Remote sensing identification method for seed maize based on automatic plot identification and texture analysis

Abstract

This invention provides a remote sensing identification method for seed maize based on automatic plot identification and texture analysis. The method includes: inputting high-resolution satellite imagery of a target area into a target identification model to identify maize plots, thereby obtaining maize plots within the target area; the target identification model is trained using samples and their corresponding labels; the samples and their corresponding labels are obtained through visual interpretation of the original high-resolution satellite imagery; and identifying seed maize plots within the target area based on the texture features of the maize plots. This invention improves the accuracy of seed maize plot identification and enables rapid and accurate acquisition of seed maize planting distribution and area estimation, providing technical support for the monitoring and supervision of seed maize.

Description

Technical Field

[0001] This invention relates to the field of crop identification technology, and in particular to a remote sensing identification method for seed maize based on automatic plot identification and texture analysis. Background Technology

[0002] As a major food crop in my country, corn requires rapid and accurate monitoring of its seed production area to ensure agricultural seed supply security and strengthen seed production supervision. Currently, the area of ​​traditional corn and seed corn is typically obtained by seed management departments, which is heavily influenced by human factors, resulting in low efficiency and slow speed.

[0003] Remote sensing technology, with its advantages of speed, wide coverage, and real-time acquisition of rich information over large areas of the earth's surface, effectively compensates for the shortcomings of traditional technologies and has become an important means of obtaining crop information. Combining information such as planting methods, phenological characteristics, and spectral and textural features of remote sensing images with those of seed maize, objectively and timely obtaining information on the planting area and spatial distribution of seed maize is an inevitable choice for achieving precise monitoring of seed maize production.

[0004] Previous monitoring studies on seed maize were based on medium- to high-resolution remote sensing data, using vegetation indices and texture structure features to extract data at the pixel scale. However, the estimation of seed maize planting area based on pixel-scale texture feature analysis, according to the seed maize planting pattern, suffers from low accuracy. Summary of the Invention

[0005] To address the problems existing in the prior art, this invention provides a remote sensing identification method for seed maize based on automatic plot identification and texture analysis.

[0006] This invention provides a remote sensing identification method for seed maize based on automatic plot identification and texture analysis, comprising:

[0007] High-resolution satellite imagery of the target area is input into a target recognition model to identify cornfields within the target area. The target recognition model is trained using samples and their corresponding labels. The samples and their corresponding labels are obtained through visual interpretation of the original high-resolution satellite imagery.

[0008] Based on the texture features of the cornfields within the target area, seed cornfields are identified within the cornfields of the target area.

[0009] In some embodiments, inputting high-resolution satellite imagery of the target area into a target recognition model for cornfield identification to obtain cornfields within the target area includes:

[0010] The high-resolution satellite imagery is segmented into multiple images of the target size;

[0011] The multiple images are input into the target recognition model to identify cornfields, thereby obtaining the cornfields corresponding to the multiple images;

[0012] The cornfields corresponding to the multiple images are stitched together to obtain the cornfields within the target area.

[0013] In some embodiments, after inputting high-resolution satellite imagery of the target area into a target recognition model to identify cornfields and obtain the cornfields within the target area, the method further includes:

[0014] Perform image morphological opening operations on the cornfields within the target area to determine the completely segmented cornfields;

[0015] The patches in the completely segmented cornfield are deleted.

[0016] In some embodiments, identifying seed corn plots within the target area based on the texture features of corn plots within the target area includes:

[0017] Based on a window of a preset size, the red band of the image of the cornfield within the target area is traversed to determine the gray-level co-occurrence matrix.

[0018] The entropy feature of the gray-level co-occurrence matrix is ​​determined as the texture feature;

[0019] Based on the texture features, seed corn plots are identified in the corn plots within the target area.

[0020] In some embodiments, identifying seed corn plots within the target area based on the texture features includes:

[0021] Based on the average value of the texture features and the box plot analysis method, the segmentation threshold is determined;

[0022] Based on the aforementioned segmentation threshold, cornfields within the target area are classified into two categories to determine seed cornfields.

[0023] This invention also provides a remote sensing identification device for seed maize based on automatic plot identification and texture analysis, comprising:

[0024] The first identification module is used to input high-resolution satellite imagery of the target area into the target identification model to identify cornfields within the target area; the target identification model is trained using samples and their corresponding labeling; the samples and their corresponding labeling are obtained through visual interpretation of the original high-resolution satellite imagery.

[0025] The second identification module is used to identify seed corn plots in the cornfields within the target area based on the texture features of the cornfield plots within the target area.

[0026] In some embodiments, the first identification module is specifically used for:

[0027] The high-resolution satellite imagery is segmented into multiple images of the target size;

[0028] The multiple images are input into the target recognition model to identify cornfields, thereby obtaining the cornfields corresponding to the multiple images;

[0029] The cornfields corresponding to the multiple images are stitched together to obtain the cornfields within the target area.

[0030] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and running on the processor, wherein when the processor executes the program, it implements the seed corn remote sensing identification method based on automatic plot identification and texture analysis as described above.

[0031] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the seed corn remote sensing identification method based on automatic plot identification and texture analysis as described above.

[0032] The present invention also provides a computer program product, including a computer program that, when executed by a processor, implements the seed corn remote sensing identification method based on automatic land parcel identification and texture analysis as described above.

[0033] The present invention provides a remote sensing identification method for seed corn based on automatic plot identification and texture analysis. It automatically identifies corn plots in high-resolution satellite images of target areas using a target identification model, and remotely identifies seed corn plots based on texture features. This improves the identification accuracy of seed corn plots and enables rapid and accurate acquisition of seed corn planting distribution and area estimation, providing technical support for the monitoring and supervision of seed corn. Attached Figure Description

[0034] To more clearly illustrate the technical solutions in this invention 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 some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0035] Figure 1 This is one of the flowcharts of the remote sensing identification method for seed maize based on automatic plot identification and texture analysis provided in this embodiment of the invention;

[0036] Figure 2 This is the second flowchart of the remote sensing identification method for seed maize based on automatic plot identification and texture analysis provided in this embodiment of the invention.

[0037] Figure 3 This is one of the schematic diagrams of the automatic land parcel identification results provided in an embodiment of the present invention;

[0038] Figure 4 This is a second schematic diagram of the automatic land parcel identification results provided in an embodiment of the present invention;

[0039] Figure 5 This is a schematic diagram of the structure of the remote sensing identification device for seed maize based on automatic plot identification and texture analysis provided in an embodiment of the present invention;

[0040] Figure 6 This is a schematic diagram of the structure of the electronic device provided in an embodiment of the present invention. Detailed Implementation

[0041] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0042] The terms "first," "second," etc., used in the specification and claims of this invention are used to distinguish similar objects and not to describe a specific order or sequence. It should be understood that such terms can be used interchangeably where appropriate so that embodiments of the invention can be implemented in orders other than those illustrated or described herein, and the objects distinguished by "first" and "second" are generally of the same class, not limited in number; for example, the first object can be one or more. Furthermore, in the specification and claims, "and / or" indicates at least one of the connected objects, and the character " / " generally indicates that the preceding and following objects are in an "or" relationship.

[0043] Figure 1 This is one of the flowcharts illustrating the remote sensing identification method for seed maize based on automatic plot identification and texture analysis provided in this embodiment of the invention. Figure 1 As shown in the embodiment of the present invention, the remote sensing identification method for seed maize based on automatic plot identification and texture analysis includes:

[0044] Step 101: Input the high-resolution satellite image of the target area into the target recognition model to identify cornfields, thereby obtaining the cornfields within the target area; the target recognition model is trained using samples and their corresponding label; the samples and their corresponding label are obtained through visual interpretation of the original high-resolution satellite image.

[0045] Step 102: Based on the texture features of the cornfields within the target area, identify seed cornfields within the target area.

[0046] It should be noted that the executing entity of the remote sensing identification method for seed corn based on automatic plot identification and texture analysis provided by this invention can be an electronic device, a component in an electronic device, an integrated circuit, or a chip. The electronic device can be a mobile electronic device or a non-mobile electronic device. For example, mobile electronic devices can be mobile phones, tablets, laptops, handheld computers, in-vehicle electronic devices, wearable devices, ultra-mobile personal computers (UMPCs), netbooks, or personal digital assistants (PDAs), etc., while non-mobile electronic devices can be servers, network-attached storage (NAS), personal computers (PCs), televisions (TVs), ATMs, or self-service machines, etc., and this invention does not impose specific limitations.

[0047] In step 101, high-resolution satellite imagery of the target area is input into the target recognition model to identify cornfields, thereby obtaining cornfields within the target area. The target recognition model is trained using samples and their corresponding labeling. The samples and their corresponding labeling are obtained through visual interpretation of the original high-resolution satellite imagery.

[0048] Visual interpretation, also known as visual analysis or visual interpretation, is a type of remote sensing image interpretation and is the reverse process of remote sensing imaging. It refers to the process by which professionals acquire information about specific targets and features from remote sensing images through direct observation or with the aid of interpreting instruments. Visual interpretation involves using the human eye (or optical instruments), relying on the interpreter's knowledge, experience, and relevant data, and through analysis, reasoning, and judgment, to extract useful information from remote sensing images.

[0049] For the acquired raw high-resolution satellite images, through visual interpretation, cornfields in certain areas of the study area were manually vectorized to obtain the raw high-resolution satellite images.

[0050] The acquired cornfield vector files were converted into raster files, and the original remote sensing images were cropped, enhanced, and segmented into 512*512 images as samples and their corresponding labels. Training and test sets were then constructed based on these samples and their corresponding labels.

[0051] Input the training set into U 2 The -net network is trained by setting various hyperparameters and tested on a test set to obtain a trained target recognition model. This target recognition model is used to extract cornfield patches from high-resolution satellite imagery.

[0052] In some embodiments, inputting high-resolution satellite imagery of the target area into a target recognition model for cornfield identification to obtain cornfields within the target area includes:

[0053] The high-resolution satellite imagery is segmented into multiple images of the target size;

[0054] The multiple images are input into the target recognition model to identify cornfields, thereby obtaining the cornfields corresponding to the multiple images;

[0055] The cornfields corresponding to the multiple images are stitched together to obtain the cornfields within the target area.

[0056] The high-resolution satellite imagery of the target area was captured and then cut into multiple images of size 512*512.

[0057] Then, the segmented images are input into the trained target recognition model for cornfield identification, which can extract the raster data of cornfields corresponding to multiple images.

[0058] Multiple images corresponding to cornfields are stitched together to obtain cornfield data within the target area.

[0059] Optionally, after inputting high-resolution satellite imagery of the target area into the target recognition model for cornfield identification to obtain the cornfields within the target area, the method further includes:

[0060] Perform image morphological opening operations on the cornfields within the target area to determine the completely segmented cornfields;

[0061] The patches in the completely segmented cornfield are deleted.

[0062] Image morphological opening operations are performed on the segmented cornfield data to break the abnormal connections between the plots, i.e., some plots in the image that cannot be completely segmented, to obtain the fully segmented cornfield plots.

[0063] At the same time, small patches in the image are filtered out and deleted. The vector polygon data of the cornfield is obtained by using the raster-to-vector method, that is, the independent and complete cornfield.

[0064] In step 102, seed corn plots are identified in the corn plots within the target area based on the texture features of the corn plots within the target area.

[0065] In some embodiments, identifying seed corn plots within the target area based on the texture features of corn plots within the target area includes:

[0066] Based on a window of a preset size, the red band of the image of the cornfield within the target area is traversed to determine the gray-level co-occurrence matrix.

[0067] The entropy feature of the gray-level co-occurrence matrix is ​​determined as the texture feature;

[0068] Based on the texture features, seed corn plots are identified in the corn plots within the target area.

[0069] By setting a 7x7 window, the red band of the image of the cornfield within the target area is traversed, and a gray-level co-occurrence matrix is ​​generated based on the window size.

[0070] The entropy features of the gray-level co-occurrence matrix are extracted to obtain the gray-level co-occurrence matrix entropy feature image of the target area. The entropy features can be used as remote sensing identification texture features for seed corn and field corn.

[0071] In some embodiments, identifying seed corn plots within the target area based on the texture features includes:

[0072] Based on the average value of the texture features and the box plot analysis method, the segmentation threshold is determined;

[0073] Based on the aforementioned segmentation threshold, cornfields within the target area are classified into two categories to determine seed cornfields.

[0074] By calculating the average entropy characteristics within maize plots within the target area, and using box plot analysis, the segmentation thresholds for the entropy characteristics of seed maize and field maize were statistically determined.

[0075] Based on this segmentation threshold, all plots are classified into two categories to obtain classification maps of seed maize and field maize in the study area.

[0076] Finally, the classified plots are stitched together to form a remote sensing identification map of seed corn and field corn in the target area.

[0077] The seed corn remote sensing identification method based on automatic plot identification and texture analysis provided in this invention uses a target identification model to automatically identify corn plots in high-resolution satellite images of the target area, and performs remote sensing identification of seed corn plots based on texture features. This improves the identification accuracy of seed corn plots and enables rapid and accurate acquisition of seed corn planting distribution and area estimation, providing technical support for the monitoring and supervision of seed corn.

[0078] Figure 2 This is the second flowchart illustrating the remote sensing identification method for seed maize based on automatic plot identification and texture analysis provided in this embodiment of the invention. Figure 2 As shown in the embodiment of the present invention, the remote sensing identification method for seed maize based on automatic plot identification and texture analysis includes:

[0079] (1) Training set construction

[0080] Based on the original high-resolution satellite imagery, cornfields in certain areas of the study area were manually vectorized through visual interpretation. The vector files of the cornfields were converted into raster files and then cropped and enhanced with the original remote sensing imagery to form 512*512 images, which were used as training set images and labels.

[0081] (2) Based on U 2 Automatic extraction of cornfields segmented by the -net network.

[0082] Input the training set into U 2 The -net network was trained by setting various hyperparameters. The high-resolution satellite images were cut into multiple regions of 512*512 size. The trained network model was used to input the segmented image dataset into the network model to achieve image segmentation of the images and obtain raster data of cornfields. Then, the blocks were stitched together to obtain segmented cornfield data of the study area.

[0083] (3) Morphological treatment of cornfields

[0084] Image morphological opening operations are performed on the segmented cornfield data to break up some incompletely segmented fields in the image; at the same time, small patches in the image are filtered out and deleted, and vector polygonal cornfield data are obtained by raster-to-vector conversion.

[0085] (4) Extraction of gray-level co-occurrence matrix entropy features at the plot scale

[0086] Set a 7x7 window to traverse the red band of the plot image, calculate and generate a gray-level co-occurrence matrix based on the window, and extract the entropy features of the matrix to obtain the gray-level co-occurrence matrix entropy feature image of the study area.

[0087] (5) Identification of seed corn

[0088] The average entropy characteristics within maize plots were calculated. Box plot analysis was used to determine the segmentation thresholds for the entropy characteristics of seed maize and field maize. Based on these segmentation thresholds, all plots were classified into two categories to obtain classification maps of seed maize and field maize in the study area.

[0089] The following describes the remote sensing identification method for seed maize based on automatic land parcel identification and texture analysis provided by the embodiments of the present invention in a specific scenario.

[0090] First, the study area and data were determined. The study area is located in the western part of a district of a city in a certain province. 0.5-meter resolution images were collected on August 5, 2022.

[0091] The experimental results of the remote sensing identification method for seed maize based on automatic plot identification and texture analysis provided in this embodiment of the invention include:

[0092] (1) Based on U 2 Cornfield segmentation results using the -net network

[0093] U 2 After training, the -net network takes a 512x512 pixel image set from the images to be classified and inputs it into the trained deep learning network for automatic land parcel extraction. Figure 3 This is one of the schematic diagrams of the automatic land parcel identification results provided in the embodiments of the present invention. Figure 4 This is the second schematic diagram of the automatic land parcel identification results provided in this embodiment of the invention. The land parcel extraction results are as follows: Figure 3 and Figure 4 As shown.

[0094] (2) Calculation results of gray-level co-occurrence matrix entropy characteristics at the plot scale

[0095] At the plot scale, the gray-level co-occurrence matrix of the images was statistically calculated and entropy features were extracted. Box plot statistical analysis revealed that the entropy features of seed maize plots were significantly higher than those of field maize. Entropy features can be effectively used to identify seed maize.

[0096] (3) Seed maize identification results

[0097] Based on the box plot analysis results and accuracy evaluation, the entropy feature threshold for seed maize was determined to be 1.55. After classifying the plots according to the threshold, the plots were then mosaicked and fused to obtain the seed maize identification results for the image areas of the study area.

[0098] (4) Evaluation of classification accuracy

[0099] The classification results were validated based on the obtained validation sample points. The overall accuracy was 94.52%, and the kappa coefficient was 0.886. The specific classification accuracy results are shown in Table 1.

[0100] Table 1

[0101]

[0102]

[0103] The seed corn remote sensing identification method based on automatic plot identification and texture analysis provided in this invention provides a method for seed corn remote sensing identification that combines automatic plot identification and texture analysis based on high-resolution satellite imagery. This method enables rapid and accurate acquisition of seed corn planting distribution and area estimation, providing technical support for the monitoring and supervision of seed corn.

[0104] The following describes the remote sensing identification device for seed maize based on automatic plot identification and texture analysis provided by the present invention. The remote sensing identification device for seed maize based on automatic plot identification and texture analysis described below can be referred to in correspondence with the remote sensing identification method for seed maize based on automatic plot identification and texture analysis described above.

[0105] Figure 5 This is a schematic diagram of the remote sensing identification device for seed maize based on automatic plot identification and texture analysis provided in an embodiment of the present invention, as shown below. Figure 5 As shown in the figure, the seed corn remote sensing identification device based on automatic plot identification and texture analysis provided in this embodiment of the invention includes:

[0106] The first identification module 510 is used to input high-resolution satellite imagery of the target area into the target identification model to identify cornfields within the target area; the target identification model is trained using samples and their corresponding labeling; the samples and their corresponding labeling are obtained through visual interpretation of the original high-resolution satellite imagery.

[0107] The second identification module 520 is used to identify seed corn plots in the corn plots within the target area based on the texture features of the corn plots within the target area.

[0108] It should be noted that the seed corn remote sensing identification device based on automatic plot identification and texture analysis provided in this embodiment of the invention can realize all the method steps implemented in the above-mentioned seed corn remote sensing identification method embodiment based on automatic plot identification and texture analysis, and can achieve the same technical effect. Here, the parts that are the same as those in the method embodiment and the beneficial effects will not be described in detail.

[0109] Optionally, the first identification module 510 is specifically used for:

[0110] The high-resolution satellite imagery is segmented into multiple images of the target size;

[0111] The multiple images are input into the target recognition model to identify cornfields, thereby obtaining the cornfields corresponding to the multiple images;

[0112] The cornfields corresponding to the multiple images are stitched together to obtain the cornfields within the target area.

[0113] Optionally, it also includes: a deletion module, used for:

[0114] Perform image morphological opening operations on the cornfields within the target area to determine the completely segmented cornfields;

[0115] The patches in the completely segmented cornfield are deleted.

[0116] Optionally, the second identification module 520 is specifically used for:

[0117] Based on a window of a preset size, the red band of the image of the cornfield within the target area is traversed to determine the gray-level co-occurrence matrix.

[0118] The entropy feature of the gray-level co-occurrence matrix is ​​determined as the texture feature;

[0119] Based on the texture features, seed corn plots are identified in the corn plots within the target area.

[0120] Optionally, the second identification module 520 is specifically used for:

[0121] Based on the average value of the texture features and the box plot analysis method, the segmentation threshold is determined;

[0122] Based on the aforementioned segmentation threshold, cornfields within the target area are classified into two categories to determine seed cornfields.

[0123] Figure 6 An example is a schematic diagram of the physical structure of an electronic device, such as... Figure 6 As shown, the electronic device may include a processor 610, a communication interface 620, a memory 630, and a communication bus 640, wherein the processor 610, the communication interface 620, and the memory 630 communicate with each other through the communication bus 640. The processor 610 can call logical instructions in the memory 630 to execute a remote sensing identification method for seed corn based on automatic plot identification and texture analysis. The method includes: inputting high-resolution satellite imagery of the target area into a target identification model to identify corn plots, thereby obtaining corn plots within the target area; the target identification model is trained using samples and their corresponding sample labels; the samples and their corresponding sample labels are obtained through visual interpretation of the original high-resolution satellite imagery; and identifying seed corn plots within the target area based on the texture features of the corn plots within the target area.

[0124] Furthermore, the logical instructions in the aforementioned memory 630 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, essentially, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0125] On the other hand, the present invention also provides a computer program product, which includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the remote sensing identification method for seed corn based on automatic plot identification and texture analysis provided by the above methods. The method includes: inputting high-resolution satellite imagery of the target area into a target identification model to identify corn plots, thereby obtaining corn plots within the target area; the target identification model is trained using samples and corresponding sample labels; the samples and corresponding sample labels are obtained through visual interpretation of the original high-resolution satellite imagery; and identifying seed corn plots within the corn plots within the target area based on the texture features of the corn plots within the target area.

[0126] In another aspect, the present invention also provides a non-transitory computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the remote sensing identification method for seed corn based on automatic plot identification and texture analysis provided by the above methods. This method includes: inputting high-resolution satellite imagery of a target area into a target identification model to identify corn plots, thereby obtaining corn plots within the target area; the target identification model is trained using samples and corresponding sample labels; the samples and corresponding sample labels are obtained through visual interpretation of the original high-resolution satellite imagery; and identifying seed corn plots within the corn plots within the target area based on the texture features of the corn plots within the target area.

[0127] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0128] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0129] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for remote sensing identification of seed corn based on plot automatic identification and texture analysis, characterized in that, include: High-resolution satellite imagery of the target area is input into the target recognition model to identify cornfields, thereby obtaining cornfields within the target area; The target recognition model is trained using samples and their corresponding labeling; the samples and their corresponding labeling are obtained through visual interpretation of the original high-resolution satellite imagery. Based on the texture features of the cornfields within the target area, seed cornfields are identified within the cornfields of the target area. The step of identifying seed corn plots within the target area based on the texture features of corn plots in the target area includes: Based on a window of a preset size, the red band of the image of the cornfield within the target area is traversed to determine the gray-level co-occurrence matrix. The entropy feature of the gray-level co-occurrence matrix is ​​determined as the texture feature; Based on the texture features, seed corn plots are identified in the corn plots within the target area; The step of identifying seed corn plots within the target area based on the texture features includes: Based on the average value of the texture features and the box plot analysis method, the segmentation threshold is determined; Based on the aforementioned segmentation threshold, cornfields within the target area are classified into two categories to determine seed cornfields.

2. The method according to claim 1, wherein, The step of inputting high-resolution satellite imagery of the target area into a target recognition model to identify cornfields and obtain cornfields within the target area includes: The high-resolution satellite imagery is segmented into multiple images of the target size; The multiple images are input into the target recognition model to identify cornfields, thereby obtaining the cornfields corresponding to the multiple images; The cornfields corresponding to the multiple images are stitched together to obtain the cornfields within the target area.

3. The remote sensing identification method for seed maize based on automatic plot identification and texture analysis according to claim 1, characterized in that, After inputting high-resolution satellite imagery of the target area into the target recognition model for cornfield identification, and obtaining the cornfields within the target area, the method further includes: Perform image morphological opening operations on the cornfields within the target area to determine the completely segmented cornfields; The patches in the completely segmented cornfield are deleted.

4. A remote sensing identification device for seed maize based on automatic plot identification and texture analysis, characterized in that, include: The first identification module is used to input high-resolution satellite images of the target area into the target identification model to identify cornfields and obtain cornfields within the target area; The target recognition model is trained using samples and their corresponding labeling; the samples and their corresponding labeling are obtained through visual interpretation of the original high-resolution satellite imagery. The second identification module is used to identify seed corn plots in the corn plots within the target area based on the texture features of the corn plots within the target area. The second identification module is specifically used for: Based on a window of a preset size, the red band of the image of the cornfield within the target area is traversed to determine the gray-level co-occurrence matrix. The entropy feature of the gray-level co-occurrence matrix is ​​determined as the texture feature; Based on the texture features, seed corn plots are identified in the corn plots within the target area; The second identification module is specifically used for: Based on the average value of the texture features and the box plot analysis method, the segmentation threshold is determined; Based on the aforementioned segmentation threshold, cornfields within the target area are classified into two categories to determine seed cornfields.

5. The remote sensing identification device for seed maize based on automatic plot identification and texture analysis according to claim 4, characterized in that, The first identification module is specifically used for: The high-resolution satellite imagery is segmented into multiple images of the target size; The multiple images are input into the target recognition model to identify cornfields, thereby obtaining the cornfields corresponding to the multiple images; The cornfields corresponding to the multiple images are stitched together to obtain the cornfields within the target area.

6. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the program, it implements the remote sensing identification method for seed maize based on automatic plot identification and texture analysis as described in any one of claims 1 to 3.

7. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the remote sensing identification method for seed maize based on automatic plot identification and texture analysis as described in any one of claims 1 to 3.

8. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the remote sensing identification method for seed maize based on automatic plot identification and texture analysis as described in any one of claims 1 to 3.

Images