology provides two-dimensional image data such as RGB images and multispectral data, as well as three-dimensional data such as point clouds and depth maps.

[0025] This invention proposes a method and system for automatically extracting and segmenting multi-temporal image data from farmland trial plots based on geospatial latitude and longitude data. The method can automatically extract the boundaries of experimental plots from drone imagery and perform batch segmentation of field plots, improving extraction efficiency and accuracy. A closed mask region is constructed based on geographic coordinates, and the mask is used to batch extract regional images with unified geographic coordinates from different time series. The latitude and longitude coordinates of each plot in the experimental plot are obtained, and a latitude and longitude coordinate matrix is ​​constructed. The drone image data from different time series is then registered. A mask for all plots in the experimental plot is constructed based on the coordinate matrix, and the mask is matched to the plot experimental information. The mask for each plot is used to extract the image data for the corresponding plot from the registered multi-temporal image data and match the plot experimental information. The method is applicable to different types of sensors. In two-dimensional data, it can be used with visible light camera data, multispectral data, hyperspectral data, and thermal infrared data, and is applicable to orthophotos, multi-band spectral information, full-band spectral information, and temperature maps. In three-dimensional data, lidar and other sensors can be used to produce point cloud data, depth maps, 3D model images, and the like.

[0026] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions of the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the described embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field are within the scope of protection of the present invention.

[0027] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present invention. As used herein, unless the context clearly indicates otherwise, the singular form is intended to include the plural form. In addition, it should be understood that when the terms "comprise" and / or "include" are used in this specification, they indicate the presence of features, steps, operations, devices, components and / or combinations thereof.

[0028] In one embodiment of the present invention, a method for automatic segmentation of multi-time series and multi-source data in field test plots is provided. In this embodiment, automatic segmentation of plots of two-dimensional data such as RGB images and thermal infrared images and three-dimensional point cloud data is achieved, aiming to improve the efficiency and accuracy of boundary extraction and minimize manual intervention. Through multi-source data fusion, image preprocessing, boundary extraction and model optimization, the present invention achieves efficient and automated segmentation of field plots, which is suitable for large-scale, continuous field trials and monitoring environments, and significantly improves the efficiency and accuracy of agricultural data analysis. Specifically, in this embodiment, if Figure 1 As shown, the method includes the following steps:

[0029] 1) Obtaining original image data of different experimental plots in the field and performing image processing to obtain registered multi-time series and multi-source image data; wherein the original image data is image data with latitude and longitude information;

[0030] 2) Extract the latitude, longitude, and altitude of each test plot from the image data, construct a longitude and latitude coordinate matrix, and match the test plot information for each test plot;

[0031] 3) Constructing a mask based on the latitude and longitude coordinate matrix of each test cell, with the mask representing the spatial extent of the test cell; the mask of each test cell includes the cell boundary, coordinate information, and test information of the test cell;

[0032] 4) The constructed mask is batch-applied to the registered multi-time series and multi-source image data, and the image data of each cell is extracted through the spatial overlap between the mask and the image data.

[0033] In step 1), raw image data of different experimental plots in the field are obtained. Specifically, the image data of the field experimental plots is obtained by a collection system equipped with a high-precision GPS module to obtain image data with latitude and longitude information. The collection system is, for example, a drone.

[0034] In step 1) above, image processing, including 2D data extraction, image registration, normalization, and feature extraction, is performed to ensure that the image data is consistent with the geographic coordinate system, providing accurate image data for subsequent agricultural monitoring and analysis. Specifically, image registration involves acquiring image data from ground-based experimental plots and processing the data using coordinate information based on their coordinate range and coordinate system. This data is converted from the geographic coordinate system's longitude and latitude data to a projected coordinate system to match the image data's spatial reference system. Data with high-precision geographic coordinate information includes orthophotos, point cloud data, and spectral data.

[0035] The image data's coordinate system is converted from a local coordinate system to a projected coordinate system consistent with the target area. This conversion process ensures that the image data and the target area's coordinate system are consistent, enabling accurate image data extraction. Geographic Information System-based coordinate conversion tools can be used, which typically provide a rich set of coordinate system conversion capabilities and support conversions across multiple coordinate systems. This method is not only applicable to 3D point cloud data of corn but can also be widely applied to the processing and analysis of other types of agricultural crop and terrain data.

[0036] In this example, the field trial plots of different crop varieties, including corn, are processed using the present invention, which can process multiple data types, including hyperspectral images, thermal infrared images, multi-hyperspectral images, and depth maps, to achieve batch automated segmentation of plots. Specifically, raw image data of different varieties is first acquired and then processed and spliced ​​to generate complete image data. This process includes image denoising, splicing, and alignment to ensure the continuity and consistency of the final image, facilitating subsequent analysis and processing.

[0037] Therefore, in this embodiment, the data collection of crop varieties includes the following situations:

[0038] RGB image acquisition: Drones equipped with high-resolution RGB cameras capture visible light images of crops at appropriate flight altitudes and speeds. These images are used to analyze crop growth, health, and biomass.

[0039] Multispectral image acquisition: Drones are equipped with multispectral cameras that can capture images in multiple wavelengths (e.g., red, green, blue, near-infrared, etc.). Multispectral images can be used for crop growth monitoring, pest and disease detection, and soil moisture analysis.

[0040] Hyperspectral imagery acquisition: Drones equipped with hyperspectral sensors can acquire more detailed spectral information. Hyperspectral images can provide in-depth information on the physiological state and chemical composition of crops, making them suitable for detailed crop monitoring and analysis.

[0041] Thermal infrared image acquisition: Drones equipped with thermal infrared cameras can monitor the temperature distribution of crops. This data is important for assessing crop moisture status and heat stress response.

[0042] Point cloud acquisition: Drones equipped with lidar sensors capture 3D point cloud data of crops and terrain. This data provides information such as crop height and canopy structure, making it suitable for biomass estimation and canopy analysis.

[0043] In step 2) above, the process of extracting the longitude and latitude of each test cell from the image data includes: extracting the boundary inflection points and longitude and latitude data of each cell through RTK positioning technology, ground measurement, or selecting points in the aligned image data to determine the position and shape of each cell.

[0044] In the above step 2), in order to automatically extract cells in batches based on the latitude, longitude and altitude of the four vertices of each cell, a latitude and longitude coordinate matrix is ​​constructed and cell test information is matched for each test cell, including the following steps:

[0045] 2.1) Construct the longitude and latitude data of all plots in the experimental site into a coordinate matrix, with the starting and ending points being the same;

[0046] 2.2) Assign the set experimental information to each plot corresponding to the coordinate matrix. The experimental information includes the plot number, experimental group, crop type and experimental parameters.

[0047] In this embodiment, multi-time series and multi-source data from field experimental plots are automatically segmented based on geospatial latitude and longitude data. This approach is applicable to both the processing of two-dimensional data such as RGB images and the automatic segmentation of point cloud data. By reading LAS point cloud files and analyzing their spatial extent, the present invention verifies the point cloud's coordinate system. This process provides a preliminary check for new point cloud data processing, ensuring the accuracy of coordinate conversions and troubleshooting issues caused by coordinate system mismatches.

[0048] In the above step 3), a mask is constructed based on the latitude and longitude coordinate matrix of each experimental plot, specifically by using a polygon algorithm or a rasterization-based mask generation method to ensure that the extracted image data accurately matches the actual geographic coordinates.

[0049] In the above embodiments, after extracting the image data of each cell through the spatial overlap of the mask and the image data, a data storage step is also included, specifically: the cell image data of each time series and data source are classified and saved, and the information of the test cell and the time series information are saved as file names.

[0050] In step 4) above, by receiving input data containing the latitude and longitude coordinate points of the boundaries of each field plot, a coordinate list of closed polygons is constructed for each plot, and corresponding polygons are created using image processing and geometric algorithms. Finally, the image containing the farmland area is cropped based on these polygons, thereby accurately extracting the image portion of each plot.

[0051] Specifically, when processing point cloud data, the first step is to read and visualize millions of point cloud data points while determining the boundaries of the data. To do this, a coordinate point needs to be selected as a reference point to help set the boundaries of the visualization. The visualized point cloud data is saved for further analysis. During the coordinate system conversion process, the original coordinate system of the point cloud must first be analyzed and the corresponding file read to obtain the necessary coordinate information. Based on the analysis results, the appropriate target coordinate system is selected and the conversion is performed to ensure that the data can be correctly aligned and read. Finally, the converted point cloud data is visualized to verify the accuracy of the processing.

[0052] The present embodiment relates to the field of point cloud data processing, and adopts a point cloud data processing, extraction and visualization method based on regional boundary coordinates. With the rapid development of lidar and remote sensing technology, point cloud data has been widely used in fields such as topographic mapping, three-dimensional modeling and GIS. However, the amount of point cloud data is usually extremely large, especially in field experiments, and researchers often only need to extract point cloud data of specific areas. Existing methods usually lack efficient and accurate processing functions in regional screening. In addition, the coordinate system conversion and the precise extraction of points in the region need to be considered during the processing to ensure the accuracy and integrity of data processing. Therefore, the present invention is based on an efficient point cloud data extraction method based on regional boundary coordinates, so as to quickly screen, process and analyze point cloud data of a specific area to meet different application requirements.

[0053] In the above embodiment, taking the corn experimental plot as an example, the plot segmentation process of point cloud data is described in detail. In order to ensure the accurate adaptation of the point cloud data within the specific coordinate system, the three-dimensional point cloud data of the corn plot is first obtained by the UAV acquisition system and visualized. According to the boundary information of the experimental plot, the data of the local cell area is extracted from the overall point cloud data through masking technology to achieve accurate spatial positioning and segmentation. Furthermore, a batch processing method is used to automatically extract the point cloud data of multiple experimental plots and save them as standard format files to ensure the uniformity and ease of use of the data. Ultimately, the saved segmentation results can be applied to multiple fields such as sample plot surveys, single tree analysis, canopy structure analysis, etc., providing high-precision three-dimensional spatial data support, and providing effective data processing solutions for research and application in fields such as agriculture, forestry and geographic information systems (GIS).

[0054] In the above embodiments, Figure 2 Figure 2 shows a visualization of segmented point cloud data from a corn experimental plot. Different colors and boundary lines distinguish the segmented experimental plots, visually indicating the location of each plot and whether its boundary falls within the predetermined range. This method provides an efficient data visualization and processing tool, enabling rapid and accurate region identification and data verification in complex multi-time series and multi-source data environments, ensuring data quality and the reliability of analysis results.

[0055] In this embodiment, by cropping and processing the image, the present invention realizes the extraction and analysis of the cell image, covering the steps of reading the cell polygon coordinates, cropping the specific area image, extracting multi-band image data and normalizing pixel values, and provides convenience for further analysis through batch automated processing.

[0056] like Figure 3a As shown, by reading the boundary coordinates of the experimental plot, the image data of the specified area is extracted, and the first three bands in the image are selected to construct an RGB image. On this basis, the image data is normalized to ensure that the image data at different time points or regions have consistent scale and accuracy, providing reliable support for subsequent feature extraction and analysis. Relevant agricultural features are extracted and the image metadata is updated through a specific algorithm to ensure that the output is a standard three-band RGB format image file. By automatically loading configuration files and creating output paths, batch automated processing of multi-plot image data is achieved, significantly improving the efficiency and accuracy of data processing. The present invention optimizes the image data processing and analysis process, provides efficient and accurate data support for precision agricultural monitoring, and promotes the temporal change analysis and refined management of farmland test plots.

[0057] like Figure 3b As shown, a diverse batch display of segmented point cloud data is achieved, and the vegetation structure and terrain features are finely analyzed through height coloring. Taking corn point cloud data as an example, three-dimensional presentation is first performed to ensure that the data spatial structure is complete and within the field of view. In order to optimize rendering efficiency, downsampling processing is adopted, and a scatter plot with height color mapping is used to display vertical distribution characteristics. The viewing angle of each sub-graph can be flexibly adjusted to ensure the best display effect and accurate data interpretation. The present invention can not only accurately test the segmentation effect, but also support in-depth analysis of vegetation and terrain in different areas. It is suitable for multiple fields such as agricultural monitoring, environmental assessment and terrain survey.

[0058] In the above embodiment, efficient processing of field plot segmentation is achieved by automatically processing drone image data, as follows:

[0059] Efficient segmentation of plot boundaries: Using image data collected by drone platforms, field plot boundaries are automatically identified and segmented, significantly reducing the workload of traditional manual processing and improving efficiency.

[0060] Integration of multiple data sources: Combining multiple image and sensor data, and accurately extracting boundary coordinates, provides richer details for further analysis of crop growth within field plots.

[0061] Coordinate transformation and accuracy improvement: Through precise coordinate transformation, the coordinate system of image data is matched with the geographic coordinate system to ensure the accuracy of the data and avoid potential coordinate errors.

[0062] Visualization support: Realize 3D visualization of image data, automatically generate segmented images, and display segmentation results through a visual interface to facilitate further inspection and analysis.

[0063] Modularity and ease of use: The modular design of the segmentation process not only improves segmentation efficiency, but also combines with an easy-to-use interface design, allowing users to easily operate without professional skills.

[0064] Wide Applicability: This method is not only suitable for agricultural monitoring but also has broad application in a variety of fields, including urban planning, topographic mapping, and environmental assessment. Through automated data processing and precise segmentation, it provides efficient and highly accurate data support, meeting the monitoring needs of various fields. In agriculture, this innovative technology is particularly suitable for real-time, precise field plot monitoring.

[0065] This example accurately segments small areas from drone aerial images, achieving both high resolution and high accuracy. This method can be widely used in fields such as agricultural monitoring. The automated segmentation process significantly improves data processing efficiency and reduces the cost and time of manual intervention.

[0066] In one embodiment of the present invention, a system for automatically segmenting multi-time series and multi-source data of a field experimental plot is provided, comprising:

[0067] The image acquisition module acquires the original image data of different crop varieties in the field test plot and performs image processing to obtain registered multi-time series and multi-source image data; wherein the original image data is image data with latitude and longitude information;

[0068] The latitude and longitude extraction and matching module extracts the latitude and longitude and altitude of each test plot from the image data, constructs a latitude and longitude coordinate matrix, and matches the test plot information for each test plot;

[0069] A mask construction module constructs a mask based on the latitude and longitude coordinate matrix of each test cell, where the mask represents the spatial extent of the test cell. The mask of each test cell includes the cell boundary, coordinate information, and test information of the test cell.

[0070] The multi-time series and multi-source image data cutting module applies the constructed mask to the registered multi-time series and multi-source image data in batches, and extracts the image data of each cell through the spatial overlap between the mask and the image data.

[0071] In the above embodiment, obtaining original image data of different crop varieties in the field trial plot includes:

[0072] The image data of the field trial plot is acquired by the acquisition system, which is equipped with a high-precision GPS module to obtain image data with latitude and longitude information.

[0073] In this embodiment, the process of performing image processing includes:

[0074] Acquire data with high-precision geographic coordinate information, process the data using the coordinate information based on its coordinate range and coordinate system, and convert the longitude and latitude data in the geographic coordinate system into a projected coordinate system to match the spatial reference system of the image data; data with high-precision geographic coordinate information includes orthophotos, point cloud data, and spectral data.

[0075] In the above embodiment, the process of extracting the longitude and latitude of each test plot from the image data includes:

[0076] Through RTK positioning technology, ground measurement or point selection in the registered image data, the boundary inflection points and longitude and latitude data of each cell are extracted to determine the location and shape of each cell.

[0077] In the above embodiment, constructing a longitude and latitude coordinate matrix and matching cell test information for each test cell includes:

[0078] The latitude and longitude data of all plots in the experimental site were constructed into a coordinate matrix, with the starting point and end point in the matrix being the same;

[0079] The plot corresponding to each coordinate matrix is ​​assigned set test information, which includes the plot number, test group, crop type and test parameters.

[0080] In the above embodiment, a mask is constructed based on the latitude and longitude coordinate matrix of each test cell, specifically by using a polygon algorithm or a rasterization-based mask generation method to ensure that the extracted image data accurately matches the actual geographic coordinates.

[0081] In the above embodiment, after extracting the image data of each cell through the spatial overlap of the mask and the image data, a data storage step is also included, specifically: classifying and saving the cell image data of each time series and data source, and saving the information of the test cell and the time series information as the file name.

[0082] The system provided in this embodiment is used to execute the above-mentioned method embodiments. Please refer to the above-mentioned embodiments for specific processes and detailed contents, which will not be repeated here.

[0083] In summary, the present invention utilizes image data acquired by an unmanned aerial vehicle remote sensing platform to achieve efficient and accurate extraction of breeding plot boundaries through automated image processing. Automated image processing includes two-dimensional image data acquired using different sensors and a large amount of three-dimensional point cloud data, which usually contains millions or even billions of points. The coordinate information of each point is expressed in the form of X, Y, and Z in three-dimensional space, and may be accompanied by other attribute information such as laser reflection intensity. These data are widely used to generate terrain models, three-dimensional structures of buildings, vegetation coverage, etc. Since the amount of two-dimensional image data and point cloud data is huge, and the original data usually covers a large area, it is often necessary to extract point cloud data and image data within a specific area during the processing process to meet specific analysis needs. Through the technical solution provided by the present invention, users can quickly and accurately extract the required area from massive data, thereby providing strong data support and decision-making basis for fields such as agricultural scientific research, ecological monitoring and environmental management.

[0084] In one embodiment of the present invention, a computing device is provided. The computing device may be a terminal and may include: a processor, a communications interface, a memory, a display screen, and an input device. The processor, communications interface, and memory communicate with each other via a communications bus. The processor is configured to provide computing and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system and a computer program. When executed by the processor, the computer program implements the methods described in the above embodiments. The internal memory provides an environment for the operating system and computer program in the non-volatile storage medium to run. The communications interface is configured to communicate with an external terminal via wired or wireless communication. The wireless communication may be achieved via Wi-Fi, a network management service provider, NFC (near field communication), or other technologies. The display screen may be a liquid crystal display or an electronic ink display. The input device may be a touchscreen layer covering the display screen, or may be buttons, a trackball, or a touchpad provided on the computing device housing, or may be an external keyboard, touchpad, or mouse. The processor may invoke logic instructions stored in the memory.

[0085] In addition, the logical instructions in the above-mentioned memory can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when sold or used as an independent product. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0086] In one embodiment of the present invention, a computer program product is provided, which includes a computer program stored on a non-transitory computer-readable storage medium, and the computer program includes program instructions. When the program instructions are executed by a computer, the computer can execute the methods provided by the above-mentioned method embodiments.

[0087] In one embodiment of the present invention, a non-transitory computer-readable storage medium is provided. The non-transitory computer-readable storage medium stores server instructions. The computer instructions enable a computer to execute the methods provided in the above embodiments.

[0088] The above embodiment provides a computer-readable storage medium, whose implementation principle and technical effects are similar to those of the above method embodiment, and will not be repeated here.

[0089] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as combinations of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts and / or block diagrams. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0090] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0091] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0092] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A method for automatic segmentation of multi-time series and multi-source data in field experimental plots, characterized in that: include: Obtaining original image data of different crop varieties in the field trial plot and performing image processing to obtain registered multi-time series and multi-source image data; wherein the original image data is image data with latitude and longitude information; Extract the latitude, longitude and altitude of each test plot from the image data, construct a longitude and latitude coordinate matrix and match the plot test information for each test plot; Constructing a mask based on the latitude and longitude coordinate matrix of each test plot, with the mask representing the spatial extent of the test plot; the mask of each test plot includes the plot boundary and coordinate information and matches the test information of the test plot; The constructed masks are batch-applied to the registered multi-time series and multi-source image data, and the image data of each cell is extracted through the spatial overlap between the mask and the image data.

2. The method for automatic segmentation of multi-time series and multi-source data in field experimental plots according to claim 1, characterized in that: Obtain original image data of different crop varieties in field trial plots, including: The image data of the field trial plot is acquired by the acquisition system, which is equipped with a high-precision GPS module to obtain image data with latitude and longitude information.

3. The method for automatic segmentation of multi-time series and multi-source data in field experimental plots according to claim 2, characterized in that: Perform image processing, including: Acquire data with high-precision geographic coordinate information, process the data using the coordinate information based on its coordinate range and coordinate system, and convert the longitude and latitude data in the geographic coordinate system into a projected coordinate system to match the spatial reference system of the image data; among them, data with high-precision geographic coordinate information includes orthophotos, point cloud data, and spectral data.

4. The method for automatic segmentation of multi-time series and multi-source data in field experimental plots according to claim 1, characterized in that: The process of extracting the latitude and longitude of each experimental plot from the image data includes: Through RTK positioning technology, ground measurement or point selection in the registered image data, the boundary inflection points and longitude and latitude data of each cell are extracted to determine the location and shape of each cell.

5. The method for automatic segmentation of multi-time series and multi-source data in field experimental plots according to claim 1, characterized in that: Construct a longitude and latitude coordinate matrix and match the cell test information for each test cell, including: The latitude and longitude data of all plots in the experimental site were constructed into a coordinate matrix, with the starting point and end point in the matrix being the same; The plot corresponding to each coordinate matrix is ​​assigned set test information, which includes the plot number, test group, crop type and test parameters.

6. The method for automatic segmentation of multi-time series and multi-source data in field experimental plots according to claim 1, characterized in that: Construct a mask based on the latitude and longitude coordinate matrix of each experimental plot. Specifically, use a polygon algorithm or a rasterization-based mask generation method to ensure that the extracted image data accurately matches the actual geographic coordinates.

7. The method for automatic segmentation of multi-time series and multi-source data in field experimental plots according to claim 1, characterized in that: After extracting the image data of each cell through the spatial overlap of the mask and the image data, the process also includes a data storage step, specifically: classifying and saving the cell image data of each time series and data source, and saving the information of the test cell and the time series information as the file name.

8. A system for automatic segmentation of multi-time series and multi-source data in field experimental plots, characterized by: include: The image acquisition module acquires the original image data of different crop varieties in the field test plot and performs image processing to obtain registered multi-time series and multi-source image data; wherein the original image data is image data with latitude and longitude information; The latitude and longitude extraction and matching module extracts the latitude and longitude and altitude of each test plot from the image data, constructs a latitude and longitude coordinate matrix, and matches the test plot information for each test plot; A mask construction module constructs a mask based on the latitude and longitude coordinate matrix of each test cell, where the mask represents the spatial extent of the test cell. The mask of each test cell includes the cell boundary, coordinate information, and test information of the test cell. The multi-time series and multi-source image data cutting module applies the constructed mask to the registered multi-time series and multi-source image data in batches, and extracts the image data of each cell through the spatial overlap between the mask and the image data.

9. A computer-readable storage medium storing one or more programs, characterized in that: The one or more programs include instructions that, when executed by a computing device, cause the computing device to perform any one of the methods of claims 1 to 7 .

10. A computing device, characterized in that include: One or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, and the one or more programs include instructions for executing any one of the methods according to claims 1 to 7.

Images