A remote sensing identification method for rice-duck integrated farming mode

Abstract

The application discloses a kind of rice duck integrated breeding mode remote sensing identification method, including obtaining the remote sensing image of target area rice heading period and obtaining its normalized vegetation index, green normalized vegetation index and near infrared band;Determine rice pixel and non-rice pixel by texture analysis;Determine the proportion of the pixel of normalized vegetation index outside its upper edge and lower edge value range and the pixel of green normalized vegetation index outside its upper edge value and lower edge value range relative to the total number of plot pixels, and identify whether the plot is a first plot;According to the proportion of the number of non-rice pixels in the plot relative to its total number of pixels, identify whether the first plot is a second plot;Determine the proportion of the number of non-rice pixels in the four corner regions of the minimum bounding rectangle of the second plot relative to the total number of non-rice pixels in the second plot, and identify whether the second plot is a rice duck integrated breeding plot according to the calculation result.The present application can efficiently and accurately identify rice duck integrated breeding plot.

Description

Technical Field

[0001] This invention relates to the field of remote sensing identification of ground features, and more particularly to a remote sensing identification method for an integrated rice-duck farming model. background

[0002] The rice-duck integrated farming model is a composite ecological planting model that combines natural ecology and human intervention, based on paddy fields, centered on rice cultivation, and characterized by free-range ducks. This model utilizes the biological characteristics of ducks to weed, control pests, and fertilize in the fields, loosening the soil and enriching the fields with muddy water, saving planting and breeding costs, improving the ecological environment, and achieving sustainable rice-duck farming. With "ecological priority and green development" as its core concept, this model has formed a new path for sustainable agricultural development where humans and nature coexist harmoniously, playing a significant role in reducing costs, increasing yields, improving quality and efficiency, and protecting the ecology. Efficient and accurate monitoring and identification of rice-duck integrated farming model areas is of great significance for analyzing the improvement of quality and efficiency, distributing subsidies, and further optimizing and promoting the rice-duck integrated farming model.

[0003] Currently, remote sensing identification methods for single crop planting patterns are relatively mature, mainly including pixel-based and object-oriented methods such as decision tree classification, maximum likelihood method, K-means method, and multi-scale segmentation. These methods utilize the spectral and textural features of crops to accurately identify single crop planting patterns. However, remote sensing identification methods for integrated rice-duck farming, a composite ecological planting pattern, are relatively scarce. Firstly, integrated rice-duck farming is a newly promoted composite ecological planting technology in recent years, and research on remote sensing identification methods in this area is limited. Secondly, the growth and development characteristics of rice in integrated rice-duck farming are similar to those in traditional planting patterns, making it difficult to identify using only spectral and textural features.

[0004] The above background information is provided only to assist in understanding the inventive concept and technical solution of this invention. It does not necessarily belong to the prior art of this patent application, nor does it necessarily provide technical teaching. In the absence of clear evidence that the above information was disclosed before the filing date of this patent application, the above background information should not be used to evaluate the novelty and inventiveness of this application. Summary of the Invention

[0005] The purpose of this invention is to provide a remote sensing identification method for rice-duck integrated farming, which can efficiently and accurately identify rice-duck integrated farming plots.

[0006] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0007] A remote sensing identification method for rice-duck integrated farming model, used for remote sensing identification of rice-duck integrated farming plots, includes the following steps:

[0008] Acquire remote sensing images of rice during the heading stage in the target area;

[0009] The acquired images of the target area are overlaid to obtain the normalized vegetation index and green normalized vegetation index corresponding to each pixel in the remote sensing image.

[0010] Near-infrared bands of remote sensing images of rice during the heading stage in the target area were acquired, and rice pixels and non-rice pixels were identified through texture analysis.

[0011] The following methods are used to identify whether each rice-growing plot within the target area is the first plot: Using box plot analysis, the upper and lower edge values ​​of the Normalized Difference Vegetation Index (NDV) and the Green Normalized Difference Vegetation Index (GDI) for a rice-growing plot are calculated. A first range is set from the lower edge to the upper edge of the NV, and a second range is set from the lower edge to the lower edge of the GDI. The proportion of pixels with NV outside the first range relative to the total number of pixels in the plot is calculated, as is the proportion of pixels with GDI outside the second range relative to the total number of pixels in the plot. Based on the calculation results, the rice-growing plot is identified as the first plot or another plot.

[0012] Identify whether each of the first plots is a second plot, and identify whether the first plot is a second plot or other plots based on the ratio of the number of non-rice pixels in the first plot to the total number of pixels in the plot;

[0013] Identifying rice-duck integrated farming plots in the second plot includes obtaining the minimum bounding rectangle of the second plot, calculating the proportion of non-rice pixels in the four vertices of the minimum bounding rectangle relative to the total number of non-rice pixels in the second plot, and identifying the second plot as a rice-duck integrated farming plot or other plots based on the calculation results.

[0014] Furthermore, based on any one or a combination of the aforementioned technical solutions, identifying whether a rice-growing plot is the first plot includes the following steps:

[0015] Count the total number of pixels N in a rice planting plot;

[0016] Count the number N pixels in rice-growing plots whose normalized vegetation index (NVI) falls outside the first range. P1 And the number N of pixels whose green normalized vegetation index is outside the second range. P2 ;

[0017] Calculate the percentage S of the number of pixels with normalized vegetation index outside the first range relative to the total number of pixels in the plot. P1And calculate the percentage S of the number of pixels outside the second range of the green normalized vegetation index relative to the total number of pixels. P2 , among which, S P1 =N P1 / N,S P2 =N P2 / N;

[0018] If S P1 Greater than or equal to the first percentage threshold and S P2 If the proportion is greater than or equal to the second threshold, then the rice-growing plot is the first plot; otherwise, the rice-growing plot is another plot.

[0019] Furthermore, based on any one or a combination of the aforementioned technical solutions, identifying whether the first land parcel is the second land parcel includes the following steps:

[0020] Determine the number of non-rice pixels within the first plot;

[0021] Calculate the proportion of non-rice pixels in the first plot relative to its total number of pixels;

[0022] If the proportion of non-rice pixels to the total number of pixels is greater than or equal to the third proportion threshold, then the first plot is the second plot; otherwise, the first plot is another plot.

[0023] Furthermore, following any one or a combination of the aforementioned technical solutions, the proportion of non-rice pixels in the four vertex regions of the minimum bounding rectangle relative to the total number of non-rice pixels in the second plot is calculated, and the second plot is identified as a rice-duck integrated farming plot or other plots based on the calculation results, including the following steps:

[0024] The minimum bounding rectangle is divided into at least five partitions, of which four partitions have the same shape and area and are located at the four vertices of the minimum bounding rectangle.

[0025] The number of non-rice pixels in the four rectangular partitions located at the vertices of the minimum bounding rectangle and the total number of non-rice pixels in the second plot are counted.

[0026] Calculate the proportion of non-rice pixels in the four rectangular partitions located at the apex of the minimum bounding rectangle relative to the total number of non-rice pixels in the second plot. If the calculation result is greater than or equal to the fourth proportion threshold, then the second plot is a non-rice-duck integrated farming plot; otherwise, the second plot is another plot.

[0027] Furthermore, following any one or a combination of the aforementioned technical solutions, the minimum bounding rectangle is divided into nine equally divided rectangular partitions, including the following partitioning steps:

[0028] Divide each side of the minimum bounding rectangle into three equal parts, with three division points corresponding to each side;

[0029] Connect the equal division points on opposite sides of the minimum bounding rectangle one by one to form nine equal rectangular partitions.

[0030] Furthermore, following any one or a combination of the aforementioned technical solutions, the first percentage threshold is 4%; and / or,

[0031] The second percentage threshold is 3%.

[0032] Furthermore, based on any or a combination of the aforementioned technical solutions, the third percentage threshold is 2%.

[0033] Furthermore, based on any one or a combination of the aforementioned technical solutions, the fourth percentage threshold is 60%.

[0034] Furthermore, based on any one or a combination of the aforementioned technical solutions, determining rice pixels and non-rice pixels includes the following steps:

[0035] Acquire near-infrared band remote sensing images of rice during the heading stage in the target area;

[0036] Texture analysis is performed based on the near-infrared band of remote sensing images. The texture analysis uses a 3*3 window to select the second moment for analysis.

[0037] A certain number of rice pixels and non-rice pixels are selected, and the threshold range for distinguishing rice pixels and non-rice pixels is determined based on the second moment values ​​of the rice pixel and non-rice pixel samples.

[0038] The threshold range is used to distinguish between rice pixels and non-rice pixels. If the second moment value of a pixel is within the threshold range, it is a rice pixel; if the second moment value of a pixel is not within the threshold range, it is a non-rice pixel.

[0039] Furthermore, following any one or a combination of the aforementioned technical solutions, the minimum bounding rectangle of the second plot is obtained using element management tools.

[0040] The beneficial effects of the technical solution provided by this invention are as follows:

[0041] a. This invention identifies whether a rice-growing plot is a first plot based on the growth and development characteristics of rice and the management characteristics of ducks under the rice-duck integrated farming model, using the normalized vegetation index and green normalized vegetation index of remote sensing images. It then identifies whether the first plot is a second plot based on the proportion of non-rice pixels in the rice-duck integrated farming plot. Furthermore, it identifies whether the second plot is a rice-duck integrated farming plot based on the feature of setting up a grazing channel in one corner of the plot. This achieves accurate identification of rice-duck integrated farming model areas, and the identification results are accurate and reliable.

[0042] b. The remote sensing identification method for the rice-duck integrated farming model proposed in this invention has a simple process and adopts a hierarchical identification strategy, which makes the identification efficiency high. Attached Figure Description

[0043] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0044] Figure 1 A flowchart illustrating a remote sensing identification method for an integrated rice-duck farming model, provided as an exemplary embodiment of the present invention;

[0045] Figure 2 A drone image of rice in Miaotou Village during the heading stage, provided as an exemplary embodiment of the present invention;

[0046] Figure 3 A schematic diagram illustrating the identification results of the first plot of land in Miaotou Village, as provided in an exemplary embodiment of the present invention;

[0047] Figure 4 A schematic diagram of the texture analysis results of rice heading stage in Miaotou Village provided as an exemplary embodiment of the present invention;

[0048] Figure 5 A schematic diagram illustrating the identification results of the second plot of land in Miaotou Village, as provided in an exemplary embodiment of the present invention;

[0049] Figure 6 A schematic diagram of the zoning results of the second plot of land in Miaotou Village, provided as an exemplary embodiment of the present invention;

[0050] Figure 7 A schematic diagram illustrating the identification results of the integrated rice-duck farming plot in Miaotou Village, as provided in an exemplary embodiment of the present invention. Detailed Implementation

[0051] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0052] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, apparatus, product, or device that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or devices.

[0053] The concept of this invention is as follows: The rice-duck integrated farming model has strict technical requirements for both rice planting methods and duck raising management. Therefore, this invention proposes a remote sensing identification method for the rice-duck integrated farming model. Based on the growth and development characteristics of rice and the key technical points of duck raising under the rice-duck integrated farming model, and combined with the spectral, texture, and pattern characteristics of the target land cover, an identification model of rice-duck integrated farming plots is constructed through texture analysis and pixel statistical analysis. A hierarchical, step-by-step identification strategy is adopted to achieve efficient and accurate identification of the rice-duck integrated farming model. This invention is of great significance for further optimizing and promoting the rice-duck integrated farming model. Furthermore, the remote sensing identification method and process proposed in this invention provide reference experience and technical reserves for future more precise monitoring research.

[0054] In one embodiment of the present invention, see Figure 1 This paper provides a remote sensing identification method for rice-duck integrated farming models, used for remote sensing identification of rice-duck integrated farming plots, including the following steps:

[0055] Acquire remote sensing images of rice during the heading stage in the target area;

[0056] The acquired images of the target area are overlaid to obtain the normalized vegetation index and green normalized vegetation index corresponding to each pixel in the remote sensing image.

[0057] Near-infrared bands of remote sensing images of rice during the heading stage in the target area were acquired, and rice pixels and non-rice pixels were identified through texture analysis.

[0058] For each rice-growing plot within the target area, the process involves identifying whether it is the first plot. This includes using box plot analysis to calculate the upper and lower edge values ​​of the Normalized Difference Vegetation Index (NDV) and the Green Normalized Difference Vegetation Index (GDI) for a rice-growing plot. A first range is defined as the area from the lower edge of the NV to the upper edge, and a second range is defined as the area from the lower edge of the GDI to the lower edge. The process also involves calculating the percentage of pixels with NV outside the first range relative to the total number of pixels in the plot, and the percentage of pixels with GDI outside the second range relative to the total number of pixels in the plot. Based on these calculations, the rice-growing plot is identified as either the first plot or another plot.

[0059] Each first plot of land is identified as a second plot of land. The first plot of land is identified as a second plot of land or other plots of land based on the ratio of the number of non-rice pixels in the first plot of land to the total number of pixels in the plot of land.

[0060] Identifying rice-duck integrated farming plots in the second plot includes obtaining the minimum bounding rectangle of the second plot, calculating the proportion of non-rice pixels in the four vertices of the minimum bounding rectangle relative to the total number of non-rice pixels in the second plot, and identifying the second plot as a rice-duck integrated farming plot or other plots based on the calculation results.

[0061] This invention does not strictly limit the order of the above steps. Within a reasonable range, the order of the steps can be adapted or changed. The following is an example of one feasible order.

[0062] To acquire remote sensing images of the target area during the rice heading stage, in this embodiment, see [link to relevant documentation]. Figure 2 Miaotou Village, Wujiang District, Suzhou City, Jiangsu Province was selected as the study area. DJI Phantom 4 Multispectral was used to collect 0.05m resolution drone remote sensing images of the target area. Using texture analysis and pixel statistics methods, the rice-duck integrated farming area of ​​Miaotou Village was efficiently and accurately identified according to this scheme.

[0063] The acquired remote sensing images are overlaid to obtain the Normalized Difference Vegetation Index (NDVI) and Green Normalized Difference Vegetation Index (GNDVI) for each pixel in the remote sensing images. Using box plot analysis, the lower and upper edge values ​​of the NDVI and GNDVI for rice-growing plots are calculated. The range from the upper edge to the lower edge of the NDVI is determined to be a first range T1, and the range from the lower edge to the upper edge of the GNDVI is determined to be a second range T2.

[0064] The near-infrared band of remote sensing images of rice at the heading stage in the target area is acquired, and rice pixels and non-rice pixels are identified through texture analysis, including the following steps:

[0065] Acquire near-infrared band remote sensing images of rice during the heading stage in the target area;

[0066] See Figure 4 Texture analysis is performed based on the near-infrared band of remote sensing images. The texture analysis uses a 3*3 window to select the second moment for analysis.

[0067] A certain number of rice pixels and non-rice pixels are selected, and the threshold range for distinguishing rice pixels and non-rice pixels is determined based on the second moment values ​​of the rice pixel and non-rice pixel samples.

[0068] The threshold range is used to distinguish between rice pixels and non-rice pixels. If the second moment value of a pixel is within the threshold range, it is a rice pixel; if the second moment value of a pixel is not within the threshold range, it is a non-rice pixel.

[0069] See Figure 1 Within the target area, arbitrarily select a rice-growing plot to be identified, and use the first range T1 and the second range T2 to identify the rice-growing plot as the first plot or other plots, including the following steps:

[0070] Count the total number of pixels N in a rice planting plot;

[0071] If the pixel NDVI value If the NDVI value of a pixel is not within the first range, then the pixel is marked as P1; otherwise, it is marked as other. Count the number N pixels in the rice-growing plot whose normalized vegetation index is within the first range T1. P1 If the pixel GNDVI value If the pixel's GNDV value is not within the second range, then the pixel is marked as P2; otherwise, it is marked as other. Count the number N pixels in the rice-growing area whose normalized green vegetation index is outside the second range T2. P2 ;

[0072] Calculate the percentage S of the number of pixels outside the first range of the normalized vegetation index relative to the total number of pixels. P1 And calculate the percentage S of the number of pixels outside the second range of the green normalized vegetation index relative to the total number of pixels. P2 , among which, S P1 and S P2 The calculation formulas are as follows:

[0073]

[0074]

[0075] Where, N P1 N represents the number of pixels P1 within a rice-growing plot. P2 P2 represents the number of pixels within a rice-growing plot, and N represents the total number of pixels within the rice-growing plot.

[0076] If S P1 Greater than or equal to the first percentage threshold and S P2 If the percentage is greater than or equal to the second percentage threshold, then the rice-growing plot is designated as the first plot; otherwise, the rice-growing plot is designated as another plot. In this embodiment, preferably, the first percentage threshold is 4%, and the second percentage threshold is 3%. That is, when S... P1 ≥4% and S P2 When the percentage is ≥3%, the rice-growing plot is designated as the first plot; otherwise, it is designated as another plot. In this embodiment, the identification results for whether each plot in the target area is the first plot are as follows: Figure 3 As shown.

[0077] After identifying each rice-growing plot within the target area as a first plot, the selected first plots are further identified to determine if they qualify as second plots. (See also...) Figure 1 Identifying the first plot of land as the second plot or other plots based on the ratio of the number of non-rice pixels to the total number of pixels within the first plot includes the following steps:

[0078] Determine the number N of non-rice pixels within the first plot. Pother and the total number of pixels N in the plot;

[0079] Calculate the ratio S of the number of non-rice pixels within the first plot relative to its total number of pixels. Pother If the proportion of non-rice pixels to the total number of pixels in the first plot is greater than or equal to a third proportion threshold, then the first plot is the second plot; otherwise, the first plot is another plot. The proportion S of the number of non-rice pixels to the total number of pixels in the first plot. Pother The calculation formula is:

[0080]

[0081] Where, N Pother denoted as the number of non-rice pixels within the first plot, and N as the total number of pixels within the first plot.

[0082] In this embodiment, preferably, the third percentage threshold is 2%. That is, when S Pother If the percentage is ≥2%, the rice-growing plot is identified as the second plot; otherwise, the rice-growing plot is identified as another plot. In this embodiment, the identification results for whether each plot in the target area is the second plot are as follows: Figure 5 As shown.

[0083] Furthermore, based on the characteristic of setting up a grazing passage at one corner of the rice-duck farming plot, the second plot can be identified as either a rice-duck integrated farming plot or another plot. The identification method is as follows:

[0084] Using the feature management tool, obtain the minimum bounding rectangle of the second plot;

[0085] Based on the shape and orientation of the minimum bounding rectangle, it is divided into at least five partitions, four of which have the same shape and area and are located at the four vertices of the minimum bounding rectangle. Preferably, see [link to relevant documentation]. Figure 6 In this embodiment, the minimum bounding rectangle is divided into nine equal rectangular partitions. Specifically, each side of the minimum bounding rectangle is trisected, with three division points on each side. The division points on opposite sides of the minimum bounding rectangle are connected one-to-one to form nine equal rectangular partitions. The four partitions located at the four vertices of the minimum bounding rectangle are labeled as region A1, and the remaining regions are labeled as region A2. It should be noted that other reasonable methods can also be used to divide the minimum bounding rectangle into regions. For example, according to certain proportional or length requirements, four regions similar to the minimum bounding rectangle can be divided at the four corners of the minimum bounding rectangle, and these four regions can be labeled as region A1, while the remaining regions in the minimum bounding rectangle can be labeled as region A2.

[0086] Count the number N non-rice pixels within the four rectangular partitions (A1 region) located at the vertices of the minimum bounding rectangle. A1 And the total number of non-rice pixels N1 in the second plot;

[0087] Calculate the percentage S of the number of non-rice pixels within the four rectangular partitions located at the vertices of the minimum bounding rectangle relative to the total number of non-rice pixels within the minimum bounding rectangle. A1 If S A1 If the percentage is greater than or equal to the fourth percentage threshold, then the second plot is a non-rice-duck integrated farming plot; otherwise, the second plot is another type of plot. Wherein, S A1The calculation formula is:

[0088]

[0089] Where, N A1 N1 represents the number of non-rice pixels in area A1, and N1 represents the total number of non-rice pixels in the second plot.

[0090] In this embodiment, preferably, the fourth percentage threshold is 60%. That is, when S A1 If ≥60%, the second plot is identified as a rice-duck integrated farming plot; otherwise, the second plot is identified as another type of plot. In this embodiment, the identification results for whether each plot in the target area is a rice-duck integrated farming plot are as follows: Figure 7 As shown.

[0091] In another embodiment of the present invention, unlike the above embodiment which first identifies and filters out the first plot among all rice-growing plots, then identifies and filters out the second plot among all the first plots, and finally identifies and filters out the rice-duck integrated farming plot among all the second plots, this embodiment identifies each rice-growing plot within the target area to determine whether it is the first plot, the second plot, or the rice-duck integrated farming plot: if a rice-growing plot is the first plot, it is further identified to determine whether it is the second plot; if it is the second plot, it is further identified to determine whether it is the rice-duck integrated farming plot. During the identification process, if it is determined that the plot is another plot, another rice-growing plot is selected for sequential identification to determine whether it is the first plot, the second plot, or the rice-duck integrated farming plot.

[0092] In summary, this invention identifies whether a rice-growing plot is the first plot based on the growth and development characteristics of rice and the management characteristics of ducks under the rice-duck integrated farming model, using the normalized vegetation index and green normalized vegetation index of remote sensing images. It then identifies whether the first plot is the second plot based on the proportion of non-rice pixels in the rice-duck integrated farming plot. If so, it further identifies whether the second plot is a rice-duck integrated farming plot based on the characteristic that a grazing channel needs to be set up in one corner of the plot. Through hierarchical identification, this invention can efficiently and accurately identify rice-duck integrated farming plots.

[0093] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0094] The above description is only a specific embodiment of this application. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of this application, and these improvements and modifications should also be considered within the scope of protection of this application.

Claims

1. A remote sensing identification method for rice-duck integrated farming model, used for remote sensing identification of rice-duck integrated farming plots, characterized in that, Includes the following steps: Acquire remote sensing images of rice during the heading stage in the target area; The acquired images of the target area are overlaid to obtain the normalized vegetation index and green normalized vegetation index corresponding to each pixel in the remote sensing image. Near-infrared bands of remote sensing images of rice during the heading stage in the target area were acquired, and rice pixels and non-rice pixels were identified through texture analysis. The following method is used to identify whether each rice planting plot in the target area is the first plot: obtain the normalized vegetation index and green normalized vegetation index of all pixels in the rice planting plot, use the box plot analysis method to calculate the upper edge value and lower edge value of the normalized vegetation index and green normalized vegetation index of a rice planting plot, and set the lower edge value to the upper edge value of the normalized vegetation index to the first range and the lower edge value to the upper edge value of the green normalized vegetation index to the second range. Calculate the proportion of the number of pixels with normalized vegetation index outside the first range relative to the total number of pixels in the plot, and calculate the proportion of the number of pixels with green normalized vegetation index outside the second range relative to the total number of pixels in the plot, and identify the rice planting plot as the first plot or other plots based on the calculation results. Identify whether each of the first plots is a second plot, and identify whether the first plot is a second plot or other plots based on the ratio of the number of non-rice pixels in the first plot to the total number of pixels in the plot; Identifying rice-duck integrated farming plots in the second plot includes obtaining the minimum bounding rectangle of the second plot, calculating the proportion of non-rice pixels in the four vertices of the minimum bounding rectangle relative to the total number of non-rice pixels in the second plot, and identifying the second plot as a rice-duck integrated farming plot or other plots based on the calculation results.

2. The remote sensing identification method for the rice-duck integrated farming model according to claim 1, characterized in that, Identifying whether a rice-growing plot is the first plot includes the following steps: Count the total number of pixels N in a rice planting plot; Count the number N pixels in rice-growing plots whose normalized vegetation index (NVI) falls outside the first range. P1 And the number N of pixels whose green normalized vegetation index is outside the second range. P2 ; Calculate the percentage S of the number of pixels with normalized vegetation index outside the first range relative to the total number of pixels in the plot. P1 And calculate the percentage S of the number of pixels outside the second range of the green normalized vegetation index relative to the total number of pixels. P2 , among which, S P1 =N P1 / N,S P2 =N P2 / N; If S P1 Greater than or equal to the first percentage threshold and S P2 If the proportion is greater than or equal to the second threshold, then the rice-growing plot is the first plot; otherwise, the rice-growing plot is another plot.

3. The remote sensing identification method for the rice-duck integrated farming model according to claim 1, characterized in that, Identifying whether the first plot of land is the second plot of land includes the following steps: Determine the number of non-rice pixels within the first plot; Calculate the proportion of non-rice pixels in the first plot relative to its total number of pixels; If the proportion of non-rice pixels to the total number of pixels is greater than or equal to the third proportion threshold, then the first plot is the second plot; otherwise, the first plot is another plot.

4. The remote sensing identification method for the rice-duck integrated farming model according to claim 1, characterized in that, Calculate the proportion of non-rice pixels in the four vertex corner regions of the minimum bounding rectangle relative to the total number of non-rice pixels in the second plot, and identify the second plot as a rice-duck integrated farming plot or other plots based on the calculation results, including the following steps: The minimum bounding rectangle is divided into at least five partitions, of which four partitions have the same shape and area and are located at the four vertices of the minimum bounding rectangle. The number of non-rice pixels in the four rectangular partitions located at the vertices of the minimum bounding rectangle and the total number of non-rice pixels in the second plot are counted. Calculate the proportion of non-rice pixels in the four rectangular partitions located at the apex of the minimum bounding rectangle relative to the total number of non-rice pixels in the second plot. If the calculation result is greater than or equal to the fourth proportion threshold, then the second plot is a rice-duck integrated farming plot; otherwise, the second plot is another plot.

5. The remote sensing identification method for the rice-duck integrated farming model according to claim 4, characterized in that, Dividing the minimum bounding rectangle into nine equally divided rectangular partitions includes the following steps: Divide each side of the minimum bounding rectangle into three equal parts, with three division points corresponding to each side; Connect the equal division points on opposite sides of the minimum bounding rectangle one by one to form nine equal rectangular partitions.

6. The remote sensing identification method for the rice-duck integrated farming model according to claim 2, characterized in that, The first percentage threshold is 4%; and / or, The second percentage threshold is 3%.

7. The remote sensing identification method for the rice-duck integrated farming model according to claim 3, characterized in that, The third percentage threshold is 2%.

8. The remote sensing identification method for the rice-duck integrated farming model according to claim 4, characterized in that, The fourth percentage threshold is 60%.

9. The remote sensing identification method for the rice-duck integrated farming model according to claim 1, characterized in that, Determining rice pixels from non-rice pixels involves the following steps: Acquire near-infrared band remote sensing images of rice during the heading stage in the target area; Texture analysis is performed based on the near-infrared band of remote sensing images. The texture analysis uses a 3*3 window to select the second moment for analysis. A certain number of rice pixels and non-rice pixels are selected, and the threshold range for distinguishing rice pixels and non-rice pixels is determined based on the second moment values ​​of the rice pixel and non-rice pixel samples. The threshold range is used to distinguish between rice pixels and non-rice pixels. If the second moment value of a pixel is within the threshold range, it is a rice pixel; if the second moment value of a pixel is not within the threshold range, it is a non-rice pixel.

10. The remote sensing identification method for the rice-duck integrated farming model according to claim 1, characterized in that, Using the feature management tool, obtain the minimum bounding rectangle of the second plot.

Images