A Calculation Method and Device for the Planting Area of Torreya grandis Fort. cv. Merrillii Based on Age Characteristics

Abstract

The present invention discloses a method and device for calculating the planting area of Torreya grandis according to age characteristics. According to the different physiological characteristics of Torreya grandis at different ages, the whole Torreya grandis is divided into two categories: adult forest and juvenile forest. After extraction and merging respectively, the overall distribution of Torreya grandis is obtained. For the adult forest, a sub-meter high-resolution image is used as the base map for annotation, and a deep learning image recognition framework is used for recognition. For the juvenile forest, Sentinel multi-spectral data combined with vegetation coverage is used as the basic data, and the random forest method is used to extract young forests, and the misclassified fields are removed by using the characteristic that the peak vegetation index of the fields is extremely high throughout the year. Finally, the planting patches of Torreya grandis are merged, and the number of pixel points is counted under the Albers equal-area projection. Combining the spatial resolution information and slope information, the planting area of Torreya grandis is obtained, and the planting area of Torreya grandis close to the statistical data rather than the projected area is obtained, which is of great significance for studying the planting distribution of Torreya grandis and guiding the production of Torreya grandis.

Description

Technical Field

[0001] The present invention relates to the field of remote sensing change monitoring, and particularly to a method and device for calculating the planting area of Torreya grandis Fort. cv. Merrillii based on age characteristics. Background Art

[0002] In the prior art, there are detection methods for agricultural crops, mainly by counting the crops existing in a region and their corresponding planting sub-regions, and counting the annual yield per unit area and the comprehensive quality coefficient of crop fruits in each historical year for each planting sub-region corresponding to each crop. Then, the average annual yield per unit area and the average comprehensive quality coefficient of crop fruits for each planting sub-region corresponding to each crop are counted, so as to sort the planting sub-regions corresponding to each crop, and thus analyze and obtain the suitable planting sub-regions corresponding to each crop according to the sorting results, realizing the monitoring and analysis of the regional crop planting situation, and providing a reference basis for regional managers to adjust the crop planting areas according to the suitable planting sub-regions corresponding to each crop.

[0003] There is also an extraction algorithm for soybean planting areas in the prior art, mainly by preprocessing the remote sensing image data Sentinel-2, and selecting G bands beneficial to soybean extraction from the preprocessed remote sensing image data; calculating the normalized difference vegetation index and the normalized difference water index, and the Sentinel-2 data contains G + 2 data; using the USDA of a certain year provided by the GEE platform for preprocessing; constructing a training set and a test set using the Sentinel-2 data containing G + 2 data and the preprocessed USDA; inputting the training set into three classifiers for training, inputting the test set into the trained classifiers for classification, and obtaining the soybean planting areas according to the classification results. It is used to solve the problems of low algorithm efficiency and long time consumption caused by huge remote sensing data volume and insufficient accuracy of a single algorithm for extracting soybeans.

[0004] The rapid expansion of Torreya grandis forests has caused a series of environmental problems, and there is an urgent need to accurately obtain information on the spatiotemporal distribution of Torreya grandis plantings. However, due to the complex landscape characteristics of Torreya grandis forests, there is currently no suitable method to extract distribution information of Torreya grandis forests. Torreya grandis forests have significant differences in canopy characteristics and ground cover characteristics. One type is composed of tall trees with a large crown area and irregular distribution; the other type is mostly short shrubs with small individual trees and a very regular distribution. The understory can be divided into two sub-categories: grass cover and no grass cover, and the overall stand cover is low. Currently, only a few studies have conducted remote sensing information extraction on Torreya grandis. Some are based on IKONOS high-resolution satellite imagery, while others combine spectral information, vegetation indices, and texture information, and use object-oriented multi-scale segmentation, nearest neighbor classification, or decision tree multi-scale classification. This type of research is mostly used to obtain the distribution information of old Torreya grandis forests at a local scale, and does not reflect the distribution information of young Torreya grandis forests that have been expanded over a large area. Moreover, the classification method used is still the traditional classification method that covers all land types, and is not specifically for extracting thematic information on Torreya grandis. Summary of the invention

[0005] The present invention provides a method for calculating the planting area of Torreya grandis based on age characteristics, which aims to solve the problem of incomplete extraction of Torreya grandis forest information and a large number of misclassifications and omissions; at the same time, it solves the situation where the area calculated from the extracted Torreya grandis patches deviates greatly from the planting area.

[0006] To achieve the above object, the present invention adopts the following technical solutions:

[0007] Acquire remote sensing multispectral data to be identified and a raster layer of an old Torreya grandis forest, wherein the remote sensing multispectral data is used to identify a young Torreya grandis forest;

[0008] Performing a preliminary screening of the remote sensing multispectral data using a random forest algorithm to obtain a first screening result of the remote sensing multispectral data;

[0009] Successively calculating the normalized vegetation index and vegetation coverage of the remote sensing multispectral data, and performing overlay analysis on the first screening result to obtain a second screening result of the remote sensing multispectral data;

[0010] The planting area of Torreya grandis is calculated based on the second screening result and the raster layer of the Torreya grandis old forest.

[0011] Preferably, the remote sensing multispectral data is preliminarily screened using a random forest algorithm to obtain a first screening result of the remote sensing multispectral data, specifically comprising:

[0012] Using the random forest algorithm to screen the time series data with cloud cover less than a certain threshold in the remote sensing multispectral data and perform median synthesis as a random forest classification base map;

[0013] Set the parameters of the random forest algorithm, mark the young Torreya grandis forests in the random forest classification base map, and train to generate a trainer;

[0014] Use the trainer to perform binary classification on the random forest classification base map to obtain a first screening result.

[0015] Preferably, calculate the normalized difference vegetation index and vegetation coverage of the remote sensing multispectral data successively, and perform overlay analysis in combination with the first screening result to obtain a second screening result of the remote sensing multispectral data, including:

[0016] Calculate the normalized difference vegetation index and vegetation coverage of the remote sensing multispectral data successively, and the calculation formulas are:

[0017] NDVI = (NIR - RED) / (NIR + RED) (1);

[0018] FVC = (NDVI - NDVImin) / (NDVImax - NDVImin) (2);

[0019] Among them, NDVI is the value of the normalized difference vegetation index, NIR is the near-infrared band, RED is the red band, FVC is the vegetation coverage, NDVImin is the value of the normalized difference vegetation index of pure bare soil, and NDVImax is the value of the normalized difference vegetation index of high vegetation coverage;

[0020] Select the layers in the remote sensing multispectral data where the normalized difference vegetation index value and the vegetation coverage are less than the preset threshold, and perform overlay analysis with the first screening result, and take the intersection to obtain the second screening result.

[0021] Preferably, calculating the planting area of Torreya grandis according to the second screening result and the raster layer of old Torreya grandis forests includes:

[0022] Use the raster calculator to perform overlay analysis on the second screening result and the raster layer of old Torreya grandis forests to obtain the raster result of the spatial distribution of Torreya grandis forests;

[0023] According to the Albers equal-area projection method, count the number of pixel points in the raster result of the spatial distribution of Torreya grandis forests, and combine the spatial resolution information and slope information to obtain the planting area of Torreya grandis.

[0024] A method and device for calculating the planting area of Torreya grandis based on age characteristics, including:

[0025] An acquisition module: used to acquire the remote sensing multispectral data to be recognized and the raster layer of old Torreya grandis forests, and the remote sensing multispectral data is used to identify young Torreya grandis forests;

[0026] The first screening module: used to preliminarily screen the remote sensing multispectral data by using the random forest algorithm to obtain the first screening result of the remote sensing multispectral data;

[0027] The second screening module: used to calculate the normalized difference vegetation index and vegetation coverage of the remote sensing multispectral data successively, and perform overlay analysis in combination with the first screening result to obtain the second screening result of the remote sensing multispectral data;

[0028] The calculation module: used to calculate the planting area of Torreya grandis according to the second screening result and the raster layer of the old Torreya grandis forest.

[0029] Preferably, the first screening module includes:

[0030] The map generation unit: used to screen the time-series data with cloud amount less than a certain threshold in the remote sensing multispectral data by using the random forest algorithm and perform median synthesis as the random forest classification base map;

[0031] The training unit: used to set the parameters of the random forest algorithm, and mark the young Torreya grandis forests in the random forest classification base map to train and generate a trainer;

[0032] The classification unit: used to perform binary classification on the random forest classification base map by using the trainer to obtain the first screening result.

[0033] Preferably, the second screening module includes:

[0034] The calculation subunit: used to calculate the normalized difference vegetation index and vegetation coverage of the remote sensing multispectral data successively, and the calculation formulas are:

[0035] NDVI = (NIR - RED) / (NIR + RED) (1);

[0036] FVC = (NDVI - NDVImin) / (NDVImax - NDVImin) (2);

[0037] where NDVI is the value of the normalized difference vegetation index, NIR is the near-infrared band, RED is the red band, FVC is the vegetation coverage, NDVImin is the value of the normalized difference vegetation index of pure bare soil, and NDVImax is the value of the normalized difference vegetation index of high vegetation coverage;

[0038] The screening subunit: used to select the layers in the remote sensing multispectral data where the values of the normalized difference vegetation index and the vegetation coverage are less than the preset thresholds, and perform overlay analysis with the first screening result to take the intersection to obtain the second screening result.

[0039] Preferably, the calculation module includes:

[0040] Overlay unit: used to perform overlay analysis on the second screening result and the grid layer of the ancient Torreya grandis forest using a grid calculator to obtain the grid result of the spatial distribution of Torreya grandis forests;

[0041] Statistics unit: used to count the number of pixel points in the grid result of the spatial distribution of Torreya grandis forests according to the Albers equal-area projection method, and combine the spatial resolution information and slope information to obtain the planting area of Torreya grandis.

[0042] A method and device for calculating the planting area of Torreya grandis based on age characteristics, including a memory and a processor, where the memory is used to store one or more computer instructions, and wherein the one or more computer instructions are executed by the processor to implement a method for calculating the planting area of Torreya grandis based on age characteristics as described in any one of the above.

[0043] A computer-readable storage medium storing a computer program, where the computer program causes a computer to implement a method for calculating the planting area of Torreya grandis based on age characteristics as described in any one of the above when executed.

[0044] The present invention has the following beneficial effects:

[0045] First point: For the problem of extracting the target. In this technical solution, by separately extracting young forests (shrub type) and adult forests (arbor type) and finally merging to obtain the total Torreya grandis forest patch map, the problems of incomplete extraction and high misclassification rate when extracting a certain type are avoided.

[0046] Second point: For the problem of classification method. In this technical solution, feature information extraction is deliberately carried out for ground objects such as Torreya grandis, including information such as crown shape and vegetation coverage, rather than using a conventional classification method covering all land types.

[0047] Third point: For the problem of the classification method and data source types used. In the selection of the method, this technical solution includes but is not limited to the random forest algorithm, and also combines spectral features, phenological characteristics, and texture information; in terms of the data sources used, there are sub-meter Google Earth images, multi-spectral Sentinel-2 images, and 30-meter resolution terrain data.

[0048] Fourth point: For the problem of projected area and slope area. Based on the result of obtaining the projected area, this technical solution supplements terrain data and calculates the slope area using the Pythagorean theorem to narrow the data gap with the statistically obtained planting area. Description of the Drawings

[0049] Figure 1 is a flowchart of a method for calculating the planting area of Torreya grandis based on age characteristics implemented in an embodiment of the present invention;

[0050] Figure 2It is the preliminary screening result diagram of Torreya grandis sapling forests in the embodiments of the present invention;

[0051] Figure 3 It is the specific flowchart of the embodiments of the present invention;

[0052] Figure 4 It is the schematic diagram of a device for calculating the planting area of Torreya grandis based on age characteristics in the embodiments of the present invention;

[0053] Figure 5 It is the schematic diagram of an electronic device for a device that implements a method for calculating the planting area of Torreya grandis based on age characteristics in the embodiments of the present invention. Specific embodiments

[0054] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative efforts belong to the scope of protection of the present invention.

[0055] The terms "first", "second", etc. in the claims and the description of this application are used to distinguish similar objects, and do not necessarily need to be used to describe a specific order or sequence. It should be understood that such terms can be interchanged under appropriate circumstances. This is only a way of distinguishing when describing objects with the same attributes in the embodiments of this application. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusion, so that a process, method, system, product or device including a series of units does not have to be limited to those units, but may include other units that are not clearly listed or are inherent to these processes, methods, products or devices.

[0056] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the technical field to which this application belongs. The terms used in the description of this application in this application are only for the purpose of describing specific embodiments, and are not intended to limit this application.

[0057] Embodiment 1

[0058] As Figure 1 and 2 shown, a method for calculating the planting area of Torreya grandis based on age characteristics includes the following steps:

[0059] S110. Obtain the remotely sensed multispectral data to be recognized and the raster layer of the old Torreya grandis forests, and the remotely sensed multispectral data is used to identify the Torreya grandis sapling forests;

[0060] S120. Use the random forest algorithm to preliminarily screen the remote sensing multispectral data to obtain the first screening result of the remote sensing multispectral data;

[0061] S130. Calculate the normalized difference vegetation index and vegetation coverage of the remote sensing multispectral data in sequence, and perform overlay analysis in combination with the first screening result to obtain the second screening result of the remote sensing multispectral data;

[0062] S140. Calculate the planting area of torreya grandis according to the second screening result and the raster layer of the old torreya grandis forest.

[0063] In Embodiment 1, since the existing technology has been disclosed in the old torreya grandis forest extraction algorithm, first use the Training sample manager function in ArcGISpro for annotation, and circle the torreya grandis as the training sample; secondly, use the ExportTraining Data For Deep Learning function to export the training data; finally, use the Detectg Object UsingDeep Learning to execute the recognition process to obtain the recognition result of adult torreya grandis trees. This recognition result is a vector, and the vector needs to be converted into a raster result. Here, in order to be unified with the young torreya grandis forest, the extraction result of the old torreya grandis forest is resampled to a resolution of 10m, so that the raster layer of the old torreya grandis forest is consistent with the spatial resolution of the young torreya grandis forest layer, and the raster layer of the old torreya grandis forest can be directly obtained.

[0064] Then obtain the remote sensing multispectral data to be recognized for identifying the young torreya grandis forest. Adopt the binary classification method, which is only divided into the young torreya grandis forest and other ground objects. Here, the extraction of the old torreya grandis forest is not considered, and it is marked on the map. Then set the parameters of the random forest. The main parameters are the number of feature selections, the maximum tree depth, and the number of trees. Here, the number of feature selections is all, the tree depth is set to 100, and the number of trees is set to 100, and 50% of the previous annotations are used as the training set. The remaining 30% is used as the validation set. When the recall rate is greater than 80%, the trainer is completed, and the base map is binary classified with this trainer to obtain the preliminary screening result of the young torreya grandis forest. As Figure 2 shown, the preliminary screening result shows the patches of torreya grandis. Then calculate the vegetation normalized index and vegetation coverage. The vegetation coverage is calculated from the vegetation normalized index, and the calculation formula is as follows:

[0065] NDVI = (NIR - RED) / (NIR + RED) (1);

[0066] FVC = (NDVI - NDVImin) / (NDVImax - NDVImin) (2);

[0067] Among them, NDVI is the normalized difference vegetation index value, NIR is the near-infrared band, RED is the red light band, FVC is the fractional vegetation cover, NDVImin is the normalized difference vegetation index value of pure bare soil, and NDVImax is the normalized difference vegetation index value of high vegetation cover; the calculation methods of NDVImin and NDVImax here are to extract 1000 bare soil pixels and vegetation pixels, extract their NDVI respectively, and use the mean value as NDVImin and NDVImax.

[0068] After obtaining the FVC layer, set the condition that when FVC < 0.8, it is the existence area of Torreya grandis sapling forests. Overlay analysis is performed on this layer and the preliminary screening results of Torreya grandis (that is, simultaneously satisfying that the random forest result is Torreya grandis and FVC < 0.8) to obtain the second screening result. The screening result here is to remove the evergreen forests mixed in the Torreya grandis sapling forests. The evergreen forests include Chinese fir forests, bamboo forests, and old Torreya grandis forests. There are also interferences from fields in the identification of Torreya grandis sapling forests.

[0069] Finally, remove the interference of fields in the identification of Torreya grandis sapling forests, which is reflected in that the annual peak NDVI of the fields is greater than that of the Torreya grandis sapling forests. Therefore, calculate the NDVI values of each month and perform annual maximum value synthesis to obtain the NDVImax raster image. After obtaining the NDVImax layer, set the condition that when NDVImax < 0.8, it is the non-field area. Overlay analysis is performed on this layer and the layer of the second screening result (that is, simultaneously satisfying that the random forest result is Torreya grandis, FVC < 0.8, and NDVImax < 0.8) to obtain the final screening result of Torreya grandis sapling forests. This result is a raster image with a spatial resolution of 10 m.

[0070] According to the different physiological characteristics of Torreya grandis at different ages, the present invention divides Torreya grandis as a whole into two categories: adult forests and sapling forests. After extraction and combination respectively, the overall distribution of Torreya grandis is obtained. For adult forests, sub-meter high-resolution images are used as the base map for annotation, and deep learning image recognition frameworks are used for recognition; for sapling forests, Sentinel multispectral data combined with fractional vegetation cover are used as basic data, and the random forest method is used to extract sapling forests, and the fields misclassified are removed by using the characteristic that the peak vegetation index of the fields is extremely high throughout the year. After the extraction of adult forests and sapling forests is completed respectively, the planting patches of Torreya grandis are combined, and the number of pixel points is counted under the Albers equal-area projection. Combining the spatial resolution information and slope information, the planting area of Torreya grandis is obtained.

[0071] Embodiment 2

[0072] As Figure 3 shown, a method for calculating the planting area of Torreya grandis based on age characteristics includes:

[0073] When obtaining the remote sensing multispectral data to be recognized, select Sentinel-2A data on the open-source platform Google Earth Engine, and screen the time-series data with cloud cover less than 20% in 2020 for median synthesis as the basis for random forest classification. Excessive cloud cover will lead to inaccurate remote sensing images. In this solution, cloud-free data in September 2020 is selected to calculate NDVI, and then the vegetation coverage FVC is calculated from NDVI. The data used here is September data, and only the vegetation coverage differences of different crops in September are considered. According to the investigation, for the old Torreya grandis forest, September is the fruit harvest period. After manual weeding under the forest, there is basically no vegetation coverage, but the overall canopy cover of the forest stand is relatively high, still reaching 50%. At other times, there is grass coverage under the forest, and the cover is higher; the overall canopy cover of the young Torreya grandis forest is relatively low, reaching 40% when there is grass coverage and about 5% when there is no grass coverage. For example, in Figure 2 Five locations are selected within the area in

[0074] to calculate the average value of FVC in September (plots without crops are selected additionally), and the results are obtained as 0.856 for adult forests, 0.971 for bamboo forests, 0.991 for Chinese fir forests, 0.475 for young forests, 0.897 for plots with crops, and 0.575 for plots without crops.

[0075] Then, the final extraction results of the young Torreya grandis forest are combined with the old Torreya grandis forest layer to summarize the overall distribution map of the Torreya grandis forest. The overlay analysis is realized using the raster calculator, that is: when a pixel is a young forest or an old forest, it is assigned a value of 1 for marking, and the rest of the area is marked with a value of 0. After completing this step, the raster result of the spatial distribution of the Torreya grandis forest is obtained. Summing up the pixel values of the entire image and multiplying by the pixel area can obtain the projected area of the Torreya grandis planting area. When summing up the pixel values of the entire image here, in the case of slope, the planted area of the slope is larger than the projected area. Therefore, for the calculation of the slope, a 30m resolution DEM raster image is obtained from the Geospatial Data Cloud, and resampled to 10m resolution to align with the Torreya grandis distribution raster. Use the slope calculation function in ArcMap to calculate the slope data from the DEM data. According to the Pythagorean theorem, there is: bottom area ÷ inclined plane area = COS(slope), and since the slope and the bottom area are known, the inclined plane area can be calculated.

[0075] For example: The original Torreya grandis distribution layer is marked with 0 and 1, where 1 represents the presence of Torreya grandis. When the slope at this location is 60°, then in the newly generated raster layer, the value at the corresponding location is 2, and the corresponding area is 200 square meters (spatial resolution of 10 meters, area is 10 * 10 * 2). After execution, a raster image of the inclined area of the Torreya grandis forest is obtained. At this time, the sum of the pixel values of all pixels in the image is calculated and multiplied by 100 to obtain the total planting area of Torreya grandis. Here, a pixel represents a 10 * 10 square area. If a large-scale topographic map is obtained, higher-precision terrain information can be generated from the topographic map. The present invention uses remote sensing multispectral data, sub-meter-level image data, and terrain data to extract arbor-type Torreya grandis and shrub-type Torreya grandis respectively, and uses the terrain data to assist in calculating the inclined planting area of Torreya grandis. This method can extract relatively complete Torreya grandis ground objects and obtain the planting area of Torreya grandis close to the statistical data rather than the projected area, which is of great significance for studying the planting distribution of Torreya grandis and guiding Torreya grandis production.

[0076] Example 3

[0077] As Figure 4 shown, a method and device for calculating the planting area of Torreya grandis based on age characteristics includes:

[0078] Acquisition module 10: used to acquire the remote sensing multispectral data to be identified and the raster layer of the old Torreya grandis forest, and the remote sensing multispectral data is used to identify young Torreya grandis forests;

[0079] First screening module 20: used to preliminarily screen the remote sensing multispectral data using the random forest algorithm to obtain the first screening result of the remote sensing multispectral data;

[0080] Second screening module 30: used to calculate the normalized difference vegetation index and vegetation coverage of the remote sensing multispectral data in sequence, and perform overlay analysis in combination with the first screening result to obtain the second screening result of the remote sensing multispectral data;

[0081] Calculation module 40: used to calculate the planting area of Torreya grandis according to the second screening result and the raster layer of the old Torreya grandis forest.

[0082] An embodiment of the above device may be: The acquisition module 10 acquires the remotely sensed multispectral data to be recognized and the raster layer of the old torreya grandis forest, and the remotely sensed multispectral data is used to identify young torreya grandis forests; The first screening module 20 preliminarily screens the remotely sensed multispectral data acquired by the acquisition module 10 by using the random forest algorithm to obtain the first screening result of the remotely sensed multispectral data; The second screening module 30 calculates the normalized difference vegetation index and vegetation coverage of the remotely sensed multispectral data successively, and performs overlay analysis in combination with the first screening result obtained by the first screening module 20 to obtain the second screening result of the remotely sensed multispectral data; The calculation module 40 calculates the planting area of torreya grandis according to the second screening result obtained by the second screening module 30 and the raster layer of the old torreya grandis forest.

[0083] Example 4

[0084] As Figure 5 shown, an electronic device includes a memory 501 and a processor 502. The memory 501 is used to store one or more computer instructions. Among them, the one or more computer instructions are executed by the processor 502 to implement the above-mentioned method for calculating the planting area of torreya grandis based on age characteristics.

[0085] Those skilled in the art can clearly understand that for the convenience and simplicity of description, the specific working process of the above-described electronic device can refer to the corresponding process in the foregoing method embodiments and will not be elaborated herein.

[0086] A computer-readable storage medium storing a computer program, where the computer program causes a computer to execute to implement the above-mentioned method for calculating the planting area of torreya grandis based on age characteristics.

[0087] Exemplarily, the computer program can be divided into one or more modules / units. One or more modules / units are stored in the memory 501 and executed by the processor 502, and the data I / O interface transmission is completed by the input interface 505 and the output interface 506 to complete the present invention. One or more modules / units can be a series of computer program instruction segments capable of performing specific functions, and the instruction segments are used to describe the execution process of the computer program in the computer device.

[0088] The computer device can be a computing device such as a desktop computer, a notebook, a palm computer, and a cloud server. The computer device may include, but is not limited to, a memory 501 and a processor 502. Those skilled in the art can understand that this embodiment is only an example of the computer device and does not constitute a limitation on the computer device. It may include more or fewer components, or combine certain components, or different components. For example, the computer device may further include an input device 507, a network access device, a bus, etc.

[0089] The processor 502 can be a central processing unit (CPU), or it can also be other general-purpose processors 502, digital signal processors 502 (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor 502 can be a microprocessor 502, or the processor 502 can also be any conventional processor 502, etc.

[0090] The memory 501 can be an internal storage unit of the computer device, such as the hard disk or memory of the computer device. The memory 501 can also be an external storage device of the computer device, such as a plug-in hard disk equipped on the computer device, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. Further, the memory 501 can also include both the internal storage unit and the external storage device of the computer device. The memory 501 is used to store computer programs and other programs and data required by the computer device. The memory 501 can also be used to temporarily store data in the output device 508, and the aforementioned storage media include various media that can store program codes, such as USB flash drives, mobile hard disks, read-only memory ROM503, random access memory RAM504, diskettes, or optical discs.

[0091] The above are only specific embodiments of the present invention, but the technical features of the present invention are not limited thereto. Any changes or modifications made by those skilled in the art within the scope of the present invention are covered by the patent scope of the present invention.

Claims

1. A calculation method for the planting area of Torreya grandis based on age characteristics, characterized in that, The age characteristics divide Torreya grandis forests into young Torreya grandis forests and old Torreya grandis forests, including: Obtain the remote sensing multispectral data to be recognized and the raster layer of the old Torreya grandis forest, where the remote sensing multispectral data is used to identify young Torreya grandis forests; Use the random forest algorithm to preliminarily screen the remote sensing multispectral data to obtain the first screening result of the remote sensing multispectral data; Calculate the normalized difference vegetation index (NDVI) and the fractional vegetation cover (FVC) of the remote sensing multispectral data in sequence, and perform overlay analysis in combination with the first screening result to obtain the second screening result of the remote sensing multispectral data; Calculate the planting area of Torreya grandis according to the second screening result and the raster layer of the old Torreya grandis forest; The step of calculating the normalized difference vegetation index (NDVI) and the fractional vegetation cover (FVC) of the remote sensing multispectral data in sequence, and performing overlay analysis in combination with the first screening result to obtain the second screening result of the remote sensing multispectral data includes: Calculate the normalized difference vegetation index (NDVI) and the fractional vegetation cover (FVC) of the remote sensing multispectral data in sequence, and the calculation formulas are: （1）； （2）； where NDVI is the value of the normalized difference vegetation index, NIR is the near-infrared band, RED is the red band, FVC is the fractional vegetation cover, NDVImin is the value of the normalized difference vegetation index of pure bare soil, and NDVImax is the value of the normalized difference vegetation index of high vegetation cover; Select the layers in the remote sensing multispectral data where the values of the normalized difference vegetation index and the fractional vegetation cover are less than the preset thresholds, and perform overlay analysis with the first screening result, and take the intersection to obtain the second screening result; The step of calculating the planting area of Torreya grandis according to the second screening result and the raster layer of the old Torreya grandis forest includes: Use the raster calculator to perform overlay analysis on the second screening result and the raster layer of the old Torreya grandis forest to obtain the raster result of the spatial distribution of Torreya grandis forests; Count the number of pixel points in the raster result of the spatial distribution of Torreya grandis forests, and combine the spatial resolution information and the slope information to obtain the planting area of Torreya grandis; When obtaining the remote sensing multispectral data to be recognized, select Sentinel-2A data on the open-source platform Google Earth Engine, screen the time-series data with cloud cover less than 20% in 2020 for median synthesis as the base map for random forest classification, screen the cloud-free data in September 2020 to calculate NDVI, and then calculate the fractional vegetation cover FVC from NDVI; Combine the final extraction result of young Torreya grandis forests with the layer of old Torreya grandis forests to summarize the overall distribution map of Torreya grandis forests; Use the raster calculator to implement overlay analysis, that is: when a pixel point is a young forest or an old forest, assign a value of 1 for marking, and mark the remaining areas with a value of 0 to obtain the raster result of the spatial distribution of Torreya grandis forests; sum the values of all image pixel points and multiply by the pixel area to obtain the projected area of the Torreya grandis planting area.

2. The Torreya grandis planting area calculation method based on age characteristics according to claim 1, characterized in that The step of using the random forest algorithm to preliminarily screen the remote sensing multispectral data to obtain the first screening result of the remote sensing multispectral data specifically includes: Use the random forest algorithm to screen the time-series data with cloud cover less than a certain threshold in the remote sensing multispectral data and perform median synthesis as the base map for random forest classification; Set the parameters of the random forest algorithm, and mark young Torreya grandis forests in the base map for random forest classification, and train to generate a trainer; Use the trainer to perform binary classification on the random forest classification base map to obtain a first screening result.

3. A Torreya grandis planting area calculation device based on age characteristics, characterized in that, Including An acquisition module: used to acquire remote sensing multispectral data to be recognized and a raster layer of Torreya grandis old forests, where the remote sensing multispectral data is used to identify young Torreya grandis forests; A first screening module: used to preliminarily screen the remote sensing multispectral data using the random forest algorithm to obtain a first screening result of the remote sensing multispectral data; A second screening module: used to calculate the normalized difference vegetation index and vegetation coverage of the remote sensing multispectral data in sequence, and perform overlay analysis in combination with the first screening result to obtain a second screening result of the remote sensing multispectral data; A calculation module: used to calculate the planting area of Torreya grandis according to the second screening result and the raster layer of Torreya grandis old forests; The second screening module includes: A calculation subunit: used to calculate the normalized difference vegetation index and vegetation coverage of the remote sensing multispectral data in sequence, and the calculation formulas are: （1）； （2）； Among them, NDVI is the value of the normalized difference vegetation index, NIR is the near-infrared band, RED is the red band, FVC is the vegetation coverage, NDVImin is the value of the normalized difference vegetation index of pure bare soil, and NDVImax is the value of the normalized difference vegetation index of high vegetation coverage; A screening subunit: used to select the layers in the remote sensing multispectral data where the value of the normalized difference vegetation index and the vegetation coverage are less than a preset threshold, and perform overlay analysis with the first screening result, and take the intersection to obtain a second screening result; The calculation module includes: An overlay unit: used to perform overlay analysis on the second screening result and the raster layer of Torreya grandis old forests using a raster calculator to obtain a raster result of the spatial distribution of Torreya grandis forests; A statistics unit: used to count the number of pixel points in the raster result of the spatial distribution of Torreya grandis forests, and combine the spatial resolution information and slope information to obtain the planting area of Torreya grandis; When acquiring the remote sensing multispectral data to be recognized, select sentinel-2A data on the open-source platform Google earth engine, screen the time-series data with a cloud cover less than 20% in 2020 for median synthesis as the random forest classification base, screen the cloud-free data in September 2020 to calculate NDVI, and then calculate the vegetation coverage FVC from NDVI; Combine the final extraction result of young Torreya grandis forests with the Torreya grandis old forest layer to summarize the overall distribution map of Torreya grandis forests; Use a raster calculator to implement overlay analysis, that is: when a pixel point is a young forest or an old forest, assign a value of 1 for marking, and mark the remaining areas with a value of 0 to obtain a raster result of the spatial distribution of Torreya grandis forests; sum the values of all image pixel points and multiply by the pixel point area to obtain the projected area of the Torreya grandis planting area.

4. The Torreya grandis planting area calculation device based on age characteristics according to claim 3, wherein The first screening module includes: A map generation unit: used to screen the time-series data with a cloud cover less than a certain threshold in the remote sensing multispectral data using the random forest algorithm and perform median synthesis as the random forest classification base map; A training unit: used to set the parameters of the random forest algorithm, and mark young Torreya grandis forests in the random forest classification base map, and train to generate a trainer; Taxonomic unit: used to perform binary classification on the random forest classification base map by using the trainer to obtain a first screening result.

5. A torreya grandis planting area calculation device based on age characteristics, characterized in that, It includes a memory and a processor. The memory is used to store one or more computer instructions. Among them, the one or more computer instructions are executed by the processor to implement a method for calculating the planting area of Torreya grandis based on age characteristics as described in claim 1 or 2.

6. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a computer, it implements a method for calculating the planting area of Torreya grandis based on age characteristics as described in claim 1 or 2.

Images