Field crop classification method and system based on growth peak multi-source remote sensing data

Abstract

The invention provides a field crop classification method and system based on growth peak multi-source remote sensing data, and the method comprises the steps: collecting the multi-source remote sensing data of a field crop growth season, and carrying out the preprocessing of the multi-source remote sensing data, and obtaining the preprocessed multi-source remote sensing data; performing time synthesis processing on the pre-processed partial multi-source remote sensing data to obtain time synthesis remote sensing data; determining a crop growth peak period in the time synthesis remote sensing data by adopting a method based on a vegetation index median threshold value; in the peak period of crop growth, space-time spectrum features are extracted for crop classification; and constructing a random forest classification model based on the spatio-temporal spectrum features, performing training to obtain an optimized classification model, and generating a field crop classification map by using the optimized classification model. The remote sensing data of the crop growth peak period are analyzed in a centralized mode, field crops can be accurately and efficiently classified, and the method is suitable for large-scale agricultural monitoring and management and has high timeliness and adaptability.

Description

Technical Field

[0001] The present invention relates to the field of agricultural remote sensing technology, and in particular to a field crop classification method and system based on multi-source remote sensing data during growth peak period. Background Art

[0002] In the field of traditional agricultural monitoring and crop classification, the application of remote sensing technology provides a large-scale and efficient monitoring method.

[0003] However, existing crop classification methods usually fail to fully utilize the critical period of peak crop growth, which is the stage when crop growth characteristics are most significant. Therefore, data collection outside the peak growth period may lead to insufficient information and low classification accuracy.

[0004] Therefore, developing a classification method specifically for the peak growth period of crops is the key to improving classification accuracy and applicability. Summary of the invention

[0005] The invention provides a field crop classification method and system based on multi-source remote sensing data during the growth peak period, so as to solve the defects existing in the prior art.

[0006] In a first aspect, the present invention provides a method for classifying field crops based on multi-source remote sensing data during peak growth period, comprising: Collecting multi-source remote sensing data during the growing season of field crops, and preprocessing the multi-source remote sensing data to obtain preprocessed multi-source remote sensing data; Performing time synthesis processing on part of the preprocessed multi-source remote sensing data to obtain time synthesis remote sensing data; Determining the peak growth period of crops in the temporally synthesized remote sensing data by using a vegetation index median threshold method; During the peak growth period of the crop, spatiotemporal spectral features are extracted for crop classification; A random forest classification model is constructed based on the spatiotemporal spectral features and trained to obtain an optimized classification model, and a field crop classification map is generated using the optimized classification model.

[0007] According to a field crop classification method based on multi-source remote sensing data during the growth peak period provided by the present invention, multi-source remote sensing data of field crops during the growing season is collected, and the multi-source remote sensing data is preprocessed to obtain preprocessed multi-source remote sensing data, including: Acquire Sentinel-1, Sentinel-2 and digital elevation model during the growing season of field crops as the multi-source remote sensing data; The multi-source remote sensing data is subjected to clipping, mean synthesis, linear interpolation and SG filtering smoothing processing to obtain the pre-processed multi-source remote sensing data.

[0008] According to a field crop classification method based on multi-source remote sensing data during the peak growth period provided by the present invention, time synthesis processing is performed on part of the pre-processed multi-source remote sensing data to obtain time-synthesized remote sensing data, including: The pre-processed Sentinel-1 and Sentinel-2 remote sensing data in the growing season of the field crops are respectively averaged and synthesized by using a growing season synthesis method to obtain a synthetic image representing the entire growing season for constructing spatial features and spectral features.

[0009] According to a field crop classification method based on multi-source remote sensing data during the peak growth period provided by the present invention, time synthesis processing is performed on part of the pre-processed multi-source remote sensing data to obtain time-synthesized remote sensing data, and the method further includes: The pre-processed Sentinel-1 and Sentinel-2 remote sensing data in the growing season of the field crops are synthesized once every ten days by a ten-day synthesis method to generate a ten-day synthetic image for extracting time series features.

[0010] According to a field crop classification method based on multi-source remote sensing data of peak growth period provided by the present invention, a vegetation index median threshold method is used to determine the crop peak growth period in the time-synthesized remote sensing data, including: Analyze the ten-day synthetic image data throughout the entire phenological period and calculate the normalized vegetation index at each time point; Statistically sort all normalized vegetation indices and obtain the median; All time periods corresponding to values ​​higher than the median are determined to be the peak growth periods of the crops.

[0011] According to a field crop classification method based on multi-source remote sensing data during the peak growth period provided by the present invention, spatiotemporal spectral features are extracted for crop classification during the peak growth period of the crop, comprising: Calculating the normalized vegetation index, enhanced vegetation index and red edge position index during the crop growth peak period in the ten-day synthetic image, and using the normalized vegetation index, the enhanced vegetation index and the red edge position index as time series features; Calculating the B2, B3, B4, B5, B11 and B12 band data representing the peak growth period of crops in the synthetic image of the entire growing season, and using the B2, B3, B4, B5, B11 and B12 band data as spectral features; The GLCM texture features, backscatter coefficient and slope in the digital elevation model of the crop growth peak in the synthetic image representing the entire growing season are calculated, and contrast, moment of inertia, variance of difference, backscatter coefficient and slope are used as spatial features.

[0012] According to a field crop classification method based on multi-source remote sensing data during the peak growth period provided by the present invention, a random forest classification model is constructed based on the spatiotemporal spectral features and trained to obtain an optimized classification model, and a field crop classification map is generated using the optimized classification model, including: Constructing a random forest classification model, wherein the random forest classification model uses the spatiotemporal spectral features extracted during the peak growth period of crops; Training, parameter optimization and verification of the random forest classification model to obtain a trained optimized classification model; The optimized classification model is used to classify field crops to generate the field crop classification map.

[0013] In a second aspect, the present invention also provides a field crop classification system based on multi-source remote sensing data during peak growth period, comprising: An acquisition preprocessing module is used to acquire multi-source remote sensing data during the growing season of field crops, and preprocess the multi-source remote sensing data to obtain preprocessed multi-source remote sensing data; A synthesis module is used to perform time synthesis processing on part of the pre-processed multi-source remote sensing data to obtain time-synthesized remote sensing data; A determination module, used for determining the peak growth period of crops in the temporal synthetic remote sensing data by using a vegetation index median threshold method; An extraction module, used for extracting spatiotemporal spectral features for crop classification during the peak growth period of the crop; A classification module is used to construct a random forest classification model based on the spatiotemporal spectral features and train it to obtain an optimized classification model, and use the optimized classification model to generate a field crop classification map.

[0014] In a third aspect, the present invention also provides an electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, a field crop classification method based on multi-source remote sensing data during peak growth periods as described above is implemented.

[0015] In a fourth aspect, the present invention further provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements any of the above-described methods for classifying field crops based on multi-source remote sensing data during peak growth periods.

[0016] The field crop classification method and system based on multi-source remote sensing data during the peak growth period provided by the present invention can accurately and efficiently classify field crops by centrally analyzing remote sensing data during the peak growth period of crops, and is suitable for large-scale agricultural monitoring and management, and has strong timeliness and adaptability. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0018] Figure 1 This is one of the flow diagrams of the field crop classification method based on multi-source remote sensing data during the peak growth period provided by the present invention; Figure 2 This is the second flow chart of the field crop classification method based on multi-source remote sensing data during the peak growth period provided by the present invention; Figure 3 It is a structural schematic diagram of a field crop classification system based on multi-source remote sensing data during the peak growth period provided by the present invention; Figure 4 It is a structural schematic diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0019] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0020] Figure 1 FIG. 1 is one of the flow charts of the field crop classification method based on multi-source remote sensing data during the peak growth period provided by an embodiment of the present invention. Figure 1 As shown, including: Step 100: collecting multi-source remote sensing data during the growing season of field crops, and preprocessing the multi-source remote sensing data to obtain preprocessed multi-source remote sensing data; Step 200: performing time synthesis processing on the pre-processed part of the multi-source remote sensing data to obtain time synthesis remote sensing data; Step 300: Determine the crop growth peak period in the temporal synthetic remote sensing data by using a vegetation index median threshold method; Step 400: extracting spatiotemporal spectral features for crop classification during the crop growth peak period; Step 500: construct a random forest classification model based on the spatiotemporal spectral features and train it to obtain an optimized classification model, and use the optimized classification model to generate a field crop classification map.

[0021] In the embodiment of the present invention, a certain province in China is used as the research area, five field crops such as wheat, corn, peanuts, soybeans, and rice are used as research objects, and 2019 is the research time.

[0022] Specifically, if Figure 2 As shown in the figure, multi-source remote sensing data such as Sentinel-1 SAR, Sentinel-2 multispectral and SRTM digital elevation model covering the growing season of field crops are collected and preprocessed; the preprocessed remote sensing data Sentinel-1 SAR and Sentinel-2 multispectral are temporally synthesized using two methods: growing season synthesis and ten-day synthesis; a method based on the median threshold of vegetation index (such as NDVI) is used to determine the peak growth period of crops; within the determined peak growth period of crops, key time-space-spectral features are extracted for subsequent crop classification; a random forest classification model combining the time-space-spectral features extracted during the peak growth period of crops is used, and the optimized model is used to classify the field crops in the area to generate a field crop classification map.

[0023] The present invention can accurately and efficiently classify field crops by centrally analyzing remote sensing data during the peak growth period of crops, and is suitable for large-scale agricultural monitoring and management, and has strong timeliness and adaptability.

[0024] Based on the above embodiment, step 100 includes: Acquire Sentinel-1, Sentinel-2 and SRTM digital elevation models during the growing season of field crops as the multi-source remote sensing data; The multi-source remote sensing data is subjected to clipping, mean synthesis, linear interpolation and SG filtering smoothing processing to obtain the pre-processed multi-source remote sensing data.

[0025] Specifically, the long-time series radar image Sentinel-1 and the optical image Sentinel-2 in the embodiment of the present invention are both downloaded through the GEE (Google Earth Engine) cloud platform.

[0026] Here, Sentinel-1, Sentinel-2 and SRTM digital elevation models covering the growing season of field crops (the growing season for wheat is from October to June of the following year, and the growing season for rice, peanuts, soybeans and corn is from June to October) are selected as remote sensing data sources, and cropping, mean synthesis, linear interpolation and SG filtering smoothing are performed to make the processed image data more consistent and smooth, thereby reducing the impact of noise and data missing and improving the quality and continuity of time series data.

[0027] Based on the above embodiment, step 200 includes: The pre-processed Sentinel-1 and Sentinel-2 remote sensing data in the growing season of the field crops are respectively averaged and synthesized by using a growing season synthesis method to obtain a synthetic image representing the entire growing season for constructing spatial features and spectral features.

[0028] The pre-processed Sentinel-1 and Sentinel-2 remote sensing data in the growing season of the field crops are synthesized once every ten days by a ten-day synthesis method to generate a ten-day synthetic image for extracting time series features.

[0029] Specifically, the embodiment of the present invention adopts a growing season synthesis method to average and synthesize all optical and radar images within the growth cycle of field crops to generate a synthetic image representing the entire growing season. The growing season synthetic image is used to support the subsequent construction of spatial features and spectral features. In this embodiment, the growing season synthetic image of wheat is an average synthesis of all images from October 2019 to June 2020, and the growing season synthetic image of rice, peanuts, soybeans, and corn is an average synthesis of all images from June 2019 to October 2019; All optical and radar images during the growing season are synthesized once every ten days using a ten-day synthesis method to generate ten-day synthetic images, which are mainly used to extract subsequent temporal features. In this embodiment, the ten-day synthetic image of wheat is an average synthesis of all images from October 2019 to June 2020 at a frequency of ten days, and the synthetic images of the growing season of rice, peanuts, soybeans, and corn are an average synthesis of all images from June 2019 to October 2019 at a frequency of ten days.

[0030] Based on the above embodiment, step 300 includes: Analyze the ten-day synthetic image data throughout the entire phenological period and calculate the normalized vegetation index at each time point; Statistically sort all normalized vegetation indices and obtain the median; All time periods corresponding to values ​​higher than the median are determined to be the peak growth periods of the crops.

[0031] Specifically, the embodiment of the present invention analyzes the remote sensing data of the entire phenological period (i.e., the entire period from the beginning of crop growth to harvest) within the ten-day synthetic image, and calculates the Normalized Difference Vegetation Index (NDVI) at each time point; All calculated NDVI values ​​for this period are statistically sorted and the median of these values ​​is found. The median is used as a statistical dividing point to divide the NDVI values ​​into two parts: values ​​below the median and values ​​above the median; On this basis, the time periods corresponding to all NDVI values ​​higher than the median are selected and defined as the peak growth period of crops.

[0032] Based on the above embodiment, step 400 includes: Calculating the normalized vegetation index, enhanced vegetation index and red edge position index during the crop growth peak period in the ten-day synthetic image, and using the normalized vegetation index, the enhanced vegetation index and the red edge position index as time series features; Calculating the B2, B3, B4, B5, B11 and B12 band data representing the peak growth period of crops in the synthetic image of the entire growing season, and using the B2, B3, B4, B5, B11 and B12 band data as spectral features; The GLCM texture features, backscatter coefficient and slope in the digital elevation model of the crop growth peak in the synthetic image representing the entire growing season are calculated, and contrast, moment of inertia, variance of difference, backscatter coefficient and slope are used as spatial features.

[0033] Specifically, the embodiment of the present invention calculates the Normalized Difference Vegetation Index (NDVI), Enhanced Vegetation Index (EVI), and Red Edge Position Index (REP) during the peak growth period in a ten-day synthetic image, and uses NDVI, EVI, and REP as time series features; Calculate the six bands of B2, B3, B4, B5, B11, and B12 during the peak growth period in the growing season synthetic image, and use B2, B3, B4, B5, B11, and B12 as spectral features; The GLCM texture features, backscatter coefficient and slope in the digital elevation model during the peak growth period in the growing season synthetic image were calculated, and contrast, moment of inertia, variance of difference (Dvar), backscatter coefficient (NRCS) and slope were used as spatial features.

[0034] Based on the above embodiment, step 500 includes: Constructing a random forest classification model, wherein the random forest classification model uses the spatiotemporal spectral features extracted during the peak growth period of crops; Training, parameter optimization and verification of the random forest classification model to obtain a trained optimized classification model; The optimized classification model is used to classify field crops to generate the field crop classification map.

[0035] Specifically, the embodiment of the present invention constructs a random forest classification model, which comprehensively uses the time-space-spectral features extracted from the peak growth period of crops, including NDVI, EVI, REP, B2, B3, B4, B5, B11, B12, Contrast, Inertia, Dvar, NRCS, Slope and other features. The sample data is randomly divided into a training set and a test set in a ratio of 8:2, and the random forest model is trained, parameter optimized and verified to obtain a trained random forest model. The optimized model is used to accurately classify the field crops in the region, and a classification map of field crops in the province in 2019 is generated.

[0036] Compared with the prior art, the present invention has the following beneficial effects: The accuracy and adaptability of crop classification are improved by focusing on the key features of the peak growth period of crops and reducing the interference of data from irrelevant periods. The model is more suitable for large-scale agricultural monitoring and is suitable for different regions and different types of crops. An efficient data processing process is provided, from data collection to the generation of classification results. The entire process is highly automated and meets the needs of the current development of intelligent agriculture.

[0037] The following describes the field crop classification system based on multi-source remote sensing data during the peak growth period provided by the present invention. The field crop classification system based on multi-source remote sensing data during the peak growth period described below and the field crop classification method based on multi-source remote sensing data during the peak growth period described above can be referenced to each other.

[0038] Figure 3 is a schematic diagram of the structure of a field crop classification system based on multi-source remote sensing data during the peak growth period provided by an embodiment of the present invention. Figure 3 As shown, it includes: a collection preprocessing module 31, a synthesis module 32, a determination module 33, an extraction module 34 and a classification module 35, wherein: The acquisition and preprocessing module 31 is used to acquire multi-source remote sensing data during the growing season of field crops, and preprocess the multi-source remote sensing data to obtain preprocessed multi-source remote sensing data; the synthesis module 32 is used to perform time synthesis processing on part of the preprocessed multi-source remote sensing data to obtain time-synthesized remote sensing data; the determination module 33 is used to determine the peak growth period of crops in the time-synthesized remote sensing data using a vegetation index median threshold method; the extraction module 34 is used to extract spatiotemporal spectral features for crop classification during the peak growth period of crops; the classification module 35 is used to construct a random forest classification model based on the spatiotemporal spectral features and train it to obtain an optimized classification model, and use the optimized classification model to generate a field crop classification map.

[0039] Figure 4 An example of a physical structure diagram of an electronic device is shown in FIG. Figure 4As shown, the electronic device may include: a processor 410, a communication interface 420, a memory 430 and a communication bus 440, wherein the processor 410, the communication interface 420 and the memory 430 communicate with each other through the communication bus 440. The processor 410 may call the logic instructions in the memory 430 to execute a field crop classification method based on multi-source remote sensing data of a growth peak period, the method comprising: collecting multi-source remote sensing data of a field crop growing season, pre-processing the multi-source remote sensing data to obtain pre-processed multi-source remote sensing data; performing time synthesis processing on part of the pre-processed multi-source remote sensing data to obtain time synthesis remote sensing data; determining the crop growth peak period in the time synthesis remote sensing data by using a vegetation index median threshold method; extracting spatiotemporal spectral features for crop classification during the crop growth peak period; constructing a random forest classification model based on the spatiotemporal spectral features and training to obtain an optimized classification model, and generating a field crop classification map by using the optimized classification model.

[0040] In addition, the logic instructions in the above-mentioned memory 430 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on this understanding, the technical solution of the present invention, in essence, 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, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0041] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, it is implemented to execute the field crop classification method based on multi-source remote sensing data during the peak growth period provided by the above-mentioned methods. The method includes: collecting multi-source remote sensing data during the growing season of field crops, preprocessing the multi-source remote sensing data to obtain preprocessed multi-source remote sensing data; performing time synthesis processing on part of the preprocessed multi-source remote sensing data to obtain time-synthesized remote sensing data; using a vegetation index median threshold method to determine the crop growth peak period in the time-synthesized remote sensing data; extracting spatiotemporal spectral features for crop classification during the crop growth peak period; constructing a random forest classification model based on the spatiotemporal spectral features and training it to obtain an optimized classification model, and using the optimized classification model to generate a field crop classification map.

[0042] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0043] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0044] 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 embodiments of the present invention.

Claims

1. A method for classifying field crops based on multi-source remote sensing data during peak growth period, characterized in that: include: Collecting multi-source remote sensing data during the growing season of field crops, and preprocessing the multi-source remote sensing data to obtain preprocessed multi-source remote sensing data; Performing time synthesis processing on part of the preprocessed multi-source remote sensing data to obtain time synthesis remote sensing data; Determining the peak growth period of crops in the temporal synthetic remote sensing data by using a vegetation index median threshold method; During the peak growth period of the crop, spatiotemporal spectral features are extracted for crop classification; A random forest classification model is constructed based on the spatiotemporal spectral features and trained to obtain an optimized classification model, and a field crop classification map is generated using the optimized classification model.

2. The method for classifying field crops based on multi-source remote sensing data during peak growth period according to claim 1, characterized in that: Collecting multi-source remote sensing data during the growing season of field crops, and preprocessing the multi-source remote sensing data to obtain preprocessed multi-source remote sensing data, including: Acquire Sentinel-1, Sentinel-2 and digital elevation model during the growing season of field crops as the multi-source remote sensing data; The multi-source remote sensing data is subjected to clipping, mean synthesis, linear interpolation and SG filtering smoothing processing to obtain the pre-processed multi-source remote sensing data.

3. The method for classifying field crops based on multi-source remote sensing data during peak growth period according to claim 1, characterized in that: Temporal synthesis processing is performed on part of the preprocessed multi-source remote sensing data to obtain temporal synthetic remote sensing data, including: The pre-processed Sentinel-1 and Sentinel-2 remote sensing data in the growing season of the field crops are respectively averaged and synthesized by using a growing season synthesis method to obtain a synthetic image representing the entire growing season for constructing spatial features and spectral features.

4. The method for classifying field crops based on multi-source remote sensing data during peak growth period according to claim 3, characterized in that: Performing time synthesis processing on the pre-processed part of the multi-source remote sensing data to obtain time-synthesized remote sensing data, further comprising: The pre-processed Sentinel-1 and Sentinel-2 remote sensing data in the growing season of the field crops are synthesized once every ten days by a ten-day synthesis method to generate a ten-day synthetic image for extracting time series features.

5. The method for classifying field crops based on multi-source remote sensing data during peak growth period according to claim 1, characterized in that: The peak growth period of crops in the temporal synthetic remote sensing data is determined by using a vegetation index median threshold method, including: Analyze the ten-day synthetic image data throughout the entire phenological period and calculate the normalized vegetation index at each time point; Statistically sort all normalized vegetation indices and obtain the median; All time periods corresponding to values ​​higher than the median are determined to be the peak growth periods of the crops.

6. The method for classifying field crops based on multi-source remote sensing data during peak growth period according to claim 4, characterized in that: During the peak growth period of the crop, spatiotemporal spectral features are extracted for crop classification, including: Calculating the normalized vegetation index, enhanced vegetation index and red edge position index during the crop growth peak period in the ten-day synthetic image, and using the normalized vegetation index, the enhanced vegetation index and the red edge position index as time series features; Calculating the B2, B3, B4, B5, B11 and B12 band data representing the peak growth period of crops in the synthetic image of the entire growing season, and using the B2, B3, B4, B5, B11 and B12 band data as spectral features; The GLCM texture features, backscatter coefficient and slope in the digital elevation model of the crop growth peak in the synthetic image representing the entire growing season are calculated, and contrast, moment of inertia, variance of difference, backscatter coefficient and slope are used as spatial features.

7. The method for classifying field crops based on multi-source remote sensing data during peak growth period according to claim 1, characterized in that: A random forest classification model is constructed based on the spatiotemporal spectral features and trained to obtain an optimized classification model, and a field crop classification map is generated using the optimized classification model, including: Constructing a random forest classification model, wherein the random forest classification model uses the spatiotemporal spectral features extracted during the peak growth period of crops; Training, parameter optimization and verification of the random forest classification model to obtain a trained optimized classification model; The optimized classification model is used to classify field crops to generate the field crop classification map.

8. A field crop classification system based on multi-source remote sensing data during peak growth period, characterized in that: include: An acquisition preprocessing module is used to acquire multi-source remote sensing data during the growing season of field crops, and preprocess the multi-source remote sensing data to obtain preprocessed multi-source remote sensing data; A synthesis module is used to perform time synthesis processing on part of the pre-processed multi-source remote sensing data to obtain time-synthesized remote sensing data; A determination module, used for determining the peak growth period of crops in the temporal synthetic remote sensing data by using a vegetation index median threshold method; An extraction module, used for extracting spatiotemporal spectral features for crop classification during the peak growth period of the crop; A classification module is used to construct a random forest classification model based on the spatiotemporal spectral features and train it to obtain an optimized classification model, and use the optimized classification model to generate a field crop classification map.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the field crop classification method based on multi-source remote sensing data during the peak growth period is implemented as described in any one of claims 1 to 7.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for classifying field crops based on multi-source remote sensing data during the peak growth period as claimed in any one of claims 1 to 7 is implemented.

Images