Grain crop harvesting progress monitoring method, device, equipment and medium

Abstract

The invention discloses a grain crop harvesting progress monitoring method, device, equipment and medium, and relates to the technical field of agricultural remote sensing monitoring, and the method comprises the steps: obtaining grain crop type distribution diagram data and multispectral remote sensing image data of a preset number of days before harvesting, calculating a red band depth index and a surface water index, the reference reflectivity of the field parcels is not harvested; during harvesting, acquiring multispectral remote sensing image data to be monitored at the current moment, and performing difference operation on the red light and short wave infrared band reflectivity of each pixel to be monitored and the reference reflectivity of the corresponding unharvested field to obtain a harvesting index; fitting a parameter intersection point through a double-cluster Gaussian mixture model based on the harvest index to obtain an optimal threshold value; and finally, judging a harvesting state according to the harvesting index and an optimal threshold value to obtain a monitoring result. According to the invention, large-range and near-real-time grain crop harvesting progress monitoring is realized, and the monitoring efficiency and accuracy are improved.

Description

Technical Field

[0001] The present application relates to the field of agricultural remote sensing monitoring technology, and in particular to a method, device, equipment and medium for monitoring the progress of grain crop harvesting. Background Art

[0002] Harvesting, the final stage of grain crop cultivation, is a critical step in achieving yield from the crop's total biomass. In the era of agricultural mechanization, large-scale, near-real-time monitoring of harvest progress is crucial. Keeping up-to-date on harvest progress allows for dynamic optimization of harvesting equipment scheduling, effectively addressing the impact of extreme weather conditions like heavy rain and hail on harvesting operations, and ensuring smooth harvesting.

[0003] Currently, large-scale crop harvest progress monitoring relies primarily on field surveys and uploading of agricultural machinery operation information. While this method provides relatively accurate information, it has numerous drawbacks. First, field surveys require significant manpower, material resources, and time, resulting in low efficiency. Second, uploading agricultural machinery operation information may result in incomplete data collection, failing to accurately and timely reflect the overall harvest progress. The development of satellite-based remote sensing technology offers another viable approach for large-scale grain crop harvest progress monitoring. Grain crops, such as rice, corn, and wheat, are typically planted in continuous fields. During harvest, aboveground biomass is removed, and the relatively moist photosynthetically active canopy transforms into dry stubble or bare soil after harvest, triggering significant spectral changes that provide critical signals for remote sensing to capture harvest events. However, no remote sensing-based algorithms exist for grain crop harvest progress monitoring.

[0004] Therefore, there is an urgent need for a remote sensing-based grain crop harvest progress monitoring method to solve the problems existing in the monitoring methods of related technologies and realize large-scale, near-real-time harvest progress monitoring. Summary of the Invention

[0005] The purpose of this application is to provide a method, device, equipment and medium for monitoring the harvest progress of grain crops, which can solve the problems of existing large-scale crop harvest progress monitoring relying on field surveys and uploading of agricultural machinery operation information, which is time-consuming and labor-intensive, and incomplete data collection. It can realize large-scale, near-real-time monitoring of the harvest progress of grain crops, improve monitoring efficiency and accuracy, provide strong support for dynamic optimization of harvesting equipment scheduling and response to extreme weather impacts, and give full play to the advantages of remote sensing technology in the field of agricultural monitoring.

[0006] To achieve the above objectives, this application provides the following solutions: In a first aspect, the present application provides a method for monitoring the progress of grain crop harvest, comprising: Obtaining grain crop type distribution map data and multispectral remote sensing image data for a preset number of days before harvest; Based on the distribution map data of grain crops and multispectral remote sensing image data for a preset number of days before harvest, the reference reflectance of unharvested fields of each grain crop type is determined by calculating the red band depth index and the surface moisture index; the grain crop types include corn, rice, and wheat; During the harvest period, obtain the multispectral remote sensing image data to be monitored at the current moment; The harvest index of each grain crop type at the current moment is obtained by performing difference calculations on the red light band reflectance and shortwave infrared band reflectance of each monitored pixel in the multispectral remote sensing image data to be monitored at the current moment and the reference reflectance of the unharvested field of the corresponding grain crop type. Based on the harvest index of each type of grain crop at the current moment, the intersection of two Gaussian distribution parameters is fitted using a two-cluster Gaussian mixture model to obtain the optimal threshold of the harvest index of each type of grain crop at the current moment; Based on the harvest index of each type of grain crop at the current moment and the corresponding optimal threshold of the harvest index, the harvest status of the grain crop corresponding to each monitored pixel at the current moment is judged to obtain the grain crop harvest progress monitoring result.

[0007] In a second aspect, the present application provides a device for monitoring the progress of grain crop harvesting, comprising: The raw data acquisition module is used to obtain the grain crop type distribution map data and multispectral remote sensing image data for a preset number of days before harvest; A reference reflectance determination module is configured to determine the reference reflectance of unharvested fields of each type of grain crop by calculating a red band depth index and a surface moisture index based on grain crop type distribution map data and multispectral remote sensing image data for a preset number of days before harvest; the grain crop types include corn, rice, and wheat; The module for acquiring remote sensing image data to be monitored is used to acquire the multispectral remote sensing image data to be monitored at the current moment during the harvest period; The harvest index calculation module is used to obtain the harvest index of each type of grain crop at the current moment by performing a difference operation between the red light band reflectance and the shortwave infrared band reflectance of each pixel to be monitored in the multispectral remote sensing image data to be monitored at the current moment and the reference reflectance of the unharvested field of the corresponding grain crop type; An optimal threshold determination module is used to obtain the optimal threshold of the harvest index of each type of grain crop at the current moment by fitting the intersection of two Gaussian distribution parameters through a double-cluster Gaussian mixture model based on the harvest index of each type of grain crop at the current moment; The harvest progress monitoring module is used to determine the harvest status of each grain crop corresponding to the monitored pixel at the current moment based on the harvest index of each grain crop type at the current moment and the corresponding optimal threshold of the harvest index, and obtain the grain crop harvest progress monitoring results.

[0008] In a third aspect, the present application provides a computer device comprising: 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 the steps of the method for monitoring the progress of grain crop harvesting as described above.

[0009] In a fourth aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method for monitoring the progress of grain crop harvesting as described above.

[0010] According to the specific embodiments provided in this application, this application has the following technical effects: The present application provides a method, device, equipment and medium for monitoring the progress of grain crop harvest. By obtaining grain crop type distribution map data and multispectral remote sensing image data for a preset number of days before harvest, the method solves the problems of traditional monitoring methods such as single data source, difficulty in obtaining data, and difficulty in covering large areas, thereby achieving efficient and convenient acquisition of large-scale monitoring data. By calculating the red band depth index and surface moisture index based on the above data, the reference reflectivity of unharvested fields of each grain crop type is determined, and the harvest index is obtained by calculating the difference between the pixel reflectivity and the reference reflectivity in the data to be monitored at the current moment, thereby solving the problem that the crop harvest feature information in the remote sensing data is not prominent and difficult to extract accurately, thereby achieving effective quantitative characterization of crop harvest status related features. By determining the optimal threshold value by fitting the parameter intersection of a two-cluster Gaussian mixture model based on the harvest index, and judging the harvest status of each grain crop corresponding to the pixel to be monitored based on this threshold value, the method solves the problem of fuzzy harvest status judgment criteria and insufficient accuracy, thereby achieving accurate and reliable grain crop harvest progress monitoring. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0012] Figure 1 This is a diagram of the application environment of a method for monitoring the progress of grain crop harvesting in one embodiment of the present application.

[0013] Figure 2A flowchart of a method for monitoring the progress of grain crop harvesting provided in one embodiment of the present application.

[0014] Figure 3 A schematic diagram of the main process of a method for monitoring the progress of grain crop harvesting provided in another embodiment of the present application.

[0015] Figure 4 A schematic diagram of the spectrum comparison between harvested and unharvested fields provided in one embodiment of the present application; wherein, Figure 4 (a) Schematic diagram showing the distribution of harvested and unharvested rice fields in the rice planting area; Figure 4 (b) indicates Figure 4 (a) Comparison of spectral reflectance of different types of fields in the rice planting area; Figure 4 (c) Schematic diagram showing the distribution of harvested and unharvested fields in the corn planting area; Figure 4 (d) indicates Figure 4 (c) Comparison of spectral reflectance of different types of fields in the corn planting area; Figure 4 (e) Schematic diagram showing the distribution of harvested and unharvested fields in the wheat planting area; Figure 4 (f) indicates Figure 4 (e) Comparison of spectral reflectance of different types of fields in the wheat planting area.

[0016] Figure 5 A schematic diagram of a method for determining the optimal threshold value of the harvest index provided in one embodiment of the present application.

[0017] Figure 6 A schematic diagram of the research area provided in one embodiment of the present application.

[0018] Figure 7 for Figure 6 Spatial distribution of harvest dates in the study area.

[0019] Figure 8 A schematic diagram of the functional modules of a device for monitoring the progress of grain crop harvesting provided in one embodiment of the present application.

[0020] Figure 9 A schematic diagram of the structure of a computer device provided in one embodiment of the present application. DETAILED DESCRIPTION

[0021] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0022] In order to make the above-mentioned purposes, features and advantages of the present application more obvious and easy to understand, the present application is further described in detail below with reference to the accompanying drawings and specific implementation methods.

[0023] The method for monitoring the progress of grain crop harvest provided in the embodiment of the present application can be applied to Figure 1 In the application environment shown. Among them, the terminal 101 communicates with the server 102 through the network. The data storage system can store the data that the server 102 needs to process. The data storage system can be set up separately, integrated on the server 102, or placed on the cloud or other servers. The terminal 101 can send the food crop type distribution map data and the multispectral remote sensing image data of the preset number of days before harvest to the server 102. After the server 102 receives the food crop type distribution map data and the multispectral remote sensing image data of the preset number of days before harvest, the server 102 determines the reference reflectance of the unharvested field of each food crop type based on the food crop type distribution map data and the multispectral remote sensing image data of the preset number of days before harvest by calculating the red band depth index and the surface moisture index; the food crop types include corn, rice, and wheat; during the harvest period, the multispectral remote sensing image data to be monitored at the current moment is obtained; by The red light band reflectivity and shortwave infrared band reflectivity of each monitored pixel in the spectral remote sensing image data are respectively subtracted from the reference reflectivity of the unharvested field of the corresponding grain crop type to obtain the harvest index of each grain crop type at the current moment; based on the harvest index of each grain crop type at the current moment, the intersection of two Gaussian distribution parameters is fitted using a double-cluster Gaussian mixture model to obtain the optimal threshold value of the harvest index of each grain crop type at the current moment; based on the harvest index of each grain crop type at the current moment and the corresponding optimal threshold value of the harvest index, the harvest status of the grain crop corresponding to each monitored pixel at the current moment is determined to obtain the grain crop harvest progress monitoring result. The server 102 can feed back the obtained grain crop harvest progress monitoring result to the terminal 101. In addition, in some embodiments, the method for monitoring the progress of grain crop harvesting can also be implemented independently by the server 102 or the terminal 101. For example, the terminal 101 can directly perform harvesting progress monitoring and processing on the grain crop type distribution map data and multispectral remote sensing image data for a preset number of days before harvest, or the server 102 can obtain the grain crop type distribution map data and multispectral remote sensing image data for a preset number of days before harvest from the data storage system, and perform harvesting progress monitoring and processing on the grain crop type distribution map data and multispectral remote sensing image data for a preset number of days before harvest.

[0024] The terminal 101 may be, but is not limited to, various desktop computers, laptop computers, smart phones, tablet computers, IoT devices, and portable wearable devices. The server 102 may be implemented as an independent server or a server cluster consisting of multiple servers, or a cloud server.

[0025] In an exemplary embodiment, Figure 2 As shown, a method for monitoring the progress of grain crop harvest is provided. The method is executed by a computer device, specifically a computer device such as a terminal or a server, or a terminal and a server. In the embodiment of the present application, the method is applied to Figure 1 Taking the server 102 in FIG. 1 as an example, the method includes the following steps 201 to 206. In which: Step 201: Acquire grain crop type distribution map data and multispectral remote sensing image data for a preset number of days before harvest.

[0026] Step 202, based on the grain crop type distribution map data and multispectral remote sensing image data of a preset number of days before harvest, the reference reflectance of the unharvested field of each grain crop type is determined by calculating the red band depth index and the surface moisture index; the grain crop types include corn, rice, and wheat.

[0027] Step 203: During the harvest period, obtain the multispectral remote sensing image data to be monitored at the current moment.

[0028] Step 204 is to obtain the harvest index of each type of grain crop at the current moment by performing difference calculations on the red light band reflectivity and the shortwave infrared band reflectivity of each pixel to be monitored in the multispectral remote sensing image data to be monitored at the current moment and the reference reflectivity of the unharvested field of the corresponding grain crop type.

[0029] Step 205 , based on the harvest index of each type of food crop at the current moment, the intersection of two Gaussian distribution parameters is fitted using a two-cluster Gaussian mixture model to obtain the optimal threshold value of the harvest index of each type of food crop at the current moment.

[0030] Step 206 , based on the harvest index of each type of grain crop at the current moment and the corresponding optimal threshold value of the harvest index, the harvest status of each grain crop corresponding to the monitored pixel at the current moment is determined to obtain a grain crop harvest progress monitoring result.

[0031] By implementing the above steps 201 to 206, the present application can make full use of the computing power of computer equipment to efficiently process food crop type distribution map data and multispectral remote sensing image data. By accurately calculating the red band depth index and the surface moisture index to determine the reference reflectance, and using the difference operation to obtain the harvest index, it is possible to accurately capture the changes in the spectral characteristics of the crop harvest status. On this basis, the class balance buffer is determined and the dual-cluster Gaussian mixture model is used to obtain the optimal threshold, and then the harvest status is judged, realizing large-scale, near-real-time monitoring of the food crop harvest progress, effectively solving the problems of time-consuming and labor-intensive traditional monitoring methods and incomplete data collection, improving the accuracy and efficiency of monitoring, and providing strong support for scientific decision-making in agricultural production.

[0032] In another exemplary embodiment of the present application, step 202 specifically includes: The red band depth index and surface moisture index are calculated using the following formulas: .

[0033] .

[0034] .

[0035] in, Indicates the red band depth index; 、 、 and Respectively represent the reflectivity of the red light band, the green light band, the near infrared band and the short-wave infrared band; represents the interpolated reflectance of the red light band; 、 、 and represent the central wavelength of the near-infrared band, the central wavelength of the green light band, and the central wavelength of the red light band respectively; Represents the surface moisture index.

[0036] Based on the distribution map data of grain crop types at a preset number of days before harvest, pixels whose red band depth index and surface moisture index were greater than the preset unharvested threshold in each grain crop type were selected as reference unharvested pixels.

[0037] The average red light band reflectance and shortwave infrared band reflectance of all reference unharvested pixels corresponding to each food crop type were calculated respectively to obtain the average red light band reflectance and average shortwave infrared band reflectance corresponding to each food crop type.

[0038] The average red light band reflectance and the average shortwave infrared band reflectance corresponding to each food crop type are respectively used as the red light band reference reflectance and the shortwave infrared band reference reflectance of the corresponding food crop type.

[0039] In another exemplary embodiment of the present application, the harvest index is calculated in step 204 using the following formula: .

[0040] in, Harvest Index; and Represent the reflectivity of red light band and short-wave infrared band respectively; and They represent the reference reflectivity of the red light band and the shortwave infrared band, respectively.

[0041] In another exemplary embodiment of the present application, step 205 specifically includes: Based on the harvest index of each type of food crop at the current moment, a class balance buffer zone of the harvest status of each type of food crop at the current moment is determined; the harvest status includes a harvested state and an unharvested state.

[0042] Based on the class-balanced buffer of each type of food crop at the current moment, the intersection of two Gaussian distribution parameters is fitted by a double-cluster Gaussian mixture model to obtain the optimal threshold of the harvest index of each type of food crop at the current moment.

[0043] In another exemplary embodiment of the present application, based on the harvest index of each type of food crop at the current moment, determining the class balance buffer zone of the harvest state of each type of food crop at the current moment specifically includes: For each crop type, the following steps are performed to obtain the class-balanced buffer zone for each crop type's harvest status at the current moment: The pixels to be monitored whose harvest index of the current grain crop type is greater than a preset high value threshold are extracted from the multispectral remote sensing image data to be monitored at the current moment as potential harvested pixels of the current grain crop type.

[0044] The spatial clustering algorithm is used to cluster the potential harvested pixels of the current grain crop type, divide the potential harvested pixels into different clusters, and obtain the spatial cluster center of each cluster.

[0045] Taking each cluster center as an anchor point, square buffers are gradually expanded outward at preset area intervals until the preset maximum area is reached, thus obtaining multiple buffers of different areas.

[0046] In each buffer zone, the harvest index distribution was fitted using a two-cluster Gaussian mixture model and a single-cluster Gaussian model, and the likelihood ratio of the two-cluster Gaussian mixture model to the single-cluster Gaussian model was calculated.

[0047] The likelihood ratios of all buffers are sorted from large to small, and the buffers whose likelihood ratios are within the preset percentage threshold are selected as the class-balanced buffers for the current grain crop type at the current moment.

[0048] In another exemplary embodiment of the present application, a two-cluster Gaussian mixture model and a single-cluster Gaussian model are used to fit the harvest index distribution in each buffer zone, and the likelihood ratio of the two-cluster Gaussian mixture model to the single-cluster Gaussian model is calculated, specifically including: The likelihood functions of the single-cluster Gaussian mixture model and the two-cluster Gaussian mixture model are calculated using the following formulas: .

[0049] .

[0050] in, and They represent the likelihood function values ​​of the single-cluster Gaussian model and the two-cluster Gaussian mixture model respectively; Indicates the first i Pixels; N represents the total number of pixels in the class balance buffer; represents the probability density function of the normal distribution; Harvest Index; represents the mean parameter of the single cluster Gaussian model; represents the variance parameter of the single cluster Gaussian model; represents the weight parameter of unharvested pixels; represents the mean parameter of unharvested pixels; represents the variance parameter of unharvested pixels; represents the weight parameter of harvested pixels; represents the mean parameter of harvested pixels; Represents the variance parameter of the harvested pixels.

[0051] For each buffer zone, the likelihood ratio of the two-cluster Gaussian mixture model to the single-cluster Gaussian model is calculated using the following formula: .

[0052] in, represents the likelihood ratio.

[0053] In another exemplary embodiment of the present application, based on the class balance buffer of each type of food crop at the current moment, the intersection of two Gaussian distribution parameters is fitted by a two-cluster Gaussian mixture model to obtain the optimal threshold value of the harvest index of each type of food crop at the current moment, specifically including: The harvest index of all pixels to be monitored within the class balance buffer zone of each type of food crop is calculated respectively.

[0054] For each type of grain crop, the expectation-maximization algorithm was used to fit the harvest index of all monitored pixels in the class-balanced buffer zone through a two-cluster Gaussian mixture model to obtain the Gaussian distribution parameters of the harvested cluster and the unharvested cluster at the current moment; the Gaussian distribution parameters included the mean, variance, and weight of each distribution.

[0055] For each type of grain crop, the probability density function of the harvested cluster and the probability density function of the unharvested cluster at the current moment are constructed according to the Gaussian distribution parameters of the harvested cluster and the unharvested cluster at the current moment.

[0056] For each type of food crop, the harvest index value corresponding to the intersection of the probability density function of the harvested cluster and the probability density function of the unharvested cluster at the current moment is used as the optimal threshold of the food crop type, and the optimal threshold of the harvest index of each food crop type at the current moment is obtained.

[0057] The following describes the application using a specific grain crop harvest progress monitoring process as an example.

[0058] The monitoring data for the current season of this application requires two input data; the monitoring data includes crop type distribution map data and multispectral remote sensing image data; specifically, it can be a crop type distribution map of the current season and a Sentinel-2 multispectral image acquired in near real time.

[0059] This method mainly includes the following steps, wherein the main process (step 1-step 2) of a method for monitoring the progress of grain crop harvest is shown in the following figure: Figure 3 shown.

[0060] Step 1: Obtain the seasonal grain crop distribution map and Sentinel-2 multispectral imagery.

[0061] Specifically, obtain a distribution map of seasonal grain crop types through mid-season remote sensing crop classification techniques or field surveys. Near-real-time Sentinel-2 multispectral remote sensing data should be downloaded from the European Space Agency (ESA) website (https: / / browser.dataspace.copernicus.eu / ) just before the harvest season begins. The remote sensing images should have a revisit period (ideally within 10 days, as otherwise progress monitoring will be difficult) and a 1.6µm shortwave infrared band to characterize moisture absorption.

[0062] Step 2: Based on the crop types and multispectral imagery from Step 1, determine the reference reflectance of unharvested fields for each food crop type.

[0063] Specifically, the Red Band Depth (RBD) and Land Surface Water Index (Land Surface Water Index) are calculated for all pixels in the image, as shown in the following formulas: .

[0064] .

[0065] .

[0066] in, Indicates the red band depth index; 、 、 and Respectively represent the reflectance of the red light band, green light band, near-infrared band, and short-wave infrared band (1.6µm) (corresponding to bands 4, 3, 8, and 11 of the Sentinel-2 sensor, respectively); Represents the interpolated reflectance of the red light band, that is, the reflectance obtained by linearly interpolating the reflectance of the green light and near-infrared bands according to wavelength. Since the fields have higher photosynthetic biomass and water content when they are not harvested, the RBD and LSWI values ​​are also correspondingly higher. Based on the crop type distribution map, the pixels with RBD and LSWI indices higher than the top 5% for each crop type are selected as representative reference unharvested pixels, and the average red light and shortwave infrared reflectance of these reference unharvested pixels are calculated as the reference red light reflectance and reference shortwave infrared reflectance for each crop type ( , ); 、 、 and represent the center wavelength of the near-infrared band (833nm), the center wavelength of the green light band (560nm), and the center wavelength of the red light band (665nm); Represents the surface moisture index. This is the reflectance obtained by linearly interpolating the reflectance of the green and near-infrared bands according to wavelength.

[0067] Step 3: Calculate the harvest index based on the reference reflectance of the unharvested fields of each grain crop type in Step 2 and the grain crop type and multispectral imagery in Step 1.

[0068] Specifically, by calculating the pixel to be monitored in the red light band (Sentinel 2 Band 4, ) and shortwave infrared bands (Sentinel-2 Band 11, ) and the reference reflectance of the unharvested field of the corresponding grain crop type obtained in step 2 ( , ) to construct the Harvest Index ( HI ), as shown in the following formula: .

[0069] Since the remote sensing pixels after harvest show obvious spectral feature differences from those of unharvested pixels in the red band (Red) and shortwave infrared band (SWIR1), a schematic diagram of the spectrum comparison between harvested and unharvested fields was drawn, as shown in the figure below. Figure 4 shown. Figure 4 (a) Schematic diagram of the distribution of harvested and unharvested fields in the rice-growing area, with different colors used to identify the specific locations and ranges of unharvested fields, harvested fields (stubble), and harvested fields (bare soil). Figure 4 (b) is Figure 4 (a) Comparison of spectral reflectance of different types of rice fields in the rice-growing area, clearly showing how the spectral reflectance of each type of field varies with wavelength in the Red and SWIR1 bands. Figure 4 (c) shows a schematic diagram of the distribution of harvested and unharvested fields in the corn-growing area, and also uses colors to distinguish fields with different harvest status. Figure 4 (d) Figure 4 (c) The comparison of spectral reflectance of different types of fields in the corn-growing area helps to analyze the differences in spectral characteristics of corn fields at different harvest stages. Figure 4 (e) is a schematic diagram of the distribution of harvested and unharvested fields in the wheat planting area, which clarifies the harvest status distribution of wheat fields. Figure 4 (f) is Figure 4 The spectral reflectance comparison chart for different types of wheat fields in a wheat-growing area (e) reflects the spectral reflectance characteristics of wheat fields under different harvest conditions. In actual analysis, the larger the harvest index (HI) of a pixel, the more likely it is that the pixel has been harvested. By combining these spectral comparison charts with the HI index, a more accurate assessment of the harvest status of a field can be achieved.

[0070] Step 4: Based on the harvest index in step 3, automatically focus on the harvested / unharvested class balancing buffer.

[0071] During the crop harvest process, the proportion of harvested and unharvested pixels in the study area cannot always remain balanced, but is in dynamic change. For example, in the early stage of harvest, the proportion of unharvested pixels is large, while in the late stage of harvest, harvested pixels dominate, resulting in an extremely unbalanced proportion of the two types. At this time, the harvest index histogram often presents a unimodal distribution, which increases the difficulty of threshold recognition. To solve this problem, a method is designed to automatically focus on a buffer containing a balanced number of harvested / unharvested pixels (with a clear bimodal distribution in the buffer) to obtain the optimal threshold, and this threshold is applied to the entire area. The schematic diagram of the method for determining the optimal threshold of the harvest index is shown in the figure. Figure 5 Specifically, pixels with harvest indices above the top 5% quantile within the entire remote sensing image (referred to as "long-tail data") were first extracted as potentially harvested pixels. Second, the DBSCAN (Density-Based Spatial Clustering of Applications with Noise) clustering algorithm was applied based on the spatial coordinates of these pixels to obtain the spatial cluster centers of potentially harvested pixels. Finally, square buffers (with areas gradually expanding from 0.5 km² to 4.5 km²) were generated using the spatial cluster centers as anchor points. Based on the HI index data within each buffer, a two-cluster Gaussian mixture model (GMM) and a single-cluster Gaussian model were used to fit their distributions.

[0072] The single cluster Gaussian model uses a single normal distribution to fit the HI index data distribution in the buffer zone.

[0073] .

[0074] Represents the harvest index under a single cluster Gaussian mixture model The probability density function of is used to describe the distribution characteristics of the harvest index in the buffer zone. is the standard notation for the probability density function of the normal distribution; represents the mean parameter of the single cluster Gaussian model; represents the variance parameter of the single cluster Gaussian model; the normal distribution The mean of ) and variance ( ) is the maximum likelihood estimate of the HI mean and variance of all pixels in the buffer zone. Represents the harvest index under a single cluster Gaussian mixture model The probability density function is used to describe the distribution characteristics of the harvest index in the buffer zone. Therefore, the likelihood function of the single cluster Gaussian mixture model is for: .

[0075] in, Indicates the first i pixels; N is the total number of pixels in the class balance buffer.

[0076] The two-cluster GMM uses two normal distributions to fit the HI index data distribution in the buffer: .

[0077] in, represents the weight parameter of unharvested pixels; represents the mean parameter of unharvested pixels; represents the variance parameter of unharvested pixels; represents the weight parameter of harvested pixels; represents the mean parameter of harvested pixels; Represents the variance parameter of the harvested pixels. and ( ) represent the weight parameters (i.e., pixel proportions) of class 0 (unharvested) and class 1 (harvested), respectively; and Represent the mean of HI of class 0 and class 1 pixels respectively; and Represent the variance of HI of class 0 and class 1 pixels respectively. The parameters of the two-cluster Gaussian mixture model ( ). It is obtained by the Expectation Maximization (EM) algorithm. The specific steps are as follows: 1) Initialization parameters: , and Take the lower 25% and upper 25% quantile values ​​of HI in the buffer zone respectively, and The variance of HI in the buffer zone is taken, and the iteration step t=0 is set.

[0078] 2) Step E: Calculate each pixel i The posterior probability of belonging to class 0 and class 1 ( and ).

[0079] .

[0080] 3) Step M: Based on and Update parameters: Weight parameter .

[0081] Mean parameter .

[0082] Variance parameter .

[0083] 4) Convergence judgment: Repeat steps 2)-3) until the estimated parameter changes of two iterations are less than a certain threshold ( ),Right now and and and and and When , the iteration stops.

[0084] Based on the estimated distribution parameters, the likelihood function of the two-cluster Gaussian mixture model can be calculated .

[0085] .

[0086] For each buffer, the likelihood ratio (LLR) of the fitted two-cluster GMM and one-cluster Gaussian distribution is calculated: .

[0087] Buffers with LLR values ​​greater than the top 5% percentile of all LLR values ​​were selected as quasi-balanced buffers, indicating that the distribution of harvest index HI in these buffers was more consistent with the bimodal distribution characteristics.

[0088] Step 5: Based on the class-balanced buffer in step 4, the optimal threshold of the harvest index is determined using a two-cluster Gaussian mixture model.

[0089] Specifically, all pixels within the class-balanced buffer obtained in step 4 are summarized, and the parameters of the two Gaussian distributions are determined by bi-cluster GMM fitting (the fitting algorithm is consistent with the EM estimation algorithm in step 4), and the HI value corresponding to the intersection of the two Gaussian distributions is used as the optimal classification threshold of the harvest index ( ). To satisfy the following equation: .

[0090] The solution is: .

[0091] in, represents the intermediate calculation amount, .

[0092] Step 6: Based on the optimal classification threshold of the harvest index in step 5, the harvested / unharvested pixels are judged to monitor the progress of grain crop harvest.

[0093] Specifically, the harvest index of all pixels to be monitored in the image and the optimal classification threshold obtained in step 5 are compared. In contrast, if the harvest index is higher than the optimal classification threshold ( ) is determined as a harvested pixel, otherwise it is determined as an unharvested pixel, thereby realizing near real-time monitoring of the progress of grain crop harvest.

[0094] like Figure 6 The figure shows a schematic diagram of the study area, where winter wheat is the primary crop, typically harvested in early to mid-June. Because the region's cropping structure is relatively stable, a 2020 winter wheat mapping result was used as the crop distribution map input. Four Sentinel-2 images from June 1, 6, 11, and 16, 2022, were acquired as remote sensing data input to monitor the progress of the 2022 winter wheat harvest. Furthermore, a field survey was conducted during the 2022 winter wheat harvest, obtaining harvest dates for 57 plots to validate the proposed method.

[0095] Based on the proposed method, winter wheat harvest recognition was performed on four Sentinel-2 images on June 1, June 6, June 11, and June 16, 2022, and the harvest date of each pixel was obtained. Figure 7 As shown, the proposed harvest period gradually shifts from south to north, consistent with the harvesting patterns in the region. Based on validation points from field surveys, the classification accuracy of harvested / unharvested categories on images from different dates is shown in Table 1 below. This demonstrates that the proposed method achieves an accuracy of over 90% for harvested pixels on images from different dates (Note: By June 16, nearly all harvests had occurred in the study area, and the field survey points did not include unharvested samples. Therefore, the F1 and average F1 values ​​for the unharvested category are not valid), demonstrating the validity of the proposed method.

[0096] Table 1 Classification accuracy of harvested / unharvested types on images of different dates

[0097] The present application also provides an application scenario that applies the above-mentioned method for monitoring the progress of grain crop harvesting. Specifically, the method for monitoring the progress of grain crop harvesting provided in this embodiment can be applied in the scenario of agricultural precision management and decision support. The scenario of agricultural precision management and decision support includes a data collection link, a data processing and analysis link, and a decision-making and execution link. The crop planting-related data enters the system from the data collection link, and after a series of data cleaning, integration, modeling and analysis, key information such as the growth status of the crops and the progress of the harvest is obtained, and then enters the decision-making and execution link. The method for monitoring the progress of grain crop harvesting provided in this embodiment belongs to the core step in the data processing and analysis link. Specifically, by obtaining the monitoring data of the current season, including crop type distribution map data and multispectral remote sensing image data, the relevant index is calculated based on these data, the reference reflectance and harvest index are determined, and then the optimal threshold of the harvest index is obtained. Finally, the harvest status of each pixel corresponding to the grain crop is judged, providing accurate and timely data support for subsequent agricultural production decisions, and facilitating the scientific and refined management of agricultural production.

[0098] Based on the same inventive concept, the embodiments of the present application also provide a device for monitoring the progress of grain crop harvests for implementing the aforementioned method for monitoring the progress of grain crop harvests. The implementation solution provided by the device is similar to the implementation solution described in the aforementioned method. Therefore, the specific limitations in the embodiments of one or more devices for monitoring the progress of grain crop harvests provided below can be found in the limitations of the method for monitoring the progress of grain crop harvests above, and will not be repeated here.

[0099] In an exemplary embodiment, Figure 8 As shown, a device for monitoring the progress of grain crop harvest is provided, comprising: The raw data acquisition module 301 is used to obtain grain crop type distribution map data and multispectral remote sensing image data for a preset number of days before harvest.

[0100] The reference reflectance determination module 302 is used to determine the reference reflectance of unharvested fields of each type of grain crop by calculating the red band depth index and the surface moisture index based on the grain crop type distribution map data and multispectral remote sensing image data of a preset number of days before harvest; the grain crop types include corn, rice, and wheat.

[0101] The module 303 for acquiring remote sensing image data to be monitored is used to acquire the multispectral remote sensing image data to be monitored at the current moment during the harvest period.

[0102] The harvest index calculation module 304 is used to obtain the harvest index of each type of grain crop at the current moment by performing a difference operation between the red light band reflectivity and the shortwave infrared band reflectivity of each pixel to be monitored in the multispectral remote sensing image data to be monitored at the current moment and the reference reflectivity of the unharvested field of the corresponding grain crop type.

[0103] The optimal threshold determination module 305 is used to obtain the optimal threshold of the harvest index of each type of food crop at the current moment by fitting the intersection of two Gaussian distribution parameters through a double-cluster Gaussian mixture model.

[0104] The harvest progress monitoring module 306 is used to determine the harvest status of each food crop corresponding to the monitored pixel at the current moment based on the harvest index of each food crop type at the current moment and the corresponding optimal threshold of the harvest index, and obtain the food crop harvest progress monitoring result.

[0105] In an exemplary embodiment, a computer device is provided. The computer device may be a server or a terminal. The internal structure diagram thereof may be as follows: Figure 9 As shown. The computer device includes a processor, a memory, an input / output interface (I / O), and a communication interface. The processor, memory, and I / O interface are connected via a system bus, and the communication interface is connected to the system bus via the I / O interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and computer program in the non-volatile storage medium. The database of the computer device is used to store grain crop harvest progress processing data. The I / O interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal via a network connection. When the computer program is executed by the processor, a method for monitoring grain crop harvest progress is implemented.

[0106] Those skilled in the art will understand that Figure 9 The structure shown in the figure is merely a block diagram of a portion of the structure related to the solution of the present application and does not constitute a limitation on the computer device to which the solution of the present application is applied. A specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement. In an exemplary embodiment, a computer device is provided, including a memory and a processor. The memory stores a computer program, and the processor implements the steps of the above-mentioned method embodiments when executing the computer program.

[0107] In an exemplary embodiment, a computer-readable storage medium is provided, storing a computer program. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.

[0108] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0109] Those skilled in the art will appreciate that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the above-mentioned embodiments. In particular, any reference to a memory, database, or other medium used in the embodiments provided in this application can include at least one of a non-volatile and a volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM may be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0110] The databases involved in the various embodiments provided herein may include at least one of a relational database and a non-relational database. Non-relational databases may include, but are not limited to, distributed databases based on blockchains. The processors involved in the various embodiments provided herein may include, but are not limited to, general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic units, data processing logic units based on quantum computing, and the like.

[0111] The technical features of the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0112] This document uses specific examples to illustrate the principles and implementation methods of this application. The description of the above examples is only intended to help understand the method and core concept of this application. At the same time, for those skilled in the art, based on the concept of this application, there may be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting this application.

Claims

1. A method for monitoring the progress of grain crop harvest, characterized in that: The method for monitoring the progress of grain crop harvesting comprises: Obtaining grain crop type distribution map data and multispectral remote sensing image data for a preset number of days before harvest; Based on the distribution map data of grain crops and multispectral remote sensing image data for a preset number of days before harvest, the reference reflectance of unharvested fields of each grain crop type is determined by calculating the red band depth index and the surface moisture index; the grain crop types include corn, rice, and wheat; During the harvest period, obtain the multispectral remote sensing image data to be monitored at the current moment; The harvest index of each grain crop type at the current moment is obtained by performing difference calculations on the red light band reflectance and shortwave infrared band reflectance of each monitored pixel in the multispectral remote sensing image data to be monitored at the current moment and the reference reflectance of the unharvested field of the corresponding grain crop type. Based on the harvest index of each type of grain crop at the current moment, the intersection of two Gaussian distribution parameters is fitted using a two-cluster Gaussian mixture model to obtain the optimal threshold of the harvest index of each type of grain crop at the current moment; Based on the harvest index of each type of grain crop at the current moment and the corresponding optimal threshold of the harvest index, the harvest status of the grain crop corresponding to each monitored pixel at the current moment is judged to obtain the grain crop harvest progress monitoring result.

2. The method for monitoring the progress of grain crop harvest according to claim 1, wherein: Based on the distribution map of grain crop types and multispectral remote sensing image data for a preset number of days before harvest, the reference reflectance of unharvested fields of each grain crop type is determined by calculating the red band depth index and surface moisture index. Specifically, the following are included: The red band depth index and surface moisture index are calculated using the following formulas: ； ； ； in, Indicates the red band depth index; 、 、 and Respectively represent the reflectivity of the red light band, the green light band, the near infrared band and the short-wave infrared band; represents the interpolated reflectance of the red light band; 、 、 and represent the central wavelength of the near-infrared band, the central wavelength of the green light band, and the central wavelength of the red light band respectively; represents the surface moisture index; Based on the distribution map data of grain crop types at a preset number of days before harvest, pixels whose red band depth index and surface moisture index were greater than the preset unharvested threshold for each grain crop type were selected as reference unharvested pixels; Calculate the average red light band reflectance and shortwave infrared band reflectance of all reference unharvested pixels corresponding to each food crop type, and obtain the average red light band reflectance and average shortwave infrared band reflectance corresponding to each food crop type; The average red light band reflectance and the average shortwave infrared band reflectance corresponding to each food crop type are respectively used as the red light band reference reflectance and the shortwave infrared band reference reflectance of the corresponding food crop type.

3. The method for monitoring the harvest progress of grain crops according to claim 1, wherein: By performing difference calculations on the red light band reflectance and shortwave infrared band reflectance of each pixel to be monitored in the multispectral remote sensing image data to be monitored at the current moment and the reference reflectance of the unharvested field of the corresponding grain crop type, the harvest index of each grain crop type at the current moment is obtained, specifically including: The harvest index was calculated using the following formula: ； in, represents the harvest index; and Represent the reflectivity of red light band and short-wave infrared band respectively; and They represent the reference reflectivity of the red light band and the shortwave infrared band, respectively.

4. The method for monitoring the progress of grain crop harvest according to claim 1, wherein: Based on the harvest index of each type of grain crop at the current moment, the intersection of two Gaussian distribution parameters is fitted by a two-cluster Gaussian mixture model to obtain the optimal threshold of the harvest index of each type of grain crop at the current moment, specifically including: Determining a class balance buffer zone for a harvest state of each type of grain crop at the current moment based on a harvest index of each type of grain crop at the current moment; the harvest state includes a harvested state and an unharvested state; Based on the class-balanced buffer of each type of food crop at the current moment, the intersection of two Gaussian distribution parameters is fitted by a double-cluster Gaussian mixture model to obtain the optimal threshold of the harvest index of each type of food crop at the current moment.

5. The method for monitoring the progress of grain crop harvest according to claim 4, wherein: Based on the harvest index of each type of grain crop at the current moment, the class balance buffer zone of the harvest status of each type of grain crop at the current moment is determined, specifically including: For each crop type, the following steps are performed to obtain the class-balanced buffer zone for each crop type's harvest status at the current moment: Extracting pixels to be monitored whose harvest index of the current grain crop type is greater than a preset high value threshold from the multispectral remote sensing image data to be monitored at the current moment as potential harvested pixels of the current grain crop type; The spatial clustering algorithm is used to cluster the potential harvested pixels of the current crop type, the potential harvested pixels are divided into different clusters, and the spatial cluster center of each cluster is obtained; Taking each cluster center as an anchor point, square buffers are gradually expanded outward at preset area intervals until the preset maximum area is reached, thus obtaining multiple buffers of different areas. In each buffer zone, the harvest index distribution was fitted using a two-cluster Gaussian mixture model and a single-cluster Gaussian model, and the likelihood ratio of the two-cluster Gaussian mixture model to the single-cluster Gaussian model was calculated. The likelihood ratios of all buffers are sorted from large to small, and the buffers whose likelihood ratios are within the preset percentage threshold are selected as the class-balanced buffers for the current grain crop type at the current moment.

6. The method for monitoring the progress of grain crop harvest according to claim 4, wherein: Based on the class-balanced buffer of each type of food crop at the current moment, the intersection of two Gaussian distribution parameters is fitted by a two-cluster Gaussian mixture model to obtain the optimal threshold of the harvest index of each type of food crop at the current moment, specifically including: Calculate the harvest index of all pixels to be monitored within the class balance buffer zone for each type of food crop; For each type of grain crop, the expectation maximization algorithm is used to fit the harvest index of all monitored pixels in the class-balanced buffer zone through a two-cluster Gaussian mixture model to obtain the Gaussian distribution parameters of the harvested cluster and the unharvested cluster at the current moment. The Gaussian distribution parameters include the mean, variance, and weight of each distribution. For each type of grain crop, the probability density function of the harvested cluster and the probability density function of the unharvested cluster at the current moment are constructed according to the Gaussian distribution parameters of the harvested cluster and the unharvested cluster at the current moment. For each type of food crop, the harvest index value corresponding to the intersection of the probability density function of the harvested cluster and the probability density function of the unharvested cluster at the current moment is used as the optimal threshold of the food crop type, and the optimal threshold of the harvest index of each food crop type at the current moment is obtained.

7. The method for monitoring the harvest progress of grain crops according to claim 5, wherein: In each buffer zone, the harvest index distribution was fitted using a two-cluster Gaussian mixture model and a single-cluster Gaussian model, and the likelihood ratio of the two-cluster Gaussian mixture model to the single-cluster Gaussian model was calculated, including: The likelihood functions of the single-cluster Gaussian mixture model and the two-cluster Gaussian mixture model are calculated using the following formulas: ； ； in, and They represent the likelihood function values ​​of the single-cluster Gaussian model and the two-cluster Gaussian mixture model respectively; Indicates the first i Pixels; N represents the total number of pixels in the class balance buffer; represents the probability density function of the normal distribution; represents the harvest index; represents the mean parameter of the single cluster Gaussian model; represents the variance parameter of the single cluster Gaussian model; represents the weight parameter of unharvested pixels; represents the mean parameter of unharvested pixels; represents the variance parameter of unharvested pixels; represents the weight parameter of harvested pixels; represents the mean parameter of harvested pixels; represents the variance parameter of the harvested pixels; For each buffer zone, the likelihood ratio of the two-cluster Gaussian mixture model to the single-cluster Gaussian model is calculated using the following formula: ； in, represents the likelihood ratio.

8. A device for monitoring the progress of grain crop harvest, characterized in that: The device for monitoring the progress of grain crop harvesting uses the method for monitoring the progress of grain crop harvesting according to any one of claims 1 to 7, and the device for monitoring the progress of grain crop harvesting comprises: The raw data acquisition module is used to obtain the grain crop type distribution map data and multispectral remote sensing image data for a preset number of days before harvest; A reference reflectance determination module is configured to determine the reference reflectance of unharvested fields of each type of grain crop by calculating a red band depth index and a surface moisture index based on grain crop type distribution map data and multispectral remote sensing image data for a preset number of days before harvest; the grain crop types include corn, rice, and wheat; The module for acquiring remote sensing image data to be monitored is used to acquire the multispectral remote sensing image data to be monitored at the current moment during the harvest period; The harvest index calculation module is used to obtain the harvest index of each type of grain crop at the current moment by performing a difference operation between the red light band reflectance and the shortwave infrared band reflectance of each pixel to be monitored in the multispectral remote sensing image data to be monitored at the current moment and the reference reflectance of the unharvested field of the corresponding grain crop type; An optimal threshold determination module is used to obtain the optimal threshold of the harvest index of each type of grain crop at the current moment by fitting the intersection of two Gaussian distribution parameters through a double-cluster Gaussian mixture model based on the harvest index of each type of grain crop at the current moment; The harvest progress monitoring module is used to determine the harvest status of each grain crop corresponding to the monitored pixel at the current moment based on the harvest index of each grain crop type at the current moment and the corresponding optimal threshold of the harvest index, and obtain the grain crop harvest progress monitoring results.

9. A computer device comprising: 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 the method for monitoring the progress of grain crop harvesting according to any one of claims 1 to 7.

10. A 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 monitoring the progress of grain crop harvesting according to any one of claims 1 to 7 is implemented.

Images