Parcel level crop classification method and device based on time sequence SAR image, and medium

Abstract

The invention discloses a time sequence SAR image-based plot-level crop classification method and device and a medium, and relates to the technical field of remote sensing. The method comprises the following steps: acquiring land parcel data; constructing a multi-level knowledge reasoning framework, and speculating crop types by utilizing attribute information and external knowledge of the land parcels to obtain reasoning confidence corresponding to the crop types planted in each land parcel; based on vector data of boundaries of the land parcels, time sequence features of each land parcel are extracted from the time sequence SAR images, the time sequence features of each land parcel are associated with crop types, and a standard crop time sequence feature library is established; and comparing the to-be-identified land parcel with the standard crop time sequence feature library, calculating the observation confidence coefficient, fusing the observation confidence coefficient with the reasoning confidence coefficient to obtain a fusion confidence coefficient, and identifying the crop type of the to-be-identified land parcel by using the fusion confidence coefficient. The method provided by the invention solves the problem that the traditional method is low in recognition precision under small samples and complex scenes.

Description

Technical Field

[0001] The present invention relates to the field of remote sensing technology, and more specifically, to a method, device and medium for classifying plot-level crops based on time-series SAR images. Background Art

[0002] At present, precision agriculture and smart agriculture have become important directions for the transformation of modern agriculture, gradually promoting the development of agriculture towards intelligence and intensification. Among them, crop species identification is an important application in precision agriculture, which plays an important role in achieving scientific agricultural management and ensuring fair agricultural subsidies.

[0003] Traditional crop identification requires regular field inspections by professionals, which has obvious limitations: first, it requires a lot of human and material resources and is costly; second, the results of such inspections are delayed and cannot be effectively monitored in real time; third, due to limited manpower, the scope of the inspection is often limited, making it difficult to achieve large-scale crop classification. In recent years, with the development of remote sensing technology, the resolution of remote sensing images has been continuously improved, while the cost of acquiring images has been continuously reduced, which has led to the recognition method based on remote sensing images gradually replacing traditional human inspections.

[0004] Early crop identification technology based on remote sensing images mainly used the threshold segmentation method, that is, to distinguish different crop types by artificially setting the feature threshold. This type of method has the following shortcomings: (1) It is highly dependent on expert experience and is highly subjective. (2) It has poor transferability and limited applicability between different regions and different crop types. (3) It is not adaptable enough to complex planting environments and is difficult to handle complex situations. In recent years, with the development of artificial intelligence, deep learning technology has been widely used in crop identification tasks. Many related technologies use algorithms to map crop types at the pixel level in remote sensing images.

[0005] However, crop identification methods based on deep learning still face the following challenges: First, crop growth is a dynamic process, and different crops have different phenological characteristics, requiring the collection of remote sensing images of multiple phases for analysis. However, existing multi-temporal remote sensing images often have low spatial resolution (usually 10-30 meters), which can lead to unclear boundary pixel attribution when mapping crop species at the pixel level, affecting the accuracy of subsequent downstream tasks. Second, when dealing with complex scenes, existing methods are prone to the unreasonable result of identifying multiple crops in one plot. Third, deep learning methods are highly dependent on a large number of high-quality training samples, but existing samples are scarce and unevenly distributed.

[0006] In addition, the existing technology still has problems such as poor applicability to cloudy and rainy areas and low accuracy in identifying small plots of planting. Summary of the invention

[0007] In order to solve the above technical problems, the present invention provides a plot-level crop classification method, device and medium based on time-series SAR images, which is based on multi-source data and knowledge-driven strategy, and aims to solve the problem of low recognition accuracy of traditional methods in small samples and complex scenes.

[0008] In a first aspect, the present invention provides a plot-level crop classification method based on time-series SAR images, the method comprising: Acquire land parcel data, wherein the land parcel data includes vector data of the boundary of the land parcel and attribute information of the land parcel; Construct a multi-level knowledge reasoning framework, use the attribute information of the plot and external knowledge to infer the crop type, and obtain the reasoning confidence corresponding to the crop type planted on each plot; Acquire time-series SAR images, extract time-series features of each plot from the time-series SAR images based on vector data of the boundaries of the plots, associate the time-series features of each plot with crop types, and establish a standard crop time-series feature library; The plot to be identified is compared with a standard crop time series feature library, the observation confidence is calculated, the reasoning confidence of the plot to be identified is calculated based on the multi-level knowledge reasoning framework, the observation confidence and the reasoning confidence are fused to obtain a fused confidence, and the crop type of the plot to be identified is identified using the fused confidence.

[0009] Furthermore, a multi-level knowledge reasoning framework is constructed to use the attribute information of the plot and external knowledge to infer the crop type, and obtain the reasoning confidence corresponding to the crop type planted on each plot, including: Integrate multi-source data and establish a crop phenology information database; wherein the multi-source data includes geographic information and spatial information of the plot, the geographic information includes land use type, plot altitude and / or slope, and the spatial information includes administrative divisions and / or natural divisions; the crop phenology information database is used to record key phenological moments of growth of different crops, and generate a phenological calendar in combination with climate data of natural areas; Based on expert knowledge, construct a set of reasoning conditions; Based on the input property information of the plot and external knowledge, the crop type is inferred according to the reasoning condition set, and the reasoning confidence corresponding to the crop type planted in the plot is output.

[0010] Furthermore, the process of inferring the crop type using the inference condition set includes: Determine a crop set planted on the plot according to the land use type of the plot; wherein the crop set includes crop types and probabilities of multiple crops; According to the altitude of the plot and the historical planting information of the administrative division to which it belongs, the crop types and probabilities of the multiple crops are screened to obtain a target crop set.

[0011] Further, a time series SAR image is obtained, and based on the vector data of the boundary of the plot, the time series feature of each plot is extracted from the time series SAR image, and the time series feature of each plot is associated with the crop type to establish a standard crop time series feature library, including: Perform radiometric calibration, terrain correction and speckle noise filtering on time-series SAR images to obtain preprocessed data; Based on the vector data of the boundary of the plot, the image features of each plot in the preprocessed data are extracted, and the image features are normalized to obtain the time series features of each plot; According to the actual planting situation of the plot, ensure that the plot covers the target crop type and target planting pattern, construct a time series characteristic curve based on the time series characteristics of each plot automatically extracted from the time series SAR image, and store the time series characteristic curve in the standard library to obtain a standard crop time series characteristic library.

[0012] Furthermore, the time-series SAR images are subjected to radiation calibration, terrain correction and speckle noise filtering to obtain preprocessed data, including: The calculation formula for radiometric calibration of time-series SAR images is: ; In the formula, is the backscattering coefficient, DN is the original digital value, A and B are the calibration parameters of the sensor; Speckle noise filtering is performed using the following formula: ; In the formula, is the pixel value after filtering, is the local mean of pixels, is the filter coefficient, is the pixel value at the corresponding position.

[0013] Furthermore, the image features are normalized by the following formula: ; In the formula, is the normalized image feature, The original characteristics of the plot. is the mean value of the plot image feature, is the standard deviation. The normalized eigenvalues ​​conform to the standard normal distribution.

[0014] Furthermore, the plot to be identified is compared with the standard crop time series feature library to calculate the observation confidence, including: Calculate the similarity between the plot to be identified and the standard crop time series feature library. The calculation formula is: ; In the formula, is the similarity between the plot and the features in the standard library, and Respectively represent the crop time series characteristics of the plot to be identified and the standard library in the first The characteristic value of the phase, is the length of the feature sequence, and the similarity range is , the closer the value is to 1, the more similar the two are.

[0015] The similarity is normalized by the following formula: ; In the formula, is a value in similarity, is the normalized value, and Represent the minimum and maximum values ​​of similarity respectively; The normalized value is used as the observation confidence.

[0016] Furthermore, the observation confidence and the reasoning confidence are fused to obtain a fused confidence, and the crop type of the to-be-identified plot is identified using the fused confidence, including: The observation confidence and the inference confidence are fused by the following formula to obtain the fusion confidence: ; In the formula, is the fusion confidence, is the inference confidence, is the observation confidence, is the inference weight, is the observation weight; A determination threshold is preset, and when the fusion confidence is greater than the determination threshold, the crop type corresponding to the fusion confidence is used as the crop type of the land parcel to be identified.

[0017] In a second aspect, the present invention provides a plot-level crop classification device based on time-series SAR images, the device comprising: A data acquisition unit is configured to acquire land parcel data, wherein the land parcel data includes vector data of a boundary of the land parcel and attribute information of the land parcel; The knowledge reasoning unit is configured to construct a multi-level knowledge reasoning framework, use the attribute information of the plot and external knowledge to infer the crop type, and obtain the reasoning confidence corresponding to the crop type planted in each plot; The feature library construction unit is configured to obtain time-series SAR images, extract the time-series features of each plot from the time-series SAR images based on the vector data of the boundaries of the plots, associate the time-series features of each plot with the crop type, and establish a standard crop time-series feature library; The crop classification unit is configured to compare the plot to be identified with a standard crop time series feature library, calculate the observation confidence, calculate the reasoning confidence of the plot to be identified based on the multi-level knowledge reasoning framework, fuse the observation confidence with the reasoning confidence to obtain a fused confidence, and use the fused confidence to identify the crop type of the plot to be identified.

[0018] In a third aspect, the present invention provides a readable storage medium, wherein the readable storage medium stores one or more programs, and the one or more programs can be executed by one or more processors to implement the method as described above.

[0019] The present invention has at least the following beneficial effects: 1. Improve recognition accuracy. The present invention effectively solves the recognition error problem caused by blurred boundaries in traditional methods. Experiments show that the overall recognition accuracy of this method exceeds 90%, which is more than 8% higher than that of traditional methods. Especially in small plots, the recognition accuracy is improved significantly, exceeding 15%.

[0020] 2. Reduce sample dependence. By introducing knowledge reasoning, the present invention significantly reduces the dependence on a large amount of labeled data. In practical applications, test results show that when the sample size is reduced by 50%, the recognition accuracy can still be maintained above 85%. This feature makes this method more applicable in areas where samples are scarce.

[0021] 3. Enhance the interpretability of inference results. The present invention constructs an interpretable decision-making framework, which enables crop identification results to have a clear logical basis through knowledge reasoning and observation result fusion. This feature not only improves the credibility of the results, but also provides decision-making support for the agricultural sector, which helps promote the implementation of precision agriculture. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 The overall flow chart of the plot-level crop classification method based on time-series SAR images according to an embodiment of the present invention is shown; Figure 2 Shows a geographical and elevational zoning map of a target area according to an embodiment of the present invention; Figure 3A local example diagram of a fine plot of a target area according to an embodiment of the present invention is shown; Figure 4 A partial example diagram of rice distribution recognition results in a target area according to an embodiment of the present invention is shown; Figure 5 A structural diagram of a plot-level crop classification device based on time-series SAR images according to an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0023] In order to enable those skilled in the art to better understand the technical solution of the present invention, the present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. The embodiments of the present invention are further described in detail below in conjunction with the accompanying drawings and specific embodiments, but are not intended to limit the present invention. For the various steps described herein, if there is no necessity for a causal relationship between each other, the order in which they are described as examples herein should not be regarded as a limitation, and those skilled in the art should know that they can be adjusted in order, as long as the logic between them is not destroyed, resulting in the inability to implement the entire process.

[0024] The embodiment of the present invention provides a plot-level crop classification method based on time-series SAR images, such as Figure 1 As shown, it is a flow chart of a plot-level crop classification method based on time-series SAR images. The plot-level crop classification method based on time-series SAR images includes the following steps S10 to S40.

[0025] S10: Acquire land parcel data, where the land parcel data includes vector data of the land parcel boundary and property information of the land parcel.

[0026] In this embodiment, the existing plot boundary data is used in combination with high-resolution images (spatial resolution better than 1 meter) to optimize and verify the plot boundaries to ensure the accuracy of the plot boundaries and provide accurate spatial constraints for subsequent plot crop identification.

[0027] S20: Construct a multi-level knowledge reasoning framework, use the attribute information of the plot and external knowledge to infer the crop type, and obtain the reasoning confidence corresponding to the crop type planted in each plot.

[0028] In this embodiment, the purpose of step S20 is to make a preliminary inference of the crop type based on knowledge reasoning. By building a multi-level knowledge reasoning framework, the crop type is inferred by comprehensively utilizing the plot attribute information and external knowledge. Through multi-dimensional knowledge fusion, the reasoning confidence corresponding to the crop planted in each plot is obtained, providing prior knowledge for subsequent identification.

[0029] In some embodiments, step S20 is implemented by the following steps S21 to S23.

[0030] S21: Integrate multi-source data and establish a crop phenology information database; wherein the multi-source data includes geographic information and spatial information of the plot, the geographic information includes land use type, plot altitude and / or slope, and the spatial information includes administrative divisions and / or natural divisions; the crop phenology information database is used to record key phenological moments of growth of different crops, and generate a phenological calendar in combination with climate data of natural areas.

[0031] Specifically, in step S21, first, integrate multi-source data, including geographical information such as land use type, land elevation, slope, and spatial information such as administrative divisions and natural divisions; for example, land use types include paddy fields, dry land, orchards, etc., and the altitude is layered at intervals of 100 meters. Secondly, establish a crop phenological information database to record the key phenological moments of growth of different crops, such as sowing period, heading period, maturity period, etc., and combine the specific climate data of different natural areas to generate an accurate and personalized phenological calendar. The phenological calendar is used to determine the crops planted on the plot each month.

[0032] S22: Construct a set of reasoning conditions based on expert knowledge.

[0033] In some embodiments, the process of inferring crop types based on the inference condition set includes: determining a crop set planted on the plot of land based on the land use type of the plot; wherein the crop set includes crop types and probabilities of multiple crops; and screening the crop types and probabilities of the multiple crops based on the altitude of the plot of land and the historical planting information of the administrative division to which it belongs, to obtain a target crop set.

[0034] For example, for the inference condition set, if the land type of the plot is an orchard, the possibility of planting aquatic crops such as rice is excluded; if the altitude of the plot is higher than 1,000 meters, the possibility of planting rice is reduced, such as less than 10%; if the plot is located in an administrative district, and the administrative district has historically mainly planted wheat, then the plot type is preferentially inferred to be wheat, such as 80%.

[0035] S23: Based on the input property information of the plot and external knowledge, the crop type is inferred according to the reasoning condition set, and the reasoning confidence corresponding to the crop type planted in the plot is output.

[0036] The purpose of step S23 is to achieve preliminary speculation on the crop type. As an example, the plot attributes and external knowledge are input into the knowledge reasoning model to obtain the confidence of the crop type of each plot. At the same time, a confidence threshold is set. If the confidence of the crop is higher than a certain threshold, it is regarded as a candidate crop. If it is lower than a certain threshold, its possibility is excluded. Otherwise, it needs to be further analyzed and judged in combination with the time series characteristics.

[0037] S30: Acquire time-series SAR images, extract the time-series features of each plot from the time-series SAR images based on the vector data of the boundaries of the plots, associate the time-series features of each plot with the crop type, and establish a standard crop time-series feature library.

[0038] In this embodiment, the purpose of step S30 is to establish a standard crop time series feature library by extracting features from time series SAR images, and to perform regional statistical analysis based on precise plot boundaries using multi-temporal SAR image sequences.

[0039] In some embodiments, step S30 may be implemented by the following steps S31 to S33 .

[0040] S31: Perform radiation calibration, terrain correction and speckle noise filtering on the time series SAR image to obtain preprocessed data.

[0041] In some embodiments, the calculation formula for performing radiation calibration on the time-series SAR image is: ; In the formula, is the backscattering coefficient, DN is the original digital value, A and B are the calibration parameters of the sensor; Speckle noise filtering is performed using the following formula: ; In the formula, is the pixel value after filtering, is the local mean of pixels, is the filter coefficient, is the pixel value at the corresponding position.

[0042] S32: Based on the vector data of the boundary of the land parcel, extract the image features of each land parcel in the preprocessed data, and normalize the image features to obtain the time series features of each land parcel.

[0043] In some embodiments, based on the boundaries of the plots, the statistical value features of each plot are extracted, and the time series features are normalized to eliminate the dimension effect. The formula is as follows: ; In the formula, is the normalized image feature, The original characteristics of the plot. is the mean value of the plot image feature, is the standard deviation. The normalized eigenvalues ​​conform to the standard normal distribution.

[0044] S33: According to the actual planting conditions of the plots, ensure that the plots cover the target crop types and target planting patterns, construct a time series characteristic curve based on the time series characteristics of each plot automatically extracted from the time series SAR image, and store the time series characteristic curve in a standard library to obtain a standard crop time series characteristic library.

[0045] During the specific implementation of step S33, the actual planting conditions of the plot samples can be obtained through field investigations to ensure that the samples cover the main crop types and typical planting patterns. The statistical value characteristic curves of the plot samples can be automatically extracted through high-temporal-resolution multi-phase SAR image sequences and stored in the standard library, thereby obtaining a standard crop time series feature library.

[0046] S40: Compare the plot to be identified with the standard crop time series feature library, calculate the observation confidence, calculate the reasoning confidence of the plot to be identified based on the multi-level knowledge reasoning framework, fuse the observation confidence with the reasoning confidence to obtain the fused confidence, and use the fused confidence to identify the crop type of the plot to be identified.

[0047] In this embodiment, the purpose of step S40 is to organically combine the temporal feature similarity with the reasoning confidence to achieve crop recognition by fusion of multi-source information.

[0048] In some embodiments, step S40 may be specifically implemented by the following steps S41 to S43.

[0049] S41: Timing feature matching.

[0050] Calculate the similarity between the plot to be identified and the sequence features in the standard crop time series feature library, and generate the observation confidence based on the cosine similarity between the two. The formula is as follows: ; In the formula, is the similarity between the plot and the features in the standard library, and Represents the crop time series characteristics of the plot to be identified and the standard library in the first The characteristic value of the phase, is the length of the feature sequence, and the similarity range is , the closer to 1, the more similar the two values ​​are; The conversion relationship between similarity and confidence is set. In order to increase the distinction between plots, the similarity is normalized and mapped to the [0,1] interval using the following formula: ; In the formula, is a value in similarity, is the normalized value, and Represent the minimum and maximum values ​​of similarity respectively.

[0051] The normalized value is used as the observation confidence.

[0052] S42: Confidence fusion.

[0053] A fusion method is designed to fuse the reasoning confidence based on knowledge reasoning and the observation confidence based on image observation according to the weights. The formula is as follows: ; In the formula, is the fusion confidence, is the inference confidence, is the observation confidence, is the inference weight, is the observation weight.

[0054] For example, when the inference weight and observation weight are both set to 0.5, assuming that the confidence of planting wheat in a certain plot based on knowledge reasoning is 85%, and the observation confidence is 95%, the final confidence after fusion is 90%.

[0055] S43: Preset a determination threshold, and when the fusion confidence is greater than the determination threshold, the crop type corresponding to the fusion confidence is used as the crop type of the to-be-identified land parcel.

[0056] For example, a judgment threshold is set, such as 80%. If the fusion confidence of a crop is higher than the threshold, it is judged as the crop type. Finally, the classification result at the plot level is output, including the plot crop type, confidence and spatial distribution.

[0057] Therefore, the method provided by the embodiment of the present invention complements the advantages of high-resolution images and SAR images, achieving a balance between spatial accuracy and temporal resolution. In addition, the introduction of the knowledge reasoning mechanism not only alleviates the problem of difficult model training under small sample conditions, but also fills the shortcomings of the data-driven method, making the reasoning results more interpretable, and enhancing the credibility and practicality of the results. Through the above technical solution, the present invention successfully solves the problems of blurred boundaries, insufficient samples, lack of interpretability of recognition results, etc. existing in traditional methods, and provides reliable technical support for precision agriculture.

[0058] The feasibility and advancement of the present invention will be further illustrated below in conjunction with specific experiments and analysis.

[0059] Based on the plot-level crop classification method based on time-series SAR images mentioned in the above embodiments, it is applied to a target area, which spans 105°11'~110°11' east longitude and 28°10'~32°13' north latitude, 470 kilometers wide from east to west and 450 kilometers long from north to south. It is located in the transition zone between the Qinghai-Tibet Plateau and the middle and lower reaches of the Yangtze River Plain. The overall trend is that the east is higher than the west, and the south is higher than the north. The altitude is between 73.1 and 2723.7 meters, and most areas are within 800 meters above sea level suitable for crop growth. The target area has a tropical monsoon humid climate, with an average annual temperature of 16~18℃ and precipitation in various regions between 940~1375mm. The landforms are mainly mountains and hills, with mountains accounting for 76%. The main soil types are purple soil, yellow soil, paddy soil, and lime (rock) soil. The main food crops are rice, corn, wheat, sorghum, sweet potatoes and potatoes. The main cash crops are rapeseed, tobacco, tea and pepper. The main fruit crops are citrus and grapefruit.

[0060] Data sources: multispectral high-resolution remote sensing image data, time series SAR data, vector data. Multispectral high-resolution remote sensing image data is used to extract fine plot boundaries and verify results. Time series SAR data is the IW mode GRD level 1 product data of Sentinel-1 satellite from January to August 2022, with a total of 12 scenes, which can ensure multi-period coverage at least once a month. Auxiliary vector data are phenological zoning, natural zoning, administrative divisions and agricultural planting fine plots obtained in the early stage. The plot attributes include plot type, reference land use type, altitude, slope, slope aspect and other statistical information.

[0061] By dividing the boundaries of the four blocks of west, central, south and northeast and the crop information of the elevation range, the sowing and growth time of different crop types in each region and elevation range are corresponded respectively, and the complete crop production zoning data of the target area is obtained.

[0062] Taking the western block as an example, the phenological calendar of major crops in the target area with an altitude of 0 to 400 m is shown in Table 1.

[0063] Table 1 Phenological calendar of major crops in the target area of ​​0-400 m above sea level in the western region Summer Crops Earliest sowing period Latest sowing date Crop maturity Rice Early April Early May September corn Late March Early April August potato Mid-December of the previous year mid january July Sweet potato Early May mid-may November Sorghum Early April mid-April September Winter Crops Earliest sowing period Latest sowing date Maturity rape Early October mid-october May of the following year potato Late August mid-september January of the following year wheat Late October mid-november May of the following year Based on the method proposed in this embodiment, crop identification is performed in multiple blocks of the target area, and a crop phenology information database and a knowledge reasoning model of the study area are established. Then, the multi-phase SAR image sequence of the study area is combined to perform temporal feature matching, and the confidence of the crop type inferred by knowledge reasoning is calculated. The crop identification results applied to the target area are shown in Figure 2. Figure 3 and Figure 4 As shown, Figure 4 The confidence in is the fusion confidence. The method proposed in the present invention has a good effect on small-area fine plots.

[0064] 1,347 samples of crop collection points in the same period were obtained, and the accuracy was evaluated based on the comparison and analysis of the verification samples and the inferred results. The accuracy calculation formula is: ; From the crop estimation accuracy table, we can see that the plot-level crop classification method based on time-series SAR images can achieve accurate classification and identification of crops. The accuracy is shown in Table 2. The overall classification accuracy of the method in this chapter reaches 90.66%, and the overall classification effect is relatively ideal. The classification accuracy of wheat, rice, and citrus is above 90%, which can meet the needs of detailed inspection of plot crops.

[0065] Table 2 Crop estimation accuracy table

[0066] It should be noted that regions one to five mentioned in Table 2 are five administrative regions within the target area.

[0067] The embodiment of the present invention also provides a plot-level crop classification device based on time-series SAR images, such as Figure 5 As shown, the device comprises: The data acquisition unit 501 is configured to acquire land parcel data, wherein the land parcel data includes vector data of the boundary of the land parcel and attribute information of the land parcel; The knowledge reasoning unit 502 is configured to construct a multi-level knowledge reasoning framework, use the attribute information of the plot and external knowledge to infer the crop type, and obtain the reasoning confidence corresponding to the crop type planted in each plot; The feature library construction unit 503 is configured to obtain time-series SAR images, extract the time-series features of each plot from the time-series SAR images based on the vector data of the boundaries of the plots, associate the time-series features of each plot with the crop type, and establish a standard crop time-series feature library; The crop classification unit 504 is configured to compare the plot to be identified with the standard crop time series feature library, calculate the observation confidence, calculate the reasoning confidence of the plot to be identified based on the multi-level knowledge reasoning framework, fuse the observation confidence with the reasoning confidence to obtain a fused confidence, and use the fused confidence to identify the crop type of the plot to be identified.

[0068] In some embodiments, the knowledge reasoning unit is further configured to: Integrate multi-source data and establish a crop phenology information database; wherein the multi-source data includes geographic information and spatial information of the plot, the geographic information includes land use type, plot altitude and / or slope, and the spatial information includes administrative divisions and / or natural divisions; the crop phenology information database is used to record key phenological moments of growth of different crops, and generate a phenological calendar in combination with climate data of natural areas; Based on expert knowledge, construct a set of reasoning conditions; Based on the input property information of the plot and external knowledge, the crop type is inferred according to the reasoning condition set, and the reasoning confidence corresponding to the crop type planted in the plot is output.

[0069] In some embodiments, the process of inferring the crop type using the inference condition set includes: Determine a crop set planted on the plot according to the land use type of the plot; wherein the crop set includes crop types and probabilities of multiple crops; According to the altitude of the plot and the historical planting information of the administrative division to which it belongs, the crop types and probabilities of the multiple crops are screened to obtain a target crop set.

[0070] In some embodiments, the feature library construction unit is further configured to: Perform radiometric calibration, terrain correction and speckle noise filtering on time-series SAR images to obtain preprocessed data; Based on the vector data of the boundary of the plot, the image features of each plot in the preprocessed data are extracted, and the image features are normalized to obtain the time series features of each plot; According to the actual planting situation of the plot, ensure that the plot covers the target crop type and target planting pattern, construct a time series characteristic curve based on the time series characteristics of each plot automatically extracted from the time series SAR image, and store the time series characteristic curve in the standard library to obtain a standard crop time series characteristic library.

[0071] In some embodiments, the feature library construction unit is further configured to: The calculation formula for radiometric calibration of time-series SAR images is: ; In the formula, is the backscattering coefficient, DN is the original digital value, A and B are the calibration parameters of the sensor; Speckle noise filtering is performed using the following formula: ; In the formula, is the pixel value after filtering, is the local mean of pixels, is the filter coefficient, is the pixel value at the corresponding position.

[0072] In some embodiments, the feature library construction unit is further configured to perform normalization processing on the image features by using the following formula: ; In the formula, is the normalized image feature, The original characteristics of the plot. is the mean value of the plot image feature, is the standard deviation. The normalized eigenvalues ​​conform to the standard normal distribution.

[0073] In some embodiments, the crop classification unit is further configured to: Calculate the similarity between the plot to be identified and the standard crop time series feature library. The calculation formula is: ; In the formula, is the similarity between the plot and the features in the standard library, and Represents the crop time series characteristics of the plot to be identified and the standard library in the first The characteristic value of the phase, is the length of the feature sequence, and the similarity range is , the closer to 1, the more similar the two values ​​are; The similarity is normalized by the following formula: ; In the formula, is a value in similarity, is the normalized value, and Represent the minimum and maximum values ​​of similarity respectively; The normalized value is used as the observation confidence.

[0074] In some embodiments, the crop classification unit is further configured to: The observation confidence and the inference confidence are fused by the following formula to obtain the fusion confidence: ; In the formula, is the fusion confidence, is the inference confidence, is the observation confidence, is the inference weight, is the observation weight; A determination threshold is preset, and when the fusion confidence is greater than the determination threshold, the crop type corresponding to the fusion confidence is used as the crop type of the land parcel to be identified.

[0075] It should be noted that the structures of the various plot-level crop classification devices based on time-series SAR images described in this embodiment belong to the same technical concept as the previously described plot-level crop classification method based on time-series SAR images, and achieve the same beneficial effects through the same principles, which will not be repeated here.

[0076] An embodiment of the present invention further provides a readable storage medium, which stores one or more programs. The one or more programs can be executed by one or more processors to implement the method described in any of the above embodiments.

[0077] The above description is intended to be illustrative rather than restrictive. For example, the above examples (or one or more of them) may be used in combination with each other. For example, a person of ordinary skill in the art may use other embodiments when reading the above description. In addition, in the above-mentioned specific embodiments, various features may be grouped together to simplify the present invention. This should not be interpreted as an intention that a feature of an invention that is not claimed for protection is necessary for any claim. On the contrary, the subject matter of the present invention may be less than all the features of an embodiment of a particular invention. Thus, the following claims are incorporated herein as examples or embodiments in the specific embodiments, wherein each claim is independently a separate embodiment, and it is considered that these embodiments may be combined with each other in various combinations or arrangements. The scope of the present invention should be determined with reference to the appended claims and the full scope of equivalent forms granted by these claims.

Claims

1. A plot-level crop classification method based on time-series SAR images, characterized in that: The method comprises: Acquire land parcel data, wherein the land parcel data includes vector data of the boundary of the land parcel and attribute information of the land parcel; Construct a multi-level knowledge reasoning framework, use the attribute information of the plot and external knowledge to infer the crop type, and obtain the reasoning confidence corresponding to the crop type planted on each plot; Acquire time-series SAR images, extract time-series features of each plot from the time-series SAR images based on vector data of the boundaries of the plots, associate the time-series features of each plot with crop types, and establish a standard crop time-series feature library; The plot to be identified is compared with a standard crop time series feature library, the observation confidence is calculated, the reasoning confidence of the plot to be identified is calculated based on the multi-level knowledge reasoning framework, the observation confidence and the reasoning confidence are fused to obtain a fused confidence, and the crop type of the plot to be identified is identified using the fused confidence.

2. The plot-level crop classification method based on time-series SAR images according to claim 1 is characterized in that: A multi-level knowledge reasoning framework is constructed to use the attribute information of the plot and external knowledge to infer the crop type, and obtain the reasoning confidence corresponding to the crop type planted on each plot, including: Integrate multi-source data and establish a crop phenology information database; wherein the multi-source data includes geographic information and spatial information of the plot, the geographic information includes land use type, plot altitude and / or slope, and the spatial information includes administrative divisions and / or natural divisions; the crop phenology information database is used to record key phenological moments of growth of different crops, and generate a phenological calendar in combination with climate data of natural areas; Based on expert knowledge, construct a set of reasoning conditions; Based on the input property information of the plot and external knowledge, the crop type is inferred according to the reasoning condition set, and the reasoning confidence corresponding to the crop type planted in the plot is output.

3. The plot-level crop classification method based on time-series SAR images according to claim 2 is characterized in that: The process of inferring the crop type according to the inference condition set includes: Determine a crop set planted on the plot according to the land use type of the plot; wherein the crop set includes crop types and probabilities of multiple crops; According to the altitude of the plot and the historical planting information of the administrative division to which it belongs, the crop types and probabilities of the multiple crops are screened to obtain a target crop set.

4. The method for crop classification at the plot level based on time-series SAR images according to claim 1, characterized in that: Acquire time-series SAR images, extract the time-series features of each plot from the time-series SAR images based on the vector data of the boundaries of the plots, associate the time-series features of each plot with the crop type, and establish a standard crop time-series feature library, including: Perform radiometric calibration, terrain correction and speckle noise filtering on time-series SAR images to obtain preprocessed data; Based on the vector data of the boundary of the plot, the image features of each plot in the preprocessed data are extracted, and the image features are normalized to obtain the time series features of each plot; According to the actual planting situation of the plot, ensure that the plot covers the target crop type and target planting pattern, construct a time series characteristic curve based on the time series characteristics of each plot automatically extracted from the time series SAR image, and store the time series characteristic curve in the standard library to obtain a standard crop time series characteristic library.

5. The method for crop classification at the plot level based on time-series SAR images according to claim 4, characterized in that: Perform radiometric calibration, terrain correction and speckle noise filtering on the time series SAR images to obtain preprocessed data, including: The calculation formula for radiometric calibration of time-series SAR images is: ； In the formula, is the backscattering coefficient, DN is the original digital value, A and B are the calibration parameters of the sensor; Speckle noise filtering is performed using the following formula: ； In the formula, is the pixel value after filtering, is the local mean of pixels, is the filter coefficient, is the pixel value at the corresponding position.

6. The method for crop classification at the plot level based on time-series SAR images according to claim 4, characterized in that: The image features are normalized by the following formula: ； In the formula, is the normalized image feature, The original characteristics of the plot. is the mean value of the plot image feature, is the standard deviation.

7. The plot-level crop classification method based on time-series SAR images according to claim 1 is characterized in that: Compare the plot to be identified with the standard crop time series feature library and calculate the observation confidence, including: Calculate the similarity between the plot to be identified and the standard crop time series feature library. The calculation formula is: ； In the formula, is the similarity between the plot and the features in the standard library, and Respectively represent the crop time series characteristics of the plot to be identified and the standard library in the first The characteristic value of the phase, is the length of the feature sequence, and the similarity range is , the closer the value is to 1, the more similar the two are; The similarity is normalized by the following formula: ； In the formula, is a value in similarity, is the normalized value, and Respectively represent the minimum and maximum values ​​in the similarity; The normalized value is used as the observation confidence.

8. The plot-level crop classification method based on time-series SAR images according to claim 1 is characterized in that: The observation confidence and the reasoning confidence are integrated to obtain a fused confidence, and the crop type of the to-be-identified plot is identified by using the fused confidence, including: The observation confidence and the inference confidence are fused by the following formula to obtain the fusion confidence: ； In the formula, is the fusion confidence, is the inference confidence, is the observation confidence, is the inference weight, is the observation weight; A determination threshold is preset, and when the fusion confidence is greater than the determination threshold, the crop type corresponding to the fusion confidence is used as the crop type of the land parcel to be identified.

9. A plot-level crop classification device based on time-series SAR images, characterized in that: The device comprises: A data acquisition unit is configured to acquire land parcel data, wherein the land parcel data includes vector data of a boundary of the land parcel and attribute information of the land parcel; The knowledge reasoning unit is configured to construct a multi-level knowledge reasoning framework, use the attribute information of the plot and external knowledge to infer the crop type, and obtain the reasoning confidence corresponding to the crop type planted in each plot; The feature library construction unit is configured to obtain time-series SAR images, extract the time-series features of each plot from the time-series SAR images based on the vector data of the boundaries of the plots, associate the time-series features of each plot with the crop type, and establish a standard crop time-series feature library; The crop classification unit is configured to compare the plot to be identified with a standard crop time series feature library, calculate the observation confidence, calculate the reasoning confidence of the plot to be identified based on the multi-level knowledge reasoning framework, fuse the observation confidence with the reasoning confidence to obtain a fused confidence, and use the fused confidence to identify the crop type of the plot to be identified. 10 . A non-transitory computer-readable storage medium storing instructions, which, when executed by a processor, perform the method according to claim 1 .

Images