Sugarcane yield prediction method, device and equipment based on remote sensing image data

Abstract

The present application relates to the field of remote sensing data analysis, and particularly relates to a sugarcane yield prediction method based on remote sensing image data, which comprises the following steps: obtaining remote sensing image data of a target sugarcane field area; extracting texture features according to the remote sensing image data to obtain texture feature data, wherein the texture feature data comprises gray level co-occurrence matrix texture feature data of several bands, and the gray level co-occurrence matrix texture feature data comprises mean feature, homogeneity feature and dissimilarity feature; calculating double difference ratio texture index according to the texture feature data to obtain double difference ratio texture index data, inputting the double difference ratio texture index data into a pre-trained sugarcane yield prediction model to predict the sugarcane yield and obtain sugarcane yield prediction data, wherein the sugarcane yield prediction model is a model constructed by using a support vector algorithm, with each type of double difference ratio texture index as an explanatory variable and the sugarcane yield as a dependent variable. The present application realizes rapid and accurate prediction of the sugarcane yield.

Description

Technical Field

[0001] This invention relates to the field of remote sensing data analysis, and in particular to a method, apparatus, device, and storage medium for predicting sugarcane yield based on remote sensing image data. Background Technology

[0002] With the development of remote sensing technology, spectral vegetation indices based on satellite imagery are widely used for crop yield estimation. However, commonly used low-to-medium resolution satellite imagery such as Sentinel-2 or Landsat series cannot capture the fine heterogeneity within a field. Furthermore, in the middle and late stages of sugarcane growth, when biomass is large, vegetation indices are prone to "saturation," leading to a significant decrease in the prediction accuracy of high-yield areas.

[0003] However, existing technologies for estimating sugarcane yield using high-resolution imagery still have the following shortcomings: Existing studies often neglect in-depth analysis of texture information or rely solely on a single, fixed texture window size and orientation for coarse analysis. Furthermore, they overlook the spatial heterogeneity and directional sensitivity of sugarcane under row-planting patterns, limiting further improvements in the accuracy of sugarcane yield prediction. Summary of the Invention

[0004] Based on this, the purpose of this invention is to provide a method, apparatus, device, and storage medium for predicting sugarcane yield based on remote sensing image data. The method involves extracting texture features from remote sensing image data, calculating the double difference ratio texture index, generating double difference ratio texture index data corresponding to the remote sensing image data, and combining this with a pre-trained sugarcane yield prediction model to predict sugarcane yield. This achieves rapid and accurate prediction of sugarcane yield while reducing costs.

[0005] In a first aspect, embodiments of this application provide a method for predicting sugarcane yield based on remote sensing image data, comprising the following steps:

[0006] Obtain remote sensing image data of the target sugarcane field area; Texture features are extracted from the remote sensing image data to obtain texture feature data of the target sugarcane field area. The texture feature data includes gray-level co-occurrence matrix texture feature data of several bands. The gray-level co-occurrence matrix texture feature data includes mean features, homogeneity features, and dissimilarity features. The mean features are used to characterize the average reflectivity of the sugarcane canopy. The homogeneity features and dissimilarity features are used to characterize the structural complexity of the sugarcane canopy and reflect the sugarcane's resistance to interference. The double difference ratio texture index is calculated based on the texture feature data to obtain the double difference ratio texture index data of the target sugarcane field area. The double difference ratio texture index data includes several types of double difference ratio texture indices. The double-difference ratio texture index data is input into a pre-trained sugarcane yield prediction model to predict sugarcane yield, thereby obtaining sugarcane yield prediction data for the target sugarcane field area. The sugarcane yield prediction model is a model constructed using support vector algorithms with various types of double-difference ratio texture indices as explanatory variables and sugarcane yield as the dependent variable.

[0007] Secondly, embodiments of this application provide a sugarcane yield prediction device based on remote sensing image data, comprising: The data acquisition module is used to acquire remote sensing image data of the target sugarcane field area; The feature extraction module is used to extract texture features from the remote sensing image data to obtain texture feature data of the target sugarcane field area. The texture feature data includes gray-level co-occurrence matrix texture feature data of several bands. The gray-level co-occurrence matrix texture feature data includes mean features, homogeneity features, and dissimilarity features. The mean features are used to characterize the average reflectivity of the sugarcane canopy. The homogeneity features and dissimilarity features are used to characterize the structural complexity of the sugarcane canopy and reflect the degree of interference resistance of the sugarcane. The index calculation module is used to calculate the double difference ratio texture index based on the texture feature data to obtain the double difference ratio texture index data of the target sugarcane field area, wherein the double difference ratio texture index data includes several types of double difference ratio texture indices. The yield prediction module is used to input the double difference ratio texture index data into a pre-trained sugarcane yield prediction model to predict sugarcane yield and obtain sugarcane yield prediction data for the target sugarcane field area. The sugarcane yield prediction model is a model constructed using support vector algorithms with various types of double difference ratio texture indices as explanatory variables and sugarcane yield as the dependent variable.

[0008] Thirdly, embodiments of this application provide a computer device, including: a processor, a memory, and a computer program stored in the memory and executable on the processor; when the computer program is executed by the processor, it implements the steps of the sugarcane yield prediction method based on remote sensing image data as described in the first aspect.

[0009] Fourthly, embodiments of this application provide a storage medium storing a computer program that, when executed by a processor, implements the steps of the sugarcane yield prediction method based on remote sensing image data as described in the first aspect.

[0010] In this application embodiment, a method, apparatus, device, and storage medium for predicting sugarcane yield based on remote sensing image data are provided. The method involves extracting texture features from remote sensing image data, calculating the double difference ratio texture index, generating double difference ratio texture index data corresponding to the remote sensing image data, and combining it with a pre-trained sugarcane yield prediction model to predict sugarcane yield. This achieves rapid and accurate prediction of sugarcane yield and reduces costs.

[0011] To better understand and implement this invention, the following detailed description is provided in conjunction with the accompanying drawings. Attached Figure Description

[0012] Figure 1 A flowchart illustrating a sugarcane yield prediction method based on remote sensing image data provided in one embodiment of this application; Figure 2 This is a flowchart illustrating step S2 of a sugarcane yield prediction method based on remote sensing image data provided in one embodiment of this application. Figure 3 This is a flowchart illustrating step S3 of a sugarcane yield prediction method based on remote sensing image data provided in one embodiment of this application. Figure 4 A flowchart illustrating a sugarcane yield prediction method based on remote sensing image data, provided as another embodiment of this application; Figure 5 This is a flowchart illustrating step S53 of a sugarcane yield prediction method based on remote sensing image data provided in one embodiment of this application. Figure 6 A flowchart illustrating step S533 of a sugarcane yield prediction method based on remote sensing image data provided in an embodiment of this application; Figure 7 A schematic diagram of the structure of a sugarcane yield prediction device based on remote sensing image data provided in one embodiment of this application; Figure 8 This is a schematic diagram of the structure of a computer device provided in one embodiment of this application. Detailed Implementation

[0013] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0014] The terminology used in this application is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. The singular forms “a,” “the,” and “the” used in this application and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used herein refers to and includes any or all possible combinations of one or more of the associated listed items.

[0015] It should be understood that although the terms first, second, third, etc., may be used in this application to describe various information, such information should not be limited to these terms. These terms are only used to distinguish information of the same type from one another. For example, without departing from the scope of this application, first information may also be referred to as second information, and similarly, second information may also be referred to as first information. Depending on the context, the word "if" as used herein may be interpreted as "when," "when," or "in response to determination."

[0016] Please see Figure 1 , Figure 1 The following is a flowchart illustrating a sugarcane yield prediction method based on remote sensing image data according to an embodiment of this application. The method includes the following steps: S1: Obtain remote sensing image data of the target sugarcane field area.

[0017] The main body executing the sugarcane yield prediction method based on remote sensing image data is the prediction device (hereinafter referred to as the prediction device). In an optional embodiment, the prediction device can be a computer device, a server, or a server cluster composed of multiple computer devices.

[0018] In this embodiment, the prediction device obtains remote sensing image data of the target sugarcane field area. Specifically, the remote sensing image data uses high-resolution satellite imagery such as Jilin-1, which can provide sub-meter level spatial details and provide a data foundation for extracting fine crop canopy textures.

[0019] S2: Extract texture features based on the remote sensing image data to obtain texture feature data of the target sugarcane field area.

[0020] In this embodiment, the prediction device extracts texture features based on the remote sensing image data to obtain texture feature data of the target sugarcane field area. The texture feature data includes gray-level co-occurrence matrix texture feature data of several bands, and the gray-level co-occurrence matrix texture feature data includes mean features, homogeneity features, and dissimilarity features.

[0021] Specifically, the prediction device employs the gray-level co-occurrence matrix (GLCM) method to extract texture features from the remote sensing image data, obtaining GLCM texture feature data for several bands. The GLCM quantifies the spatial texture features of the canopy by statistically analyzing the joint distribution of pixel gray values ​​at specified distances and directions in the image, and is one of the most classic and commonly used methods for extracting image texture features. Mean, homogeneity, and dissimilarity are classic texture indicators extracted by the GLCM, capturing canopy spatial structure information from different dimensions and reflecting differences in sugarcane growth and canopy structure. The combination of these three indicators provides a more comprehensive characterization of crop status and effectively improves the accuracy of yield prediction models.

[0022] The mean feature is the average value of all elements in the gray-level co-occurrence matrix, reflecting the overall brightness of the entire texture matrix. In images, the mean feature is used to characterize the average reflectance of the sugarcane canopy. The more vigorous the sugarcane growth and the denser the leaves, the higher the average reflectance and the higher the mean feature. It is usually positively correlated with biomass and yield.

[0023] The homogeneity and dissimilarity features are used to characterize the structural complexity of the sugarcane canopy and reflect the sugarcane's resistance to disturbance. The homogeneity feature measures the concentration of elements in the gray-level co-occurrence matrix near the main diagonal, reflecting the local uniformity of the texture. For sugarcane fields, high homogeneity usually indicates a more uniform canopy, more uniform growth, and less disturbance from pests, diseases, and lodging. Such fields tend to have more stable yields.

[0024] Dissimilarity measures the degree of difference between adjacent gray levels in a gray-level co-occurrence matrix, reflecting the degree of texture non-uniformity. The coarser the texture of a remote sensing image, the greater the difference in gray levels between adjacent pixels, and the more dispersed the matrix elements, the higher the dissimilarity value. In sugarcane fields, high dissimilarity usually corresponds to complex canopy structures, numerous gaps, withered leaves, and lodging; such fields often have lower growth and yield.

[0025] Please see Figure 2 , Figure 2 The flowchart of step S2 in the sugarcane yield prediction method based on remote sensing image data provided in one embodiment of this application includes step S21, as follows: S21: Based on the remote sensing image data and the preset gray-level co-occurrence matrix texture feature calculation algorithm, obtain the mean features, homogeneity features and dissimilarity features of each band.

[0026] The algorithm for calculating the gray-level co-occurrence matrix texture features is as follows:

[0027]

[0028]

[0029] In the formula, It is a characteristic of the mean. It is a homogeneous characteristic. It is a characteristic of dissimilarity. This represents the row number of the gray-level co-occurrence matrix. The number of columns in the gray-level co-occurrence matrix. Indicates spatial distance ,direction Under the condition, the gray value is and The normalized frequency of occurrence of adjacent pixel pairs.

[0030] In this embodiment, the prediction device obtains the mean features, homogeneity features, and dissimilarity features of each band based on the remote sensing image data and a preset gray-level co-occurrence matrix texture feature calculation algorithm.

[0031] S3: Calculate the double difference ratio texture index based on the texture feature data to obtain the double difference ratio texture index data of the target sugarcane field area.

[0032] In this embodiment, the prediction device calculates the double difference ratio texture index based on the texture feature data to obtain the double difference ratio texture index data of the target sugarcane field area. The double difference ratio texture index data includes several types of double difference ratio texture indices.

[0033] The bands include the blue band, green band, red band, and near-infrared band. Please refer to [link / reference]. Figure 3 , Figure 3 The flowchart of step S3 in the sugarcane yield prediction method based on remote sensing image data provided in one embodiment of this application includes step S31, as follows: S31: Based on the mean characteristics, homogeneity characteristics, dissimilarity characteristics of each band and the preset double difference ratio texture index calculation algorithm, obtain the mean double difference ratio texture index, the homogeneity double difference ratio texture index and the dissimilarity double difference ratio texture index.

[0034] The algorithm for calculating the double difference ratio texture index is as follows:

[0035]

[0036]

[0037] In the formula, The mean-difference ratio texture index. The homogeneity double difference ratio texture index, The dissimilarity ratio texture index is the ratio of the two differences between the two textures. The mean characteristic of the blue band is... The mean characteristics are in the near-infrared band. The mean characteristic of the green band is... The mean characteristic of the red band is... The homogeneity of the red band is a characteristic. The homogeneity characteristic of the near-infrared band The homogeneity characteristic of the blue band, The homogeneity of the green band is a characteristic. The dissimilarity characteristic of the red band, The anisotropy characteristics in the near-infrared band This is a characteristic of the dissimilarity of the blue band. This is the anisotropy characteristic of the green band.

[0038] In this embodiment, the prediction device obtains the mean double difference ratio texture index, the homogeneity double difference ratio texture index, and the dissimilarity double difference ratio texture index based on the mean characteristics, homogeneity characteristics, dissimilarity characteristics, and the preset double difference ratio texture index calculation algorithm of each band.

[0039] S4: Input the double difference ratio texture index data into the pre-trained sugarcane yield prediction model to predict sugarcane yield and obtain sugarcane yield prediction data for the target sugarcane field area.

[0040] In this embodiment, the prediction device inputs the double difference ratio texture index data into a pre-trained sugarcane yield prediction model to predict sugarcane yield and obtain sugarcane yield prediction data for the target sugarcane field area. The sugarcane yield prediction model is a model constructed using the support vector algorithm with various types of double difference ratio texture indices as explanatory variables and sugarcane yield as the dependent variable.

[0041] Texture features extracted from remote sensing image data are used to calculate the double difference ratio texture index, generating double difference ratio texture index data corresponding to the remote sensing image data. This data is then combined with a pre-trained sugarcane yield prediction model to predict sugarcane yield, achieving rapid and accurate sugarcane yield prediction while reducing costs.

[0042] Please see Figure 4 , Figure 4 A flowchart illustrating a sugarcane yield prediction method based on remote sensing image data, provided for another embodiment of this application, further includes steps S51-S53, which, prior to step S4, are as follows: S51: Obtain remote sensing image data of several sugarcane maturity stages in the sample sugarcane field area, as well as sugarcane yield label data corresponding to each sample remote sensing image data.

[0043] In this embodiment, the prediction device obtains remote sensing image data of several sugarcane maturity stages in the sample sugarcane field area, as well as sugarcane yield label data corresponding to each sample remote sensing image data.

[0044] S52: Calculate the double difference ratio texture index based on the remote sensing image data of each sample to obtain the double difference ratio texture index data of the remote sensing image data of each sample.

[0045] In this embodiment, the prediction device calculates the double difference ratio texture index based on each of the sample remote sensing image data to obtain the double difference ratio texture index data of each of the sample remote sensing image data. For specific implementation, please refer to step S31, which will not be repeated here.

[0046] S53: Input the double difference ratio texture index data of each sample remote sensing image data and the sugarcane yield label data corresponding to each sample remote sensing image data into the sugarcane yield prediction model to be trained. Use the support vector algorithm and leave-one-out cross-validation method to build and validate the model, and obtain the trained sugarcane yield prediction model.

[0047] In this embodiment, the prediction device inputs the double difference ratio texture index data of each sample remote sensing image data and the sugarcane yield label data corresponding to each sample remote sensing image data into the sugarcane yield prediction model to be trained. The support vector algorithm and leave-one-out cross-validation method are used to build and validate the model, and the trained sugarcane yield prediction model is obtained, realizing high-precision prediction of sugarcane yield.

[0048] Please see Figure 5 , Figure 5 The flowchart of S53 in the sugarcane yield prediction method based on remote sensing image data provided in one embodiment of this application includes steps S531 to S533, as follows: S531: Divide the remote sensing image data of each sample to obtain remote sensing image data of each training sample and remote sensing image data of each validation sample.

[0049] In this embodiment, the prediction device divides the remote sensing image data of each sample to obtain remote sensing image data of each training sample and remote sensing image data of each validation sample.

[0050] S532: Train the model based on the double difference ratio texture index data of each training sample remote sensing image data and the sugarcane yield label data corresponding to each training sample remote sensing image data to obtain several trained candidate models; input each validation sample remote sensing image data into each trained candidate model to predict sugarcane yield, and obtain the sugarcane yield prediction data of each validation sample remote sensing image data output by each trained candidate model.

[0051] In this embodiment, the prediction device trains the model based on the double difference ratio texture index data of each training sample remote sensing image data and the sugarcane yield label data corresponding to each training sample remote sensing image data, and obtains several trained candidate models.

[0052] The prediction device inputs the remote sensing image data of each validation sample into each trained candidate model to predict sugarcane yield, and obtains the sugarcane yield prediction data of each validation sample remote sensing image data output by each trained candidate model.

[0053] S533: Based on the sugarcane yield prediction data of each validation sample remote sensing image data output by each trained candidate model and the sugarcane yield label data corresponding to each training sample remote sensing image data, calculate the model accuracy parameters to obtain the model accuracy parameters of each trained candidate model; based on the model accuracy parameters, determine the target model from each trained candidate model as the sugarcane yield prediction model.

[0054] In this embodiment, the prediction device calculates the model accuracy parameters based on the sugarcane yield prediction data of each validation sample remote sensing image data output by each trained candidate model and the sugarcane yield label data corresponding to each training sample remote sensing image data, thereby obtaining the model accuracy parameters of each trained candidate model.

[0055] The prediction device determines the target model from the various trained candidate models based on the model accuracy parameters, and uses it as the sugarcane yield prediction model to improve the accuracy of the sugarcane yield prediction model in predicting sugarcane yield.

[0056] The model accuracy parameters include the model determination coefficient and root mean square error; please refer to [link / reference]. Figure 6 , Figure 6 The flowchart of S533 in the sugarcane yield prediction method based on remote sensing image data provided in one embodiment of this application includes step S5331, as follows: S5331: Based on the sugarcane yield prediction data of each validation sample remote sensing image data output by each trained candidate model, the sugarcane yield label data corresponding to each training sample remote sensing image data, and the preset model accuracy parameter calculation algorithm, obtain the model accuracy parameters of each trained candidate model.

[0057] The algorithm for calculating the model accuracy parameters is as follows:

[0058]

[0059] In the formula, The coefficient of determination for the model The root mean square error, For sugarcane yield labeling data, The sugarcane yield prediction data is generated from the validation sample remote sensing imagery data output by the current trained candidate model. This represents the average of the sugarcane yield label data. To verify the quantity of sample remote sensing image data.

[0060] In this embodiment, the prediction device obtains the model accuracy parameters of each trained candidate model based on the sugarcane yield prediction data of each validation sample remote sensing image data output by each trained candidate model, the sugarcane yield label data corresponding to each training sample remote sensing image data, and the preset model accuracy parameter calculation algorithm.

[0061] Please refer to Figure 7 , Figure 7 This is a schematic diagram of a sugarcane yield prediction device based on remote sensing image data according to an embodiment of this application. The device can be implemented entirely or partially through software, hardware, or a combination of both. The device 7 includes: Data acquisition module 71 is used to acquire remote sensing image data of the target sugarcane field area; The feature extraction module 72 is used to extract texture features based on the remote sensing image data to obtain texture feature data of the target sugarcane field area. The texture feature data includes gray-level co-occurrence matrix texture feature data of several bands. The gray-level co-occurrence matrix texture feature data includes mean features, homogeneity features, and dissimilarity features. The mean features are used to characterize the average reflectivity of the sugarcane canopy. The homogeneity features and dissimilarity features are used to characterize the structural complexity of the sugarcane canopy and reflect the degree of interference resistance of the sugarcane. The index calculation module 73 is used to calculate the double difference ratio texture index based on the texture feature data to obtain the double difference ratio texture index data of the target sugarcane field area, wherein the double difference ratio texture index data includes several types of double difference ratio texture indices. The yield prediction module 74 is used to input the double difference ratio texture index data into a pre-trained sugarcane yield prediction model to predict sugarcane yield and obtain sugarcane yield prediction data for the target sugarcane field area. The sugarcane yield prediction model is a model constructed using support vector algorithms with various types of double difference ratio texture indices as explanatory variables and sugarcane yield as the dependent variable.

[0062] In this embodiment, a data acquisition module obtains remote sensing image data of the target sugarcane field area; a feature extraction module extracts texture features from the remote sensing image data to obtain texture feature data of the target sugarcane field area. The texture feature data includes gray-level co-occurrence matrix texture feature data for several bands, which includes mean features, homogeneity features, and dissimilarity features. The mean feature characterizes the average reflectance of the sugarcane canopy; the homogeneity and dissimilarity features characterize the structural complexity of the sugarcane canopy, reflecting the sugarcane's resistance to interference. The index calculation module calculates the double-difference ratio texture index based on the texture feature data to obtain the double-difference ratio texture index data for the target sugarcane field area. This double-difference ratio texture index data includes several types of double-difference ratio texture indices. The yield prediction module inputs this double-difference ratio texture index data into a pre-trained sugarcane yield prediction model to predict the sugarcane yield for the target sugarcane field area. This sugarcane yield prediction model is constructed using a support vector algorithm, with each type of double-difference ratio texture index as the explanatory variable and sugarcane yield as the dependent variable. By calculating the double-difference ratio texture index based on texture feature extraction from remote sensing image data, corresponding double-difference ratio texture index data corresponding to the remote sensing image data is generated. This data, combined with the pre-trained sugarcane yield prediction model, is used to predict sugarcane yield, achieving rapid and accurate sugarcane yield prediction while reducing costs.

[0063] Please refer to Figure 8 , Figure 8 This is a schematic diagram of the structure of a computer device provided in one embodiment of this application. The computer device 8 includes: a processor 81, a memory 82, and a computer program 83 stored in the memory 82 and executable on the processor 81; the computer device can store multiple instructions, which are adapted to be loaded and executed by the processor 81. Figures 1 to 5 The method steps of the illustrated embodiment can be found in the following documentation for detailed execution. Figures 1 to 5 The specific details of the illustrated embodiments will not be elaborated here.

[0064] The processor 81 may include one or more processing cores. The processor 81 connects to various parts of the server using various interfaces and lines. It executes various functions and processes data in the sugarcane yield prediction device 7 based on remote sensing image data by running or executing instructions, programs, code sets, or instruction sets stored in the memory 82, and by calling data from the memory 82. Optionally, the processor 81 may be implemented using at least one hardware form of Digital Signal Processing (DSP), Field-Programmable Gate Array (FPGA), or Programmable Logic Array (PLA). The processor 81 may integrate one or more of the following: a Central Processing Unit (CPU), a Graphics Processing Unit (GPU), and a modem. The CPU primarily handles the operating system, user interface, and applications; the GPU is responsible for rendering and drawing the content to be displayed on the touch screen; and the modem handles wireless communication. It is understood that the modem may also be implemented as a separate chip without being integrated into the processor 81.

[0065] The memory 82 may include random access memory (RAM) or read-only memory. Optionally, the memory 82 may include a non-transitory computer-readable storage medium. The memory 82 can be used to store instructions, programs, code, code sets, or instruction sets. The memory 82 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as touch instructions), instructions for implementing the various method embodiments described above, etc.; the data storage area may store data involved in the various method embodiments described above, etc. Optionally, the memory 82 may also be at least one storage device located remotely from the aforementioned processor 81.

[0066] This application embodiment also provides a storage medium that can store multiple instructions, which are adapted to be loaded and executed by a processor as described above. Figures 1 to 5 The method steps of the illustrated embodiment can be found in the following documentation for detailed execution. Figures 1 to 5 The specific details of the illustrated embodiments will not be elaborated here.

[0067] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the functional units and modules are only for easy differentiation and are not intended to limit the scope of protection of this application. The specific working process of the units and modules in the above system can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0068] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0069] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the algorithm. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0070] In the embodiments provided by this invention, it should be understood that the disclosed apparatus / terminal devices and methods can be implemented in other ways. For example, the apparatus / terminal device embodiments described above are merely illustrative. For instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0071] 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 units can be selected to achieve the purpose of this embodiment according to actual needs.

[0072] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0073] If the integrated module / unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments of the present invention can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms.

[0074] This invention is not limited to the above-described embodiments. If any modifications or variations to this invention do not depart from the spirit and scope of this invention, and if such modifications and variations fall within the scope of the claims and equivalent technologies of this invention, then this invention also intends to include such modifications and variations.

Claims

1. A method for predicting sugarcane yield based on remote sensing image data, characterized in that, Includes the following steps: Obtain remote sensing image data of the target sugarcane field area; Texture features are extracted from the remote sensing image data to obtain texture feature data of the target sugarcane field area. The texture feature data includes gray-level co-occurrence matrix texture feature data of several bands. The gray-level co-occurrence matrix texture feature data includes mean features, homogeneity features, and dissimilarity features. The mean features are used to characterize the average reflectivity of the sugarcane canopy. The homogeneity features and dissimilarity features are used to characterize the structural complexity of the sugarcane canopy and reflect the sugarcane's resistance to interference. The double difference ratio texture index is calculated based on the texture feature data to obtain the double difference ratio texture index data of the target sugarcane field area. The double difference ratio texture index data includes several types of double difference ratio texture indices. The double-difference ratio texture index data is input into a pre-trained sugarcane yield prediction model to predict sugarcane yield, thereby obtaining sugarcane yield prediction data for the target sugarcane field area. The sugarcane yield prediction model is a model constructed using support vector algorithms with various types of double-difference ratio texture indices as explanatory variables and sugarcane yield as the dependent variable.

2. The sugarcane yield prediction method based on remote sensing image data according to claim 1, characterized in that, The step of extracting texture features from the remote sensing image data to obtain texture feature data includes the following steps: Based on the remote sensing image data and a preset gray-level co-occurrence matrix texture feature calculation algorithm, the mean feature, homogeneity feature, and dissimilarity feature of each band are obtained. The gray-level co-occurrence matrix texture feature calculation algorithm is as follows: In the formula, It is a characteristic of the mean. It is a homogeneous characteristic. It is a characteristic of dissimilarity. This represents the row number of the gray-level co-occurrence matrix. The number of columns in the gray-level co-occurrence matrix. Indicates spatial distance ,direction Under the condition, the gray value is and The normalized frequency of occurrence of adjacent pixel pairs.

3. The sugarcane yield prediction method based on remote sensing image data according to claim 2, characterized in that: The dual-difference ratio texture index includes the mean dual-difference ratio texture index, the homogeneity dual-difference ratio texture index, and the dissimilar dual-difference ratio texture index. The mean dual-difference ratio texture index is used to reflect the difference in canopy brightness and is related to leaf biomass and leaf area index. The homogeneity dual-difference ratio texture index is used to reflect the canopy uniformity and is related to the uniformity and density of sugarcane growth. The dissimilar dual-difference ratio texture index is used to reflect the roughness of canopy texture and is related to stem structure and plant height variation.

4. The sugarcane yield prediction method based on remote sensing image data according to claim 3, characterized in that: The bands include blue band, green band, red band, and near-infrared band; The step of calculating the double difference ratio texture index based on the texture feature data to obtain the double difference ratio texture index data of the target sugarcane field area includes the following steps: Based on the mean characteristics, homogeneity characteristics, and dissimilarity characteristics of each band, and a preset double-difference ratio texture index calculation algorithm, the mean double-difference ratio texture index, the homogeneity double-difference ratio texture index, and the dissimilarity double-difference ratio texture index are obtained. The double-difference ratio texture index calculation algorithm is as follows: In the formula, The mean-difference ratio texture index. The homogeneity double difference ratio texture index, The dissimilarity ratio texture index is the ratio of the two differences between the two textures. The mean characteristic of the blue band is... The mean characteristics are in the near-infrared band. The mean characteristic of the green band is... The mean characteristic of the red band is... The homogeneity characteristic of the red band, The homogeneity characteristic of the near-infrared band The homogeneity characteristic of the blue band, The homogeneity of the green band is a characteristic. The dissimilarity characteristic of the red band, The anisotropy characteristics in the near-infrared band This is a characteristic of the dissimilarity of the blue band. This is the anisotropy characteristic of the green band.

5. The sugarcane yield prediction method based on remote sensing image data according to claim 1 or 4, characterized in that, Before inputting the double-difference ratio texture index data into a pre-trained sugarcane yield prediction model to predict sugarcane yield and obtain sugarcane yield prediction data for the target sugarcane field area, the following steps are also included: Obtain remote sensing image data of several sugarcane maturity stages in the sample sugarcane field area, as well as sugarcane yield label data corresponding to each sample remote sensing image data; The double difference ratio texture index is calculated based on the remote sensing image data of each sample to obtain the double difference ratio texture index data of each sample remote sensing image data. The double difference ratio texture index data of each sample remote sensing image data and the sugarcane yield label data corresponding to each sample remote sensing image data are input into the sugarcane yield prediction model to be trained. The support vector algorithm and leave-one-out cross-validation method are used to build and validate the model to obtain the trained sugarcane yield prediction model.

6. The sugarcane yield prediction method based on remote sensing image data according to claim 5, characterized in that, The process of building and validating a sugarcane yield prediction model using support vector algorithms and leave-one-out cross-validation to obtain the trained model includes the following steps: The remote sensing image data of each sample is divided to obtain remote sensing image data of each training sample and remote sensing image data of each validation sample. The model is trained based on the double difference ratio texture index data of each training sample remote sensing image data and the sugarcane yield label data corresponding to each training sample remote sensing image data to obtain several trained candidate models; the remote sensing image data of each validation sample is input into each trained candidate model to predict sugarcane yield, and the sugarcane yield prediction data of each validation sample remote sensing image data output by each trained candidate model is obtained. Based on the sugarcane yield prediction data of each validation sample remote sensing image data output by each trained candidate model and the sugarcane yield label data corresponding to each training sample remote sensing image data, the model accuracy parameters are calculated to obtain the model accuracy parameters of each trained candidate model; based on the model accuracy parameters, the target model is determined from each trained candidate model as the sugarcane yield prediction model.

7. The sugarcane yield prediction method based on remote sensing image data according to claim 6, characterized in that: The model accuracy parameters include the model determination coefficient and the root mean square error; The step of calculating model accuracy parameters based on the sugarcane yield prediction data of each validation sample remote sensing image data and the sugarcane yield label data corresponding to each training sample remote sensing image data to obtain the model accuracy parameters of each trained candidate model includes the following steps: Based on the sugarcane yield prediction data of each validation sample remote sensing image data output by each trained candidate model, the sugarcane yield label data corresponding to each training sample remote sensing image data, and the preset model accuracy parameter calculation algorithm, the model accuracy parameters of each trained candidate model are obtained. The model accuracy parameter calculation algorithm is as follows: In the formula, The coefficient of determination for the model. The root mean square error, For sugarcane yield labeling data, The sugarcane yield prediction data is generated from the validation sample remote sensing imagery data output by the current trained candidate model. This represents the average of the sugarcane yield label data. To verify the quantity of sample remote sensing image data.

8. A sugarcane yield prediction device based on remote sensing image data, characterized in that, include: The data acquisition module is used to acquire remote sensing image data of the target sugarcane field area; The feature extraction module is used to extract texture features from the remote sensing image data to obtain texture feature data of the target sugarcane field area. The texture feature data includes gray-level co-occurrence matrix texture feature data of several bands. The gray-level co-occurrence matrix texture feature data includes mean features, homogeneity features, and dissimilarity features. The mean features are used to characterize the average reflectivity of the sugarcane canopy. The homogeneity features and dissimilarity features are used to characterize the structural complexity of the sugarcane canopy and reflect the degree of interference resistance of the sugarcane. The index calculation module is used to calculate the double difference ratio texture index based on the texture feature data to obtain the double difference ratio texture index data of the target sugarcane field area, wherein the double difference ratio texture index data includes several types of double difference ratio texture indices. The yield prediction module is used to input the double difference ratio texture index data into a pre-trained sugarcane yield prediction model to predict sugarcane yield and obtain sugarcane yield prediction data for the target sugarcane field area. The sugarcane yield prediction model is a model constructed using support vector algorithms with various types of double difference ratio texture indices as explanatory variables and sugarcane yield as the dependent variable.

9. A computer device, characterized in that, include: A processor, a memory, and a computer program stored in the memory and executable on the processor; the computer program, when executed by the processor, implements the steps of the sugarcane yield prediction method based on remote sensing image data as described in any one of claims 1 to 7.

10. A storage medium, characterized in that: The storage medium stores a computer program that, when executed by a processor, implements the steps of the sugarcane yield prediction method based on remote sensing image data as described in any one of claims 1 to 7.

Images