Crop harvesting progress monitoring method and system, electronic equipment and storage medium

Abstract

The invention discloses a crop harvesting progress monitoring method and system, electronic equipment and a storage medium, and relates to the technical field of crop harvesting progress monitoring, and the method comprises the steps: obtaining a target region satellite image and the same-period agricultural machine track data; processing the crop planting distribution data to generate a mask pattern, and processing an image; carrying out cloud detection; generating a clustering result graph based on the cloud and the crop mask; agricultural machine track points are mapped to a clustering graph, the harvesting state is judged, and a classification result graph layer is generated; and counting the harvested and total planting areas based on the image layer and the crop mask, and calculating the progress percentage. The method introduces the same-period agricultural machine trajectory data to participate in determination, reduces the dependence on an artificial sample and a supervision model, and improves the generalization ability. In combination with multi-source data processing, timeliness is enhanced, day-by-day updating and automatic monitoring are supported, and a reliable basis is provided for agricultural management.

Description

Technical Field

[0001] This invention relates to the field of crop harvesting progress monitoring technology, and in particular to a crop harvesting progress monitoring method, system, electronic device and storage medium. Background Technology

[0002] Crop harvesting progress monitoring refers to the method of dynamically acquiring and evaluating the crop harvesting process using technologies such as manual surveys and remote sensing observation. Achieving accurate, near-real-time harvesting progress monitoring helps optimize agricultural resource allocation and production layout, and is of great significance for promoting the development of precision agriculture and improving agricultural production management in my country. Traditional monitoring methods mainly rely on manual field surveys and records, which have significant subjectivity and time lag, resulting in low efficiency and difficulty in meeting the demand for large-scale, high-timeliness monitoring and updates during peak crop harvesting periods. With the development of Earth observation technology, remote sensing technology, with its advantages of wide coverage, short revisit cycles, and repeatable observations, has gradually become an important technical means of acquiring crop harvesting progress, providing new data support for agricultural decision-making and management.

[0003] Existing methods for monitoring crop harvesting progress mainly rely on the spectral features, texture features, and polarization parameters of remote sensing images for identification. For example, Xu Feifei et al. (Zhongke Hexin Remote Sensing Technology (Suzhou) Co., Ltd. A Remote Sensing Monitoring Method for Crop Harvesting Progress: 202311815092.4 [P]. 2024-04-02.) identified harvested and unharvested pixels in planting plots by constructing spectral and texture feature indices, thereby determining the crop harvesting progress. Zhao Chunjiang et al. (Beijing Agricultural Information Technology Research Center. A Method and Device for Monitoring Crop Harvesting Progress: 201410610674.3 [P]. 2015-02-18.) used radar images during the harvest period for polarization decomposition, calculating the average polarization parameters of each plot to determine whether the crop had been harvested. Zheng Jia et al. (Northeast Institute of Geography and Agroecology, Chinese Academy of Sciences. A method for monitoring maize harvesting progress with high spatiotemporal resolution: 202310059066.7 [P]. 2023-04-14.) combined Sentinel-1 radar imagery and Sentinel-2 optical imagery, comprehensively analyzed polarization characteristics and spectral reflectance curves, constructed a multi-feature set, and realized spatiotemporal dynamic mapping of maize harvesting progress.

[0004] However, the aforementioned existing methods still have significant limitations and are difficult to meet practical application needs. First, their generalization ability is poor: most existing methods rely on a large number of ground samples to build supervised models, making them highly sensitive to sample characteristics and regional environments. This makes it difficult to adapt to different crop types, regional conditions, and interannual environmental changes, resulting in insufficient model transferability and limiting automated promotion over large areas. Second, their timeliness is low: current methods are mostly based on publicly available remote sensing data such as Sentinel-1 / 2, which are limited by the satellite revisit cycle (usually 5–6 days), making daily image coverage impossible. Therefore, they cannot support daily dynamic updates of harvest progress, restricting their timeliness and practicality in agricultural production management.

[0005] Therefore, given the shortcomings of existing harvest progress monitoring methods in terms of generalization and timeliness, there is an urgent need to develop a crop harvest progress monitoring technology that can adapt to various local conditions and crop types and achieve near real-time updates, so as to improve the precision scheduling and management capabilities of agricultural production. Summary of the Invention

[0006] The technical problem to be solved by this invention is to address the shortcomings of existing technologies, and specifically provides a method, system, electronic device, and storage medium for monitoring crop harvesting progress, as detailed below: 1) In a first aspect, the present invention provides a method for monitoring crop harvesting progress, the specific technical solution of which is as follows: Acquire satellite imagery of the target area and acquire agricultural machinery trajectory data of the target area at the same imaging time as the satellite imagery; The crop planting distribution data of the target area is cropped and resampled to the same spatial range and resolution as the satellite image to obtain a crop planting mask map. The pixel values ​​of non-crop planting areas in the satellite image are set to 0 to obtain the processed satellite image. Cloud detection is performed on the processed satellite imagery to obtain cloud detection results; Based on cloud detection results and crop planting mask maps, a clustering result map containing multiple categories is generated; Map the agricultural machinery trajectory points in the agricultural machinery trajectory data to the clustering result map, determine whether each category is a harvested state or an unharvested state, and generate a classification result layer; Based on the classification result layer and crop planting mask map, the harvested area and total planted area within the target area are statistically analyzed, and the harvest progress percentage is calculated.

[0007] The beneficial effects of the crop harvesting progress monitoring method provided by this invention are as follows: By introducing agricultural machinery trajectory data contemporaneous with satellite imagery and mapping it to the clustering results map for ground feature status determination, the reliance on large amounts of manual samples and supervised models is effectively reduced. This enhances the model's adaptability and generalization ability under different regional, crop, and annual conditions, supporting large-scale automated application. The comprehensive utilization of multi-source satellite data, combined with cloud detection and crop planting masking, mitigates the limitations of long revisit cycles from single data sources, significantly improving data timeliness and availability, and providing a technical foundation for daily dynamic updates of crop harvesting progress. By generating categorized result layers and automatically calculating the harvested area and total planted area based on pixels, the percentage of harvest progress can be calculated quickly and objectively, significantly improving the accuracy and practicality of monitoring results and providing stable and reliable data support for the rational allocation of agricultural resources and production management.

[0008] Based on the above scheme, the crop harvesting progress monitoring method of the present invention can be further improved as follows.

[0009] Furthermore, cloud detection is performed on the processed satellite imagery to obtain cloud detection results, including: The normalized vegetation index (NDI) of the processed satellite imagery is calculated, and cloud detection is performed on the processed satellite imagery based on the NDI, the reflectance of the blue light band in the processed satellite imagery, and the reflectance of the near-infrared light band in the processed satellite imagery to obtain cloud detection results.

[0010] The beneficial effects of adopting the above-mentioned further scheme are as follows: By comprehensively utilizing the normalized vegetation index, blue light band reflectance, and near-infrared light band reflectance for cloud detection, the accuracy and robustness of cloud pixel recognition are significantly improved. It effectively overcomes the limitations of relying solely on spectral features, which is easily affected by factors such as surface vegetation cover and soil background, effectively distinguishing clouds from bright ground surfaces and reducing the probability of false and false cloud detections. Accurate cloud detection results ensure the reliability of subsequent cluster analysis and agricultural machinery trajectory data mapping, thus laying a high-quality data foundation for accurately determining crop harvesting status and calculating the percentage of harvest progress, and enhancing the adaptability and stability of the entire technical solution under complex observation conditions.

[0011] Furthermore, based on the cloud detection results and crop planting mask map, a clustering result map containing multiple categories is generated, including: Based on cloud detection results and crop planting mask map, the processed satellite image is masked to retain effective pixels that are not cloud pixels and are within the crop planting area. The reflectance values ​​of the blue light band, green light band, and red light band of each effective pixel are extracted and combined into a feature vector. All feature vectors are clustered to generate a clustering result map containing multiple categories.

[0012] The beneficial effects of adopting the above-mentioned further scheme are as follows: By combining cloud detection results with crop planting mask maps for collaborative masking, cloud-covered pixels and non-crop planting areas are effectively eliminated, while accurately retaining effective pixels for subsequent analysis, significantly improving data quality and processing efficiency. By extracting the blue, green, and red reflectance values ​​of each effective pixel and constructing feature vectors for clustering, the spectral response differences of crops under different harvesting states can be fully explored, thereby accurately generating clustering result maps containing multiple land cover categories. This method reduces reliance on prior human knowledge and a large number of samples, enhances the model's adaptability and automation level under different regions and crop types, and provides reliable and objective intermediate data results for subsequent agricultural machinery trajectory mapping and harvesting status determination, effectively supporting the accuracy of harvest progress percentage calculation.

[0013] Furthermore, the agricultural machinery trajectory points in the agricultural machinery trajectory data are mapped to the clustering result map to determine whether each category is a harvested or unharvested state, and a classification result layer is generated, including: The agricultural machinery trajectory points in the agricultural machinery trajectory data are mapped to the clustering result map. The number of agricultural machinery trajectory points and the number of pixels in each category in the clustering result map are counted. Based on the preset judgment logic, each category is determined to be either harvested or unharvested, and a classification result layer is generated.

[0014] The beneficial effects of adopting the above-mentioned further scheme are as follows: Mapping the agricultural machinery trajectory points in the agricultural machinery trajectory data to the clustering result map, and statistically analyzing the number of trajectory points and pixels within each category, automatically determining the crop harvesting status based on preset logic significantly improves the objectivity and automation level of status recognition. This method effectively utilizes the strong correlation between agricultural machinery operation and crop harvesting status, overcoming the limitations of traditional methods that rely solely on remote sensing features, are easily affected by environmental interference, and have weak generalization ability. By generating accurate classification result layers, a reliable data foundation is provided for subsequent calculations of harvested area and total planted area, ensuring the accuracy of the harvest progress percentage statistics, enhancing the practicality and reliability of the entire technical solution at the regional scale, and meeting the needs of agricultural production management for near real-time, large-scale monitoring.

[0015] 2) In a second aspect, the present invention also provides a crop harvesting progress monitoring system, the specific technical solution of which is as follows: It includes an image data acquisition module, a data processing module, a cloud detection module, a clustering module, a classification result layer generation module, and a harvest progress determination module; The image data acquisition module is used to: acquire satellite images of the target area, and acquire agricultural machinery trajectory data of the target area at the same imaging time as the satellite images; The data processing module is used to: crop the crop planting distribution data of the target area and resample it to the same spatial range and resolution as the satellite image to obtain a crop planting mask map, and set the pixel value of the non-crop planting area in the satellite image to 0 to obtain the processed satellite image; The cloud detection module is used to perform cloud detection on the processed satellite imagery and obtain the cloud detection results. The clustering module is used to generate clustering result maps containing multiple categories based on cloud detection results and crop planting mask maps; The classification result layer generation module is used to: map the agricultural machinery trajectory points in the agricultural machinery trajectory data to the clustering result map, determine whether each category is a harvested state or an unharvested state, and generate a classification result layer; The harvest progress determination module is used to: based on the classification result layer and crop planting mask map, count the harvested area and total planted area within the target area, and calculate the harvest progress percentage.

[0016] Based on the above solution, the crop harvesting progress monitoring system of the present invention can be further improved as follows.

[0017] Furthermore, the cloud detection module is specifically used for: The normalized vegetation index (NDI) of the processed satellite imagery is calculated, and cloud detection is performed on the processed satellite imagery based on the NDI, the reflectance of the blue light band in the processed satellite imagery, and the reflectance of the near-infrared light band in the processed satellite imagery to obtain cloud detection results.

[0018] Furthermore, the clustering module is specifically used for: Based on cloud detection results and crop planting mask map, the processed satellite image is masked to retain effective pixels that are not cloud pixels and are within the crop planting area. The reflectance values ​​of the blue light band, green light band, and red light band of each effective pixel are extracted and combined into a feature vector. All feature vectors are clustered to generate a clustering result map containing multiple categories.

[0019] Furthermore, the classification result layer generation module is specifically used for: The agricultural machinery trajectory points in the agricultural machinery trajectory data are mapped to the clustering result map. The number of agricultural machinery trajectory points and the number of pixels in each category in the clustering result map are counted. Based on the preset judgment logic, each category is determined to be either harvested or unharvested, and a classification result layer is generated.

[0020] 3) In a third aspect, the present invention also provides an electronic device, the electronic device including a processor coupled to a memory, the memory storing at least one computer program, the at least one computer program being loaded and executed by the processor, so that the electronic device implements any of the above-mentioned crop harvesting progress monitoring methods.

[0021] 4) In a fourth aspect, the present invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements any of the above-described crop harvesting progress monitoring methods.

[0022] It should be noted that the beneficial effects of the technical solutions of the second to fourth aspects of the present invention and their corresponding possible implementations can be found in the above description of the technical effects of the first aspect and its corresponding possible implementations, and will not be repeated here. Attached Figure Description

[0023] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments of the present invention will be briefly introduced below: Figure 1 This is a flowchart illustrating a method for monitoring crop harvesting progress according to an embodiment of the present invention. Figure 2 This is a schematic diagram of the structure of a crop harvesting progress monitoring system according to an embodiment of the present invention; Figure 3 This is a schematic diagram of the structure of an electronic device according to an embodiment of the present invention. Detailed Implementation

[0024] The principles and features of the present invention are described below. The examples given are only for explaining the present invention and are not intended to limit the scope of the present invention.

[0025] The technical solution of the present invention and how the technical solution of the present invention solves the above-mentioned technical problems are described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of the present invention will now be described with reference to the accompanying drawings.

[0026] like Figure 1 As shown, an embodiment of the present invention provides a method for monitoring crop harvesting progress, comprising the following steps: S1. Acquire satellite imagery of the target area and acquire agricultural machinery trajectory data of the target area at the same imaging time as the satellite imagery; Among them, satellite imagery with high temporal resolution and sufficient spatial resolution is selected. The satellite imagery includes red, green, blue and near-infrared bands to facilitate subsequent calculation of normalized vegetation index and cloud detection. The selected satellite imagery should completely cover the target area, and priority should be given to satellite imagery with less cloud coverage and high imaging quality to minimize the interference of cloud layer on effective information extraction.

[0027] In one feasible approach, the Gaofen-1 remote sensing satellite (GF-1), with its multispectral imaging capabilities, acquires raw satellite imagery of the target area. The GF-1 satellite is a crucial component of the high-resolution Earth observation system, boasting a spatial resolution of 16 meters and a revisit cycle of 1–2 days under multi-satellite networking conditions. The raw satellite imagery covers the red, green, blue, and near-infrared bands, with a single image swath width of up to 800 kilometers. A preset time node (e.g., September 24, 2024) is selected as the acquisition date for the raw satellite imagery, and GF-1 images covering the target area are downloaded as the raw satellite imagery. The preset time node is determined based on years of agricultural phenological monitoring data and verification results of agricultural machinery operation timelines. For example, the preset time node might be the time when "most summer maize in the target area has entered or is nearing maturity, the leaf color of the plants in the field changes from dark to light, and the canopy reflectance characteristics change significantly." To ensure data quality, satellite imagery with cloud cover below 10% is prioritized, and geometric correction is performed on the raw satellite imagery. Specifically, the open-source Python libraries gdal and arosics can be used for geometric correction of the raw satellite imagery. Using GDAL combined with the accompanying RPB file, a projection transformation is performed on the original satellite imagery to output a georeferenced geotiff file. Subsequently, the reference imagery is cropped according to the boundaries of the original satellite imagery to ensure spatial consistency. Finally, the coreg_local module from Arosics is used to perform sub-pixel-level registration between the original and reference images. Specifically, a sliding window and cross-correlation algorithms are employed to optimize registration accuracy, ensuring strict alignment between the original and reference images, resulting in a satellite imagery of the target area.

[0028] It should be understood that the process of acquiring satellite imagery of the target area is applicable to harvest monitoring scenarios involving multiple time phases, regions, and crops. Furthermore, this invention is not limited to the typical application of corn harvesting on September 24, 2024, but can be extended to other years, regions, and crop types. For example, remote sensing images acquired in mid-June 2024 can be used to monitor the progress of winter wheat harvesting; images acquired in early October 2024 can be used to identify the harvesting progress of spring corn or soybeans. The original satellite imagery can come from various satellite data sources with medium to high spatial resolution and high temporal resolution, including but not limited to the Gaofen series satellites (such as GF-1, GF-6, etc.), the HJ-1A / B environmental satellites, the PlanetScope constellation, and SPOT-6 / 7 satellites. The aforementioned original remote sensing images should include blue, green, red, and near-infrared bands to meet the processing requirements for calculating the Normalized Difference Vegetation Index (NDVI), cloud detection, and cluster classification.

[0029] Specifically, this involves acquiring agricultural machinery trajectory data at the same imaging time as the satellite imagery. This data is obtained from vehicle networking platforms, T-Box systems, or relevant data service interfaces, ensuring complete consistency between the agricultural machinery trajectory data and the satellite imagery's imaging time. The agricultural machinery trajectory data should include latitude and longitude coordinates, timestamps, and operational status information, providing a basis for subsequent trajectory mapping and harvesting status determination. The acquired agricultural machinery trajectory data is then cleaned and projected onto the same spatial coordinate system as the satellite imagery to ensure a unified spatial reference.

[0030] S2. Cropping and resampling the crop planting distribution data of the target area to the same spatial range and resolution as the satellite imagery, obtaining a crop planting mask map, and setting the pixel values ​​of non-crop planting areas in the satellite imagery to 0, resulting in the processed satellite imagery. Specifically: The crop planting distribution data is cropped to a spatial range completely consistent with the satellite imagery. The cropped data is then resampled to ensure its spatial resolution matches that of the satellite imagery. A binarized crop planting mask is generated based on the resampled data, and the pixel values ​​in non-crop planting areas of the satellite imagery are uniformly set to 0, thus obtaining the processed satellite imagery. Specifically: S20. The crop planting distribution data is a raster image, which consists of tens of thousands of regular, uniform grids (cells or pixels). Each grid is used to record the spatial distribution information of crops (such as corn or rice) within the target area. A cell value of 1 indicates a crop planting area. To ensure the consistency of the crop planting distribution data with the satellite image in terms of spatial range, the crop planting distribution data needs to be cropped to a spatial range that is completely consistent with the satellite image. Specifically, the open-source Python library gdal is used to extract the spatial boundary of the satellite image, and the crop planting distribution data is spatially cropped based on this spatial boundary to obtain the cropped planting distribution data. The cropped planting distribution data is completely consistent with the range of the satellite image.

[0031] S21. Extract raster parameters such as spatial resolution, starting coordinates, number of rows and columns, and projected coordinate system from the satellite imagery and use them as a template for resampling. Use Python's `rasterio.warp.reproject` method to resample the crop distribution data, obtaining resampled crop distribution data. The spatial resolution of the resampled crop distribution data is consistent with the resolution of the satellite imagery. Nearest neighbor interpolation can be used for resampling to ensure the accuracy and consistency of the classification data.

[0032] S22. Generate a binarized crop planting mask based on the resampled crop planting distribution data. Apply the crop planting mask to the satellite imagery, retaining crop planting areas with a pixel value of 1, and uniformly setting the pixel values ​​of all bands in the satellite imagery that are in non-crop planting areas (pixel values ​​not equal to 1) to 0, thereby shielding the interference from non-planting areas and obtaining the processed satellite imagery, providing clean input data for subsequent analysis.

[0033] S3. Perform cloud detection on the processed satellite imagery to obtain cloud detection results, specifically including: The normalized vegetation index (NDI) of the processed satellite imagery is calculated, and cloud detection is performed on the processed satellite imagery based on the NDI, the reflectance of the blue light band in the processed satellite imagery, and the reflectance of the near-infrared light band in the processed satellite imagery to obtain cloud detection results.

[0034] The process involves reading the reflectance information of the blue, green, red, and near-infrared bands in the processed satellite imagery and calculating the Normalized Difference Vegetation Index (NDVI) based on this information. Specifically, the NDVI of the processed satellite imagery is calculated using the following formula: in, Represents: the reflectance in the near-infrared band of the processed satellite image. Represents the reflectivity of the red band in the processed satellite image.

[0035] Clouds typically exhibit characteristics including a low Normalized Difference Vegetation Index (NDVI), high brightness, and significantly higher reflectance in the blue light band compared to other bands. Based on these spectral characteristics, rules and thresholds are set for discrimination, specifically: When the Normalized Difference Vegetation Index (NDVI) of any pixel in the processed satellite image is lower than a preset NDVI threshold, and the reflectance of that pixel in both the blue and near-infrared bands is higher than a preset reflectance threshold, that pixel can be identified as a cloud-covered pixel region, referred to as a cloud pixel. Otherwise, the pixel is identified as a non-cloud pixel, thus obtaining the cloud detection result. Specifically, the cloud detection result is a binary mask image with the same spatial range and resolution as the satellite image. In this binary mask image, each pixel is assigned a label to characterize whether it is a cloud pixel. The identified cloud regions will be masked in subsequent processing to reduce the impact of cloud interference on the analysis results.

[0036] The preset NDVI threshold can be set to 0.1, and the preset reflectivity threshold can be set to 14. The preset NDVI threshold and preset reflectivity threshold can also be adjusted according to the actual situation.

[0037] S4. Based on the cloud detection results and crop planting mask map, generate a clustering result map containing multiple categories, specifically including: S40. Based on cloud detection results and crop planting masking maps, perform masking processing on the processed satellite imagery to retain effective pixels that are not in the cloud area and are within the crop planting range. Specifically: By comprehensively utilizing cloud detection results and crop planting mask maps, the processed satellite imagery is masked to remove cloud pixels and pixels from non-crop planting areas, retaining only valid pixels that are not cloud-covered and are within the crop planting area for subsequent analysis. Specifically: The cloud detection results (where cloud pixels are identified as 0 and non-cloud pixels as 1) are logically operated and spatially overlaid with a crop planting mask image (also in binary format, where pixels in crop planting areas are identified as 1 and pixels in non-planting areas as 0). A composite mask layer is generated using logical AND operations or direct multiplication algorithms. This composite mask layer outputs a value of 1 only at positions where the corresponding pixel value in both the cloud detection results and the crop planting mask image is 1 (i.e., simultaneously satisfying the conditions of "non-cloud pixel" and "within a crop planting area"). Otherwise, it outputs a value of 0 or sets it to NoData. Subsequently, this composite mask layer is used as the final effective pixel selection template. The data is applied to the processed satellite image (where the pixel values ​​of non-crop planting areas have been set to 0). By judging each pixel individually, only the original multi-band reflectance data corresponding to pixels with a value of 1 in the composite mask layer are retained, while pixels with a value of 0 or NoData in the composite mask layer (i.e., those determined to be cloud-covered or located in non-crop planting areas) are uniformly assigned a value of 0 or NoData and removed. Finally, a satellite image containing only a subset of effective pixels that are not cloud-covered and whose spatial location is strictly within the crop planting area is output. This image completely eliminates cloud interference and the influence of non-target features, providing high-quality and highly reliable input data for subsequent feature vector construction and cluster analysis.

[0038] S41. Extract and combine the reflectance values ​​of the blue band, green band, and red band of each effective pixel into a feature vector. Cluster all feature vectors to generate a clustering result map containing multiple categories. Specifically: S410. For each effective pixel, extract the reflectance values ​​of the blue light band, green light band, and red light band respectively, and combine them to form a feature vector until the feature vector corresponding to each effective pixel is obtained. S411. The MiniBatchKMeans clustering algorithm is used to perform cluster analysis on all feature vectors to obtain the category label of each effective pixel and the cluster center of each category.

[0039] In this embodiment, the number of cluster categories is set to 10. The initial cluster centers are selected using the k-means++ strategy to ensure a reasonable distribution of all initial cluster centers and accelerate cluster convergence. To improve processing efficiency for large-scale remote sensing data, the initial sample pool size is set to 6144 (i.e., 2048 × 3). In each iteration, 2048 pixels are randomly selected from all effective pixels as a mini-batch sample to incrementally update the cluster centers. This number of samples is randomly selected in each iteration to incrementally update the cluster centers. The maximum number of iterations is set to 100, or the process terminates early when the convergence condition is met. Finally, the category label of each effective pixel and the cluster center of each category are obtained. Through this process, the model can efficiently divide all effective pixels into 10 cluster categories, each category corresponding to a spectral pattern with similar characteristics, providing a basis for subsequent harvesting state classification.

[0040] It should be noted that this step uses the MiniBatchKMeans clustering algorithm to divide all valid pixels into multiple groups and assign different category labels in an unsupervised manner. However, the specific ground objects corresponding to each category label are not yet known. The harvesting status corresponding to each category needs to be further determined through the subsequent step S5, "mapping the agricultural machinery trajectory data to the clustering result map".

[0041] The MiniBatchKMeans clustering algorithm is an efficient variant of the standard KMeans. This method first randomly selects K points from the samples as initial cluster centers. In each iteration, a small batch of samples is randomly selected, and their distances to all cluster centers are calculated. The nearest centers are assigned to these samples, and the cluster centers are incrementally updated only based on this batch of samples. This process is iterated until the maximum number of iterations is reached or convergence occurs. This approach significantly reduces the computational cost per iteration, making it suitable for high-dimensional, large-scale remote sensing data scenarios. It can achieve rapid convergence while maintaining clustering effectiveness, and is particularly suitable for efficient clustering processing of high-resolution multi-band satellite imagery.

[0042] S412. Based on the category labels obtained in S411, areas that may exhibit anomalies are identified and assigned values. The final determined category labels are then mapped back to their corresponding spatial locations in the satellite imagery, generating a complete clustering result map containing multiple categories (e.g., 10 categories). Furthermore, based on the cloud detection results, pixel areas originally identified as cloud-covered are uniformly assigned specific category labels to accurately identify cloud areas, forming a clustering result map that can be used for subsequent agricultural machinery trajectory fusion and discrimination. Specifically: By calculating the spectral feature distance of cluster centers and the standard deviation of pixels within each cluster, anomalous pixels that are more than twice the standard deviation from their respective cluster centers or located on the edge of a cluster are identified. Based on their spatial context (e.g., using majority filtering or neighboring cluster assignment rules), the category labels of these anomalous pixels are corrected or reassigned to ensure the spectral consistency and spatial continuity of the clustering results. Furthermore, a two-dimensional blank matrix with the same size as the processed satellite image and initial values ​​all set to NoData is created. Based on the row and column index positions of the valid pixels in the processed satellite image, the final category labels after anomaly detection and correction are assigned one by one. The corresponding pixel positions in the matrix are mapped back to generate a spatially complete preliminary clustering result map containing multiple categories. Then, in the preliminary clustering result map, all cloud pixels in the cloud detection result map are uniformly assigned a specific reserved category label (e.g., value 11) to clearly identify areas that cannot be effectively clustered due to cloud cover. Finally, a complete clustering result map integrating all information is output. The clustering result map not only includes multiple crop status categories based on unsupervised division of spectral features, but also clearly marks the cloud-covered areas (cloud pixels), providing a reliable and complete classification basis for subsequent spatial mapping of agricultural machinery trajectory data and harvesting status determination.

[0043] S5. Map the agricultural machinery trajectory points in the agricultural machinery trajectory data to the clustering result map, determine whether each category is a harvested or unharvested state, and generate a classification result layer; including: The agricultural machinery trajectory points in the agricultural machinery trajectory data are mapped to the clustering result map. The number of agricultural machinery trajectory points and the number of pixels in each category in the clustering result map are counted. Based on the preset judgment logic, each category is determined to be either harvested or unharvested, and a classification result layer is generated. Specifically: S50. Spatially overlay the agricultural machinery trajectory points with the clustering result map, and count the number of agricultural machinery trajectory points (i.e., the total number of agricultural machinery trajectory points falling into that category) and the number of pixels (i.e., the total number of pixels) in each category in the clustering result map. At the same time, calculate the total number of agricultural machinery trajectory points and the total number of clustered pixels in the entire target area. Among them, agricultural machinery trajectory points are the most basic units constituting agricultural machinery trajectory data, accurately recording the spatial location and status information of agricultural machinery at a corresponding moment. Agricultural machinery trajectory points include information such as latitude and longitude coordinates, timestamps, and operating status, which are divided into "operating", "driving / transferring without load" and "stopped".

[0044] S51. Based on the statistical results of S50, calculate various reference indicators, mainly including: the maximum number of agricultural machinery trajectory points in all categories (maximum number of agricultural machinery), the corresponding category label (maximum agricultural machinery category label), and the proportion of pixels in the cloud-covered area (cloud area proportion). S52. For each category, calculate its agricultural machinery point ratio (number of agricultural machinery trajectory points in this category / total number of agricultural machinery trajectory points) and pixel ratio (number of pixels in this category / total number of clustered pixels), and determine the state according to the following preset judgment logic: If the number of agricultural machinery trajectory points in a certain category exceeds the first threshold N_max, then the category is directly determined to be in a harvested state. If the percentage of pixels in a certain category is lower than the second threshold P_pixel_low and the percentage of agricultural machinery trajectory points is higher than the percentage of pixels, then it is determined to be harvested. If the absolute value of the difference between the percentage of agricultural machinery trajectory points and the percentage of pixels is lower than the third threshold Diff_close and the number of agricultural machinery trajectory points in this category exceeds the fourth threshold N_moderate, then it is determined to be harvested. If the number of agricultural machinery trajectory points in a certain category is lower than the fifth threshold N_min and the percentage of pixels is higher than the sixth threshold P_pixel_high, then it is determined to be unharvested. If the difference between the percentage of pixels in a certain category and the percentage of agricultural machinery trajectory points exceeds the seventh threshold Diff_far, then it is determined to be unharvested. For areas with a category identifier of cloud coverage (e.g., category value = 11), they are uniformly determined to be in an unharvested state.

[0045] The threshold values ​​are: first threshold N_max=80, second threshold P_pixel_low=0.1, third threshold Diff_close=0.05, fourth threshold N_moderate=10, fifth threshold N_min=5, sixth threshold P_pixel_high=0.2, and seventh threshold Diff_far=0.2. All thresholds can be set according to the actual situation.

[0046] S53. For categories determined to be in a harvested state, assign a value of 1 to the corresponding pixel; for categories determined to be in an unharvested state, assign a value of 2 to the corresponding pixel; and assign a value of 255 to the pixel corresponding to the cloud-covered area, thereby generating the final classification result layer.

[0047] The classification result layer is a type of raster data that shares the same spatial extent, resolution, and cell alignment as the original satellite imagery and clustering result map. In the classification result layer, the value of each cell no longer represents reflectance or spectral category, but rather a specific, easily understood status identifier, specifically: A value of 1 indicates that the ground area corresponding to this pixel is classified as "harvested". A value of 2 indicates that the ground area corresponding to this pixel is classified as "unharvested". A value of 255 (or other reserved values, such as 11) indicates that the area corresponding to this pixel is "cloud-covered". Since cloud cover prevents effective analysis, it is uniformly classified as "unharvested" or "unknown".

[0048] S6. Based on the classification result layer and crop planting mask map, calculate the harvested area and total planted area within the target area, and calculate the harvest progress percentage, specifically including: S60. Using the classification result layer, count the total number of pixels identified as "harvested" and, combined with the actual area of ​​the image pixels, calculate the harvested area of ​​the target region; and based on the crop planting mask map, count the total number of all non-zero pixels (i.e., crop planting areas) and convert them to obtain the total planting area. Specifically: Using the categorized result layer, the total number of pixels identified as "harvested" is counted. Combined with the spatial resolution parameters of the remote sensing image (16m × 16m), the number of pixels is converted into the corresponding ground area to calculate the harvested area of ​​the target region. Furthermore, based on the crop planting mask map, the total number of all non-zero pixels (i.e., crop planting areas) is counted and converted into the total planting area. By incorporating administrative boundaries for spatial cropping, harvested area statistics can be achieved at the provincial, municipal, and county levels.

[0049] S61. Based on the harvested area and total planted area obtained in S60, calculate the harvest progress percentage using the following formula: Harvesting progress percentage = (harvested area / total planted area) × 100% S62. Write the above statistical results into the attribute table, including fields such as: administrative division name, total planted area, harvested area, and harvest progress percentage. These results can be used to construct a thematic map of crop harvesting progress, enabling visual display and dynamic monitoring of crop harvesting status.

[0050] This invention provides a method for monitoring crop harvesting progress that integrates big data on agricultural machinery trajectories with multispectral satellite remote sensing. Compared to existing technologies, this invention constructs a multi-dimensional fusion analysis mechanism based on the spatial distribution of agricultural machinery trajectories and the clustering results of remote sensing images, significantly improving the accuracy and reliability of automatic identification of crop harvesting status. This method fully utilizes the direct evidence of harvesting activities provided by agricultural machinery data and the large-scale surface spectral information provided by remote sensing data, exhibiting stable adaptability under different crop types and regional conditions. It possesses both high-efficiency response capabilities and good potential for large-scale promotion, providing reliable near-real-time data support and intelligent decision-making basis for large-scale agricultural production management. Moreover, agricultural machinery trajectories, as a dynamic information source highly coupled with crop harvesting activities, possess significant adaptability across regions and crops. Addressing the limitations of existing remote sensing monitoring methods in terms of generalization and timeliness, this invention introduces agricultural machinery operation trajectory data as a direct criterion for characterizing actual harvesting behavior, and performs collaborative analysis and deep fusion with multi-temporal remote sensing images. By installing a T-Box (remote information processing terminal) on agricultural machinery, data such as operation paths, timestamps, and status information can be collected frequently and with high precision. This allows for the construction of a precise spatial-attribute correspondence between the agricultural machinery's trajectory and remote sensing images, supporting accurate identification of harvested areas. Through the effective integration of agricultural machinery big data and satellite remote sensing data, this invention significantly improves the accuracy, timeliness, and regional generalization ability of harvest progress monitoring, providing key technical support for the refinement and intelligentization of modern agricultural management.

[0051] Although the steps have been numbered in the above embodiments, they are only specific embodiments given by the present invention. Those skilled in the art can adjust the execution order of the steps according to the actual situation, which is also within the protection scope of the present invention. It can be understood that some embodiments may include some or all of the above embodiments.

[0052] like Figure 2 As shown, a crop harvesting progress monitoring system 200 according to an embodiment of the present invention includes an image data acquisition module 201, a data processing module 202, a cloud detection module 203, a clustering module 204, a classification result layer generation module 205, and a harvesting progress determination module 206. The image data acquisition module 201 is used to: acquire satellite images of the target area and acquire agricultural machinery trajectory data of the target area at the same imaging time as the satellite images; The data processing module 202 is used to: crop the crop planting distribution data of the target area and resample it to the same spatial range and resolution as the satellite image to obtain a crop planting mask map, and set the pixel value of the non-crop planting area in the satellite image to 0 to obtain the processed satellite image; The cloud detection module 203 is used to: perform cloud detection on the processed satellite imagery and obtain cloud detection results; Clustering module 204 is used to: generate a clustering result map containing multiple categories based on cloud detection results and crop planting mask map; The classification result layer generation module 205 is used to: map the agricultural machinery trajectory points in the agricultural machinery trajectory data to the clustering result map, determine whether each category is a harvested state or an unharvested state, and generate a classification result layer; The harvest progress determination module 206 is used to: based on the classification result layer and the crop planting mask, count the harvested area and the total planted area in the target area, and calculate the harvest progress percentage.

[0053] Optionally, in the above technical solution, the cloud detection module 203 is specifically used for: The normalized vegetation index (NDI) of the processed satellite imagery is calculated, and cloud detection is performed on the processed satellite imagery based on the NDI, the reflectance of the blue light band in the processed satellite imagery, and the reflectance of the near-infrared light band in the processed satellite imagery to obtain cloud detection results.

[0054] Optionally, in the above technical solution, the clustering module 204 is specifically used for: Based on cloud detection results and crop planting mask map, the processed satellite image is masked to retain effective pixels that are not cloud pixels and are within the crop planting area. The reflectance values ​​of the blue light band, green light band, and red light band of each effective pixel are extracted and combined into a feature vector. All feature vectors are clustered to generate a clustering result map containing multiple categories.

[0055] Optionally, in the above technical solution, the classification result layer generation module 205 is specifically used for: The agricultural machinery trajectory points in the agricultural machinery trajectory data are mapped to the clustering result map. The number of agricultural machinery trajectory points and the number of pixels in each category in the clustering result map are counted. Based on the preset judgment logic, each category is determined to be either harvested or unharvested, and a classification result layer is generated.

[0056] It should be noted that the beneficial effects of the crop harvesting progress monitoring system 200 provided in the above embodiments are the same as those of the crop harvesting progress monitoring method described above, and will not be repeated here. Furthermore, the system provided in the above embodiments is only illustrated by the division of the above functional modules. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the system can be divided into different functional modules according to the actual situation to complete all or part of the functions described above. In addition, the system and method embodiments provided in the above embodiments belong to the same concept, and their specific implementation process is detailed in the method embodiments, and will not be repeated here.

[0057] The crop harvesting progress monitoring system of the present invention can be a computer program (including program code) running on a computer device. For example, the crop harvesting progress monitoring system of the present invention is an application software that can be used to execute the corresponding steps in the crop harvesting progress monitoring method of the present invention.

[0058] In some embodiments, the crop harvesting progress monitoring system of the present invention can be implemented in a combination of hardware and software. As an example, the crop harvesting progress monitoring system of the present invention can be a processor in the form of a hardware decoding processor, which is programmed to execute the crop harvesting progress monitoring method of the present invention. For example, the processor in the form of a hardware decoding processor can be one or more application-specific integrated circuits (ASICs), DSPs, programmable logic devices (PLDs), complex programmable logic devices (CPLDs), field-programmable gate arrays (FPGAs), or other electronic components.

[0059] The modules described in the embodiments of this invention can be implemented in software or hardware. The names of the modules are not, in some cases, limiting the scope of the module itself.

[0060] An electronic device according to an embodiment of the present invention includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements any of the above-mentioned crop harvesting progress monitoring methods. That is, an electronic device according to an embodiment of the present invention may include, but is not limited to: a processor and a memory; the memory is used to store the computer program; the processor is used to execute the crop harvesting progress monitoring method shown in any embodiment of the present invention by calling the computer program.

[0061] In one alternative embodiment, an electronic device is provided, such as Figure 3 As shown, Figure 3 The illustrated electronic device 4000 includes a processor 4001 and a memory 4003. The processor 4001 and the memory 4003 are connected, for example, via a bus 4002. Optionally, the electronic device 4000 may further include a transceiver 4004, which can be used for data interaction between the electronic device and other electronic devices, such as sending and / or receiving data. It should be noted that in practical applications, the transceiver 4004 is not limited to one type, and the structure of the electronic device 4000 does not constitute a limitation on the embodiments of the present invention.

[0062] Processor 4001 may be a CPU (Central Processing Unit), a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It can implement or execute the various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of this invention. Processor 4001 may also be a combination that implements computational functions, such as including one or more microprocessor combinations, a combination of a DSP and a microprocessor, etc.

[0063] Bus 4002 may include a path for transmitting information between the aforementioned components. Bus 4002 may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus, etc. Bus 4002 can be divided into address bus, data bus, control bus, etc. For ease of representation, Figure 3 The bus 4002 is represented by only one thick line, but this does not mean that there is only one bus or one type of bus.

[0064] The memory 4003 may be ROM (Read Only Memory) or other types of static storage devices capable of storing static information and instructions, RAM (Random Access Memory) or other types of dynamic storage devices capable of storing information and instructions, or EEPROM (Electrically Erasable Programmable Read Only Memory), CD-ROM (Compact Disc Read Only Memory) or other optical disc storage, optical disc storage (including compressed optical discs, laser discs, optical discs, digital universal optical discs, Blu-ray discs, etc.), magnetic disk storage media or other magnetic storage devices, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer, but not limited thereto.

[0065] The memory 4003 stores application code (computer program) for executing the present invention, and its execution is controlled by the processor 4001. The processor 4001 executes the application code stored in the memory 4003 to implement the content shown in the foregoing method embodiments.

[0066] Among them, electronic devices can also be terminal devices, which can be any device that can install applications, including at least one of smartphones, tablets, laptops, desktop computers, smart speakers, smartwatches, smart TVs, and smart in-vehicle devices.

[0067] It should be noted that, Figure 3 The electronic device shown is merely an example and should not be construed as limiting the functionality and scope of use of the embodiments of the present invention.

[0068] An embodiment of the present invention provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements any of the above-described crop harvesting progress monitoring methods.

[0069] Alternatively, the computer-readable storage medium may be a read-only memory (ROM), a random access memory (RAM), a compact disc read-only memory (CD-ROM), magnetic tape, a floppy disk, and an optical data storage device, etc.

[0070] In an exemplary embodiment, a computer program product or computer program is also provided, which includes computer instructions stored in a computer-readable storage medium. A processor of an electronic device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the electronic device to perform any of the above-described crop harvesting progress monitoring methods.

[0071] Computer program code for performing the operations of this invention can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, and conventional procedural programming languages ​​such as C or similar languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0072] It should be understood that the flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of methods and computer program products according to various embodiments of the present invention. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0073] The computer-readable storage medium provided in this invention can be, but is not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EEPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this invention, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0074] The aforementioned computer-readable storage medium carries one or more programs, which, when executed by the electronic device, cause the electronic device to perform the method shown in the above embodiments.

[0075] The above description is merely a preferred embodiment of the present invention and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of disclosure in this invention is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the above-disclosed concept. For example, technical solutions formed by substituting the above features with (but not limited to) technical features with similar functions disclosed in this invention.

[0076] It should be noted that the terms "first," "second," etc., used in the specification and claims of this application are used to distinguish similar objects and represent a limitation on a specific order or sequence. Where appropriate, the order of use for similar objects can be interchanged so that the embodiments of this application described herein can be implemented in an order other than that shown or described.

[0077] Those skilled in the art will recognize that this invention can be implemented as a system, method, or computer program product. Therefore, this invention can be specifically implemented in the following forms: it can be entirely hardware, entirely software (including firmware, resident software, microcode, etc.), or a combination of hardware and software, generally referred to herein as a "circuit," "module," or "system." Furthermore, in some embodiments, this invention can also be implemented as a computer program product contained in one or more computer-readable media, which includes computer-readable program code.

[0078] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention.

Claims

1. A method for monitoring crop harvesting progress, characterized in that, include: Acquire satellite imagery of the target area, and acquire agricultural machinery trajectory data of the target area at the same imaging time as the satellite imagery; The crop planting distribution data of the target area is cropped and resampled to the same spatial range and resolution as the satellite image to obtain a crop planting mask map. The pixel values ​​of non-crop planting areas in the satellite image are set to 0 to obtain the processed satellite image. Cloud detection is performed on the processed satellite imagery to obtain cloud detection results; Based on the cloud detection results and the crop planting mask map, a clustering result map containing multiple categories is generated; The agricultural machinery trajectory points in the agricultural machinery trajectory data are mapped to the clustering result map, each category is determined to be either harvested or unharvested, and a classification result layer is generated. Based on the classification result layer and the crop planting mask map, the harvested area and the total planted area within the target area are statistically analyzed, and the harvest progress percentage is calculated.

2. The method for monitoring crop harvesting progress according to claim 1, characterized in that, Cloud detection is performed on the processed satellite imagery to obtain cloud detection results, including: The normalized vegetation index (NDI) of the processed satellite image is calculated, and cloud detection is performed on the processed satellite image based on the NDI, the reflectance of the blue light band in the processed satellite image, and the reflectance of the near-infrared light band in the processed satellite image to obtain cloud detection results.

3. A method for monitoring crop harvesting progress according to claim 1 or 2, characterized in that, Based on the cloud detection results and the crop planting mask map, a clustering result map containing multiple categories is generated, including: Based on the cloud detection results and the crop planting mask map, the processed satellite image is masked to retain non-cloud pixels and effective pixels within the crop planting area. The reflectance values ​​of the blue light band, green light band, and red light band of each effective pixel are extracted and combined into a feature vector. All feature vectors are clustered to generate a clustering result map containing multiple categories.

4. A method for monitoring crop harvesting progress according to claim 1 or 2, characterized in that, Mapping the agricultural machinery trajectory points in the agricultural machinery trajectory data to the clustering result map, determining whether each category is a harvested or unharvested state, and generating a classification result layer, including: The agricultural machinery trajectory points in the agricultural machinery trajectory data are mapped to the clustering result map. The number of agricultural machinery trajectory points and the number of pixels in each category in the clustering result map are counted. Based on the preset judgment logic, each category is determined to be either harvested or unharvested, and the classification result layer is generated.

5. A crop harvesting progress monitoring system, characterized in that, It includes an image data acquisition module, a data processing module, a cloud detection module, a clustering module, a classification result layer generation module, and a harvest progress determination module; The image data acquisition module is used to: acquire satellite images of the target area, and acquire agricultural machinery trajectory data of the target area that are acquired at the same time as the satellite images; The data processing module is used to: crop and resample the crop planting distribution data of the target area to the same spatial range and resolution as the satellite image to obtain a crop planting mask map, and set the pixel value of the non-crop planting area in the satellite image to 0 to obtain the processed satellite image; The cloud detection module is used to: perform cloud detection on the processed satellite imagery and obtain cloud detection results; The clustering module is used to: generate a clustering result map containing multiple categories based on the cloud detection results and the crop planting mask map; The classification result layer generation module is used to: map the agricultural machinery trajectory points in the agricultural machinery trajectory data to the clustering result map, determine whether each category is a harvested state or an unharvested state, and generate a classification result layer; The harvest progress determination module is used to: based on the classification result layer and the crop planting mask, count the harvested area and the total planted area within the target area, and calculate the harvest progress percentage.

6. The crop harvesting progress monitoring system according to claim 5, characterized in that, The cloud detection module is specifically used for: The normalized vegetation index (NDI) of the processed satellite image is calculated, and cloud detection is performed on the processed satellite image based on the NDI, the reflectance of the blue light band in the processed satellite image, and the reflectance of the near-infrared light band in the processed satellite image to obtain cloud detection results.

7. A crop harvesting progress monitoring system according to claim 5 or 6, characterized in that, The clustering module is specifically used for: Based on the cloud detection results and the crop planting mask map, the processed satellite image is masked to retain non-cloud pixels and effective pixels within the crop planting area. The reflectance values ​​of the blue light band, green light band, and red light band of each effective pixel are extracted and combined into a feature vector. All feature vectors are clustered to generate a clustering result map containing multiple categories.

8. A crop harvesting progress monitoring system according to claim 5 or 6, characterized in that, The classification result layer generation module is specifically used for: The agricultural machinery trajectory points in the agricultural machinery trajectory data are mapped to the clustering result map. The number of agricultural machinery trajectory points and the number of pixels in each category in the clustering result map are counted. Based on the preset judgment logic, each category is determined to be either harvested or unharvested, and the classification result layer is generated.

9. An electronic device, characterized in that, The method includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement a crop harvesting progress monitoring method according to any one of claims 1 to 4.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements a crop harvesting progress monitoring method according to any one of claims 1 to 4.

Images