Low-efficiency land use identification method and system based on big data analysis

Abstract

The invention relates to the technical field of land recognition based on big data analysis, and particularly discloses a low-efficiency land recognition method and system based on big data analysis, and the method comprises the steps: recognizing a land coverage type through a satellite remote sensing image technology, and obtaining a farmland land as a monitoring target region; then, dividing the target area into a plurality of sub-areas, and obtaining crop data and environment state data of each sub-area through remote sensing image analysis and ground survey; thirdly, evaluating the abnormal degree of crop production by adopting a One-Class SVM algorithm, and analyzing and evaluating the influence of the environmental state on the crops in combination with the environmental interference factor; and finally, through comprehensive analysis of production abnormity and environment interference degree of each sub-region, identifying low-efficiency land use sub-regions, and feeding back the low-efficiency land use sub-regions to a land manager for the land manager to decide whether to re-plan or take other intervention measures.

Description

Technical Field

[0001] The present invention relates to the technical field of land identification based on big data analysis, and in particular to a method and system for identifying inefficient land based on big data analysis. Background Art

[0002] As the global population continues to grow, the shortage of land resources has become more prominent, especially in some developing countries. The efficiency of agricultural land use and the rational allocation of land resources have become important issues in land management and agricultural production. Traditional land use planning methods often rely on manual surveys and empirical judgments. This method is not only labor-intensive and time-consuming, but also lacks efficient data support, making it difficult to adjust and optimize land use strategies in a rapidly changing environment.

[0003] In agricultural production, due to the impact of various factors (such as climate change, environmental pollution, backward agricultural technology, etc.) on the production capacity of land, some farmlands have been inefficiently used. Inefficient land use usually means that the land has not fully utilized its production potential, the yield is low or it has not been used for a long time, resulting in a waste of agricultural resources. In addition, environmental problems such as soil degradation and water shortage may also lead to a decline in land productivity. Traditional farmland management and planning methods are difficult to accurately identify these inefficient areas, especially in large-scale farmland and complex environments. How to scientifically evaluate land use efficiency has become an urgent problem for land managers to solve. Summary of the invention

[0004] The purpose of the present invention is to provide a method and system for identifying inefficient land use based on big data analysis to solve the problems in the above background.

[0005] The purpose of the present invention can be achieved through the following technical solutions: The inefficient land identification method based on big data analysis includes the following steps: S1: Identify land cover types based on satellite remote sensing image technology, including forest land, farmland land and urban built-up area land, and obtain farmland land as the target area for monitoring; S2: Divide the target area into several sub-areas, obtain the crop data in each sub-area, including: crop yield, output value and species, analyze the crop data in each sub-area, and evaluate the abnormality of crop production in each sub-area; S3: Obtain environmental status data in each sub-area, including air quality data and water quality data, analyze the environmental status data, and evaluate the degree of interference of the environmental status data on crops; S4: Comprehensively analyze the abnormal degree of crop production in each sub-region and the degree of interference of the environmental status data of each sub-region on crops to determine whether the current sub-region is low-efficiency land; S5: Based on the judgment result, extract the sub-areas of inefficient land use, and feed back the sub-areas of inefficient land use to the manager to determine whether the land in the sub-areas needs to be replanned.

[0006] As a further solution of the present invention: the analysis of the crop data in each sub-region and the evaluation of the abnormality of crop production in each sub-region specifically include: Get the crop yield, crop output value and crop species in each sub-region; By constructing feature vectors for crop yield, crop output value and crop species in each sub-region, the feature vectors are used as inputs of the One-Class SVM model and the One-Class SVM model is trained. According to the trained One-Class SVM model, the value of the decision function and the minimum value of the objective function are output, and the sub-region abnormality coefficient is obtained by calculating the ratio of the value of the decision function to the minimum value of the objective function; Determine whether the sub-region abnormal coefficient is greater than or equal to the preset threshold. If not, record it as a normal sub-region; if so, record it as an abnormal sub-region.

[0007] As a further solution of the present invention: the process of obtaining the sub-region abnormality coefficient is: Obtain the mean and standard deviation of crop yields in the target area, calculate the difference between the crop yield in each sub-area and the mean of the crop yield in the target area, and take the absolute value to obtain the yield deviation value; The yield deviation value is calculated by ratio with the standard deviation of the crop yield in the target area to obtain the standardized value of the crop yield in each sub-area; Obtain the mean and standard deviation of the crop output value in the target area, calculate the difference between the crop output value in each sub-area and the mean of the crop output value in the target area, and take the absolute value to obtain the output value deviation value; The output value deviation value is calculated by ratio with the standard deviation of the crop output value in the target area to obtain the standardized value of the crop output value in each sub-area; For discrete features, including crop species, one-hot encoding was used to convert each species into a binary vector; For discrete features, including crop species, one-hot encoding was used to convert each species into a binary vector; Construct the feature matrix, , Represents the feature vector of the i-th sub-region; Wherein, i represents the number of sub-regions, i=1, 2, ..., n, and n represents the maximum number of sub-regions; The feature vector Input into the One-Class SVM model for training; The goal of the model is to distinguish normal data from abnormal data by optimizing a hyperplane, which is a decision function constructed by the normal vector w and the bias term b; The calculation expression of the decision function is: ; In the formula, represents the decision function, x represents the input feature vector, represents the transpose of the normal vector w, and b represents the bias term; In the optimization process, the objective function is used to maximize the interval between normal data points and the hyperplane, pull normal data points away from the hyperplane, and minimize the number of abnormal data points; Among them, the calculation expression of the objective function is: ; In the formula, Represents maximizing the interval from normal data points to the hyperplane, v represents the preset scale factor, represents the slack variable, represents the minimum value of the objective function; After training is completed, the model will output the normal vector w and the bias term b, which will be used to predict new data points; The sub-region anomaly coefficient is obtained by calculating the ratio of the decision function value to the minimum value of the objective function. .

[0008] As a further solution of the present invention: the analysis of the environmental status data and the evaluation of the interference degree of the environmental status data on crops specifically include: Obtain air quality and water quality data in each sub-region; Process the air quality and water quality data in each sub-area, and calculate the interference factor in each sub-area based on the air quality and water quality data in each sub-area; It is determined whether the interference factor in each sub-region is greater than or equal to a preset interference factor threshold. If so, it is recorded as a high interference level sub-region; if not, it is recorded as a low interference level sub-region.

[0009] As a further solution of the present invention: the process of obtaining the interference factor is: De-mean the collected air quality and water quality data; A zero-mean series is obtained by calculating the mean of the data and subtracting the mean from each data point; The air quality and water quality data after de-averaging are converted into frequency domain data through discrete Fourier transform; Through Fourier transform, periodic fluctuations and energy components of different frequencies in environmental state data are identified; The calculation expression of the Fourier transform is: ; In the formula, Indicates frequency The corresponding Fourier coefficients are, represents the kth discrete frequency, k represents the number of frequencies, represents the data point at the mth moment, represents the data after removing the mean, m represents the number of acquisition time points, and M represents the maximum number of acquisition time points. represents the interval between two adjacent time points, and j represents an imaginary unit; Through the Fourier coefficients, we further calculate the power spectral density of the environmental data. The calculation expression is: ; In the formula, Indicates frequency The power density at By accumulating the power spectral density in the entire frequency range, the total fluctuation energy of the environmental data is obtained. ; By analyzing the power spectral density and total fluctuation energy, the interference factor is calculated through the frequency-weighted power spectrum. The calculation expression is: ; In the formula, represents the interference factor of the ith sub-region, and E represents the maximum number of frequencies.

[0010] As a further solution of the present invention: the determining whether the current sub-region is low-efficiency land specifically includes: Get the sub-region anomaly coefficient within each sub-region and interference factors ; For each sub-region, the sub-region anomaly coefficient and interference factors Perform normalization and calculate the land use score in each sub-area; The land use score is used to determine whether each sub-region is low-efficiency land; It is determined whether the land use score in each sub-region is greater than or equal to a preset threshold. If not, the current sub-region is low-efficiency land.

[0011] As a further solution of the present invention: the determination of whether the land in the sub-region needs to be replanned specifically includes: Feedback the identified inefficient land use sub-areas to land managers by generating detailed GIS maps to indicate the location, size and influencing factors of inefficient land use; For sub-areas with high interference levels, measures are taken to reduce PM2.5 concentrations, PM10 concentrations, COD concentrations and ammonia nitrogen concentrations, and the interference factors and land use scores are recalculated. If the land use score is less than the preset threshold, the current area will be replanned.

[0012] Inefficient land use identification system based on big data analysis, including: The target area determination module identifies the land cover types, including forest land, farmland land and urban built-up area land, based on satellite remote sensing image technology, and obtains farmland land as the target area for monitoring. A data collection module, the data collection module is used to obtain crop data in each sub-area, including: crop yield, output value and species; Obtain environmental status data in each sub-area, including: air quality data and water quality data; An abnormality evaluation module, which is used to analyze the crop data in each sub-region and evaluate the abnormality of crop production in each sub-region; An interference degree assessment module, wherein the interference degree assessment module is used to analyze the environmental status data and assess the interference degree of the environmental status data on the crops; A comprehensive analysis module, which is used to comprehensively analyze the abnormality of crop production in each sub-region and the interference degree of the environmental status data of each sub-region on the crops; An inefficient land identification module, wherein the inefficient land identification module identifies whether each sub-area is inefficient land according to the analysis result of the comprehensive analysis module; The low-efficiency land optimization module extracts the low-efficiency land sub-areas according to the judgment result, feeds back the low-efficiency land sub-areas to the manager, and judges whether the land in the sub-areas needs to be replanned.

[0013] Beneficial effects of the present invention: (1) The present invention can accurately identify inefficient areas in land use through an inefficient land identification method based on big data analysis. Through comprehensive analysis of remote sensing imaging technology, crop data, and environmental monitoring data, the present invention can accurately evaluate each sub-area, identify these inefficient lands, and provide corresponding optimization solutions. Through this method, managers can accurately understand which lands are not effectively used, and then take re-planning or intervention measures, such as improving soil quality, adjusting crop planting structure, or increasing infrastructure construction such as irrigation, so as to improve the land utilization rate and production efficiency of these areas; (2) By analyzing the environmental status data of each sub-region (such as air quality, ammonia nitrogen concentration, water quality indicators, etc.) and the degree of interference with crops, the invention can provide accurate risk assessment and intervention recommendations for agricultural production; crop production in many areas is affected by environmental pollution and climate change, resulting in reduced yields or damaged crop quality, and long-term reliance on traditional agricultural methods is often difficult to cope with these changes; through the technology of the present invention, areas with high interference levels can be identified in a timely manner, providing managers with improvement plans. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] The present invention will be further described below in conjunction with the accompanying drawings.

[0015] Figure 1 It is a flowchart of the specific steps of the low-efficiency land identification method based on big data analysis of the present invention; Figure 2 It is a flowchart of the low-efficiency land identification system based on big data analysis in the present invention. DETAILED DESCRIPTION

[0016] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0017] See also Figure 1 As shown, the present invention is a method for identifying inefficient land use based on big data analysis, comprising the following steps: S1: Identify land cover types based on satellite remote sensing image technology, including forest land, farmland land and urban built-up area land, and obtain farmland land as the target area for monitoring; S2: Divide the target area into several sub-areas, obtain the crop data in each sub-area, including: crop yield, output value and species, analyze the crop data in each sub-area, and evaluate the abnormality of crop production in each sub-area; S3: Obtain environmental status data in each sub-area, including air quality data (such as PM2.5, PM10) and water quality data (such as COD, ammonia nitrogen), analyze the environmental status data, and evaluate the degree of interference of the environmental status data on crops; S4: Comprehensively analyze the abnormal degree of crop production in each sub-region and the degree of interference of the environmental status data of each sub-region on crops to determine whether the current sub-region is low-efficiency land; S5: Based on the judgment result, extract the sub-areas of inefficient land use, and feed back the sub-areas of inefficient land use to the manager to determine whether the land in the sub-areas needs to be replanned.

[0018] In S1, land cover types are identified based on satellite remote sensing image technology, including forest land, farmland land and urban built-up area land, and farmland land is obtained as the target area for monitoring, including: Obtain image data within the monitoring area through satellite remote sensing technology, including reflection information of different ground objects, such as vegetation, soil, and water bodies; Use remote sensing image analysis software to pre-process images, including radiation correction, geometric correction, etc., to ensure the accuracy of image data; Using image classification technology, different areas in satellite remote sensing images are classified according to land cover types; The areas in the image are divided into forest land, farmland land, and urban built-up area land through the spectral characteristics of remote sensing data; Extract the farmland area as the target area for monitoring; Through further data screening and verification, we ensure that the extracted farmland meets the research requirements and provide accurate geographic information basis for subsequent crop data collection and analysis; In S2, the target area is divided into several sub-areas, and the crop data in each sub-area is obtained, including: crop yield, output value and species. The crop data in each sub-area is analyzed to evaluate the abnormality of crop production in each sub-area, including: Divide the target area into several identical sub-areas; Through remote sensing image analysis, ground surveys, and agricultural statistical data, the crop yield in each sub-region is obtained; The crop output value of each sub-region is calculated based on the market price and yield of the crops; Obtain the crop species planted in each sub-region and count the dominant crops in each sub-region; Standardization of collected crop yields and crop output values, including: Obtain the mean and standard deviation of crop yields in the target area, calculate the difference between the crop yield in each sub-area and the mean of the crop yield in the target area, and take the absolute value to obtain the yield deviation value; The yield deviation value is calculated by ratio with the standard deviation of the crop yield in the target area to obtain the standardized value of the crop yield in each sub-area; Obtain the mean and standard deviation of the crop output value in the target area, calculate the difference between the crop output value in each sub-area and the mean of the crop output value in the target area, and take the absolute value to obtain the output value deviation value; The output value deviation value is calculated by ratio with the standard deviation of the crop output value in the target area to obtain the standardized value of the crop output value in each sub-area; For discrete features, including crop species, one-hot encoding was used to convert each species into a binary vector; Construct the feature matrix, , Represents the feature vector of the i-th sub-region; Wherein, i represents the number of sub-regions, i=1, 2, ..., n, and n represents the maximum number of sub-regions; The feature vector Input into the One-Class SVM model for training; The goal of the model is to distinguish normal data from abnormal data by optimizing a hyperplane, which is a decision function constructed by the normal vector w and the bias term b; The calculation expression of the decision function is: ; In the formula, represents the decision function, x represents the input feature vector, represents the transpose of the normal vector w, and b represents the bias term; In the optimization process, the objective function is used to maximize the interval between normal data points and the hyperplane, pull normal data points away from the hyperplane, and minimize the number of abnormal data points; Among them, the calculation expression of the objective function is: ; In the formula, Represents maximizing the interval from normal data points to the hyperplane, v represents the preset scale factor, represents the slack variable, represents the minimum value of the objective function; After training is completed, the model will output the normal vector w and the bias term b, which will be used to predict new data points; The sub-region anomaly coefficient is obtained by calculating the ratio of the decision function value to the minimum value of the objective function. , determine whether the sub-region abnormal coefficient is greater than or equal to the preset threshold. If not, it means that the crop data of the corresponding sub-region is normal and is recorded as a normal sub-region. If so, it means that the crop data of the corresponding sub-region is abnormal and is recorded as an abnormal sub-region; It should be noted that the slack variables are used to control the proportion of abnormal points and to penalize misclassified data points. The slack variables allow some data points to deviate from the hyperplane, that is, some data points can be misclassified as abnormal to increase the flexibility of the model. The model can not only accurately classify most data points, but also tolerate the interference of some noise data, making the model more generalizable. During the optimization process, the objective function is minimized by adjusting the normal vector and bias term, as well as the slack variables, so as to find an optimal hyperplane that can distinguish normal data from abnormal data to the greatest extent.

[0019] In S3, the environmental status data in each sub-area is obtained, including air quality data (such as PM2.5, PM10) and water quality data (such as COD, ammonia nitrogen), and the environmental status data is analyzed to evaluate the degree of interference of the environmental status data on crops, including: Collect air quality data in real time through air particle monitors; The air quality data refers to the concentration of suspended particles in the air, including PM2.5 concentration and PM10 concentration; Collect water quality data in real time through COD meter and ammonia nitrogen meter, including COD concentration and ammonia nitrogen concentration; Preprocess the environmental status data, including processing missing values, detecting outliers, and data standardization; By calculating the interference factor in each sub-area, it is used to evaluate the interference degree of environmental status data on crops; The process of obtaining the interference factor is as follows: De-mean the collected air quality and water quality data; A zero-mean series is obtained by calculating the mean of the data and subtracting the mean from each data point; The air quality and water quality data after de-averaging are converted into frequency domain data through discrete Fourier transform; Through Fourier transform, we can identify periodic fluctuations in environmental state data and energy components of different frequencies; The calculation expression of the Fourier transform is: ; In the formula, Indicates frequency The corresponding Fourier coefficients are, represents the kth discrete frequency, k represents the number of frequencies, represents the data point at the mth moment, represents the data after removing the mean, m represents the number of acquisition time points, and M represents the maximum number of acquisition time points. represents the interval between two adjacent time points, and j represents an imaginary unit; Through the Fourier coefficients, we further calculate the power spectral density of the environmental data. The calculation expression is: ; In the formula, Indicates frequency The power density at By accumulating the power spectral density in the entire frequency range, the total fluctuation energy of the environmental data is obtained. ; By analyzing the power spectral density and total fluctuation energy, the interference factor is calculated through the frequency-weighted power spectrum. The calculation expression is: ; In the formula, represents the interference factor of the ith sub-region, E represents the maximum number of frequencies. It should be noted that: the power spectral density describes the energy distribution of the signal at each frequency, reflecting the periodicity and volatility of the environmental state over time; the interference factor is used to evaluate the impact of the environmental state in each sub-region on crop production; the higher the interference factor, the more significant the impact of the environmental state on crop production, which may lead to a decrease in crop yield or adverse impact on quality.

[0020] comparing the interference factor of each sub-region with a preset interference factor threshold; If the interference factor is greater than or equal to the preset interference factor threshold, it means that the environmental interference level of the corresponding sub-area is high, and it is recorded as a high interference level sub-area; If the interference factor is less than the preset interference factor threshold, it means that the environmental interference level of the corresponding sub-area is low, and it is recorded as a low interference level sub-area.

[0021] In S4, the abnormal degree of crop production in each sub-region and the degree of interference of the environmental status data of each sub-region on crops are comprehensively analyzed to determine whether the current sub-region is low-efficiency land, including: Get the sub-region anomaly coefficient within each sub-region and interference factors ; For each sub-region, the sub-region anomaly coefficient and interference factors Perform normalization and calculate the land use score in each sub-area; The land use score is used to determine whether each sub-region is low-efficiency land; Determine whether the land use score in each sub-region is greater than or equal to a preset threshold. If not, the current sub-region is low-efficiency land; The calculation expression of the land use fraction is: ; In the formula, represents the land use fraction in the i-th sub-region, and is the preset scale factor, and and All are greater than 0; It is determined whether the land use score in each sub-region is greater than or equal to a preset threshold. If not, the current sub-region is low-efficiency land.

[0022] In S5, based on the judgment result, the low-efficiency land sub-region is extracted and fed back to the manager to judge whether the land in the sub-region needs to be replanned, including: Feedback the identified inefficient land use sub-areas to land managers. The feedback of inefficient land use is not just a display of data. By generating a specific geographic information system (GIS) map, the location, size and influencing factors of inefficient land use are marked; This provides managers with a comprehensive and intuitive basis for decision-making, facilitating the precise deployment of land replanning, resource optimization, or environmental governance measures; Exemplary: For sub-areas with greater environmental disturbance, managers can propose measures to improve air or water quality, or adopt new crop planting strategies for areas with abnormal agricultural production; Based on the feedback, managers conduct a comprehensive assessment of the inefficient land sub-areas to determine whether these lands need to be replanned or other intervention measures are taken, including: Determine the degree of abnormality of crop yield deviation and output value in each sub-region; If the output deviation is too large or the output value is persistently low, it means that the agricultural production capacity of the region is seriously affected and intervention needs to be given priority; Determine the degree of interference of environmental conditions on agricultural production. If the air quality in some sub-areas is poor (e.g., high concentrations of PM2.5 and PM10) or the water quality is poor (e.g., high concentrations of COD and ammonia nitrogen), this may have a negative impact on crop growth and require priority intervention; Determine whether the current area is suitable for improvement, including: If the land area is small or restricted by geographical and climatic conditions, the possibility of replanning is low; As for the larger areas of inefficient land use, adjustments can be made by improving soil and water management, introducing suitable crop varieties, and increasing irrigation facilities; After the intervention measures are implemented, managers need to establish a long-term monitoring mechanism to regularly evaluate changes in low-efficiency land sub-areas to ensure the continued effectiveness of the intervention measures; Through continuous collection of environmental and crop data, management strategies can be dynamically adjusted and optimized, and necessary land replanning can be carried out based on the latest data to ensure efficient land use and sustainable development of agricultural production.

[0023] It should be noted that through land replanning, land resources can be optimized and waste caused by unreasonable use can be avoided, while creating better environmental conditions for agricultural production. This process not only helps to improve agricultural production efficiency, but also promotes the sustainable development of land resources, thereby achieving the goal of maximizing the benefits of land use.

[0024] See also Figure 2 As shown in the figure, the low-efficiency land identification system based on big data analysis includes: The target area determination module identifies the land cover types, including forest land, farmland land and urban built-up area land, based on satellite remote sensing image technology, and obtains farmland land as the target area for monitoring. A data collection module, the data collection module is used to obtain crop data in each sub-area, including: crop yield, output value and species; Obtain environmental status data in each sub-area, including: air quality data and water quality data; An abnormality evaluation module, which is used to analyze the crop data in each sub-region and evaluate the abnormality of crop production in each sub-region; An interference degree assessment module, wherein the interference degree assessment module is used to analyze the environmental status data and assess the interference degree of the environmental status data on the crops; A comprehensive analysis module, which is used to comprehensively analyze the abnormality of crop production in each sub-region and the interference degree of the environmental status data of each sub-region on the crops; An inefficient land identification module, wherein the inefficient land identification module identifies whether each sub-area is inefficient land according to the analysis result of the comprehensive analysis module; The low-efficiency land optimization module extracts the low-efficiency land sub-areas according to the judgment result, feeds back the low-efficiency land sub-areas to the manager, and judges whether the land in the sub-areas needs to be replanned.

[0025] The working principle of the present invention is as follows: the land cover type is identified by using satellite remote sensing imaging technology, and farmland is extracted as the monitoring area; the target area is divided into several sub-areas, and the crop data (such as yield, output value, species) and environmental data (such as air quality and water quality) of each sub-area are obtained and standardized; the abnormal degree of crop production is evaluated by using the One-ClassSVM model, and the interference factors of environmental data are analyzed by Fourier transform; the abnormal coefficient and interference factor of each sub-area are comprehensively analyzed, the land utilization score is calculated, the low-efficiency land sub-area is judged, and the analysis results are fed back to the manager to evaluate whether re-planning or intervention measures are needed; the system can achieve efficient utilization of land resources and sustainable development of agricultural production through precise data analysis and model optimization, and promote the optimal allocation of land resources.

[0026] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0027] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented by software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website site, computer, server or data center to another website site, computer, server or data center by wired or wireless (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state hard disk.

[0028] It should be understood that the term "and / or" in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. A and B can be singular or plural. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship, but it may also indicate an "and / or" relationship. Please refer to the context for specific understanding.

[0029] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0030] The above is a detailed description of an embodiment of the present invention, but the content is only a preferred embodiment of the present invention and cannot be considered to limit the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. A method for identifying inefficient land use based on big data analysis, characterized in that: The following steps are involved: S1: Identify land cover types based on satellite remote sensing image technology, including forest land, farmland land and urban built-up area land, and obtain farmland land as the target area for monitoring; S2: Divide the target area into several sub-areas, obtain the crop data in each sub-area, including: crop yield, output value and species, analyze the crop data in each sub-area, and evaluate the abnormality of crop production in each sub-area; S3: Obtain environmental status data in each sub-area, including air quality data and water quality data, analyze the environmental status data, and evaluate the degree of interference of the environmental status data on crops; S4: Comprehensively analyze the abnormal degree of crop production in each sub-region and the degree of interference of the environmental status data of each sub-region on crops to determine whether the current sub-region is low-efficiency land; S5: Based on the judgment result, extract the sub-areas of inefficient land use, and feed back the sub-areas of inefficient land use to the manager to determine whether the land in the sub-areas needs to be replanned.

2. The method for identifying inefficient land use based on big data analysis according to claim 1 is characterized in that: The analysis of the crop data in each sub-region to evaluate the abnormality of crop production in each sub-region specifically includes: Get the crop yield, crop output value and crop species in each sub-region; By constructing feature vectors for crop yield, crop output value and crop species in each sub-region, the feature vectors are used as inputs of the One-Class SVM model and the One-Class SVM model is trained. According to the trained One-Class SVM model, the value of the decision function and the minimum value of the objective function are output, and the sub-region abnormality coefficient is obtained by calculating the ratio of the value of the decision function to the minimum value of the objective function; Determine whether the sub-region abnormal coefficient is greater than or equal to the preset threshold. If not, record it as a normal sub-region; if so, record it as an abnormal sub-region.

3. The method for identifying inefficient land use based on big data analysis according to claim 2 is characterized in that: The process of obtaining the sub-region anomaly coefficient is as follows: Obtain the mean and standard deviation of crop yields in the target area, calculate the difference between the crop yield in each sub-area and the mean of the crop yield in the target area, and take the absolute value to obtain the yield deviation value; The yield deviation value is calculated by ratio with the standard deviation of the crop yield in the target area to obtain the standardized value of the crop yield in each sub-area; Obtain the mean and standard deviation of the crop output value in the target area, calculate the difference between the crop output value in each sub-area and the mean of the crop output value in the target area, and take the absolute value to obtain the output value deviation value; The output value deviation value is calculated by ratio with the standard deviation of the crop output value in the target area to obtain the standardized value of the crop output value in each sub-area; For discrete features, including crop species, one-hot encoding was used to convert each species into a binary vector; Construct the feature matrix, , Represents the feature vector of the i-th sub-region; Wherein, i represents the number of sub-regions, i=1, 2, ..., n, and n represents the maximum number of sub-regions; The feature vector Input into the One-Class SVM model for training; The goal of the model is to distinguish normal data from abnormal data by optimizing a hyperplane, which is a decision function constructed by the normal vector w and the bias term b; The calculation expression of the decision function is: ； In the formula, represents the decision function, x represents the input feature vector, represents the transpose of the normal vector w, and b represents the bias term; In the optimization process, the objective function is used to maximize the interval between normal data points and the hyperplane, pull normal data points away from the hyperplane, and minimize the number of abnormal data points; Among them, the calculation expression of the objective function is: ； In the formula, Represents maximizing the interval from normal data points to the hyperplane, v represents the preset scale factor, represents the slack variable, represents the minimum value of the objective function; After training is completed, the model will output the normal vector w and the bias term b, which will be used to predict new data points; The sub-region anomaly coefficient is obtained by calculating the ratio of the decision function value to the minimum value of the objective function. .

4. The method for identifying inefficient land use based on big data analysis according to claim 1, characterized in that: The analysis of the environmental status data to evaluate the interference degree of the environmental status data on crops specifically includes: Obtain air quality and water quality data in each sub-region; Process the air quality and water quality data in each sub-area, and calculate the interference factor in each sub-area based on the air quality and water quality data in each sub-area; It is determined whether the interference factor in each sub-region is greater than or equal to a preset interference factor threshold. If so, it is recorded as a high interference level sub-region; if not, it is recorded as a low interference level sub-region.

5. The method for identifying inefficient land use based on big data analysis according to claim 1, characterized in that: The process of obtaining the interference factor is as follows: De-mean the collected air quality and water quality data; A zero-mean series is obtained by calculating the mean of the data and subtracting the mean from each data point; The air quality and water quality data after de-averaging are converted into frequency domain data through discrete Fourier transform; Through Fourier transform, periodic fluctuations and energy components of different frequencies in environmental state data are identified; The calculation expression of the Fourier transform is: ； In the formula, Indicates frequency The corresponding Fourier coefficients are, represents the kth discrete frequency, k represents the number of frequencies, represents the data point at the mth moment, represents the data after removing the mean, m represents the number of acquisition time points, and M represents the maximum number of acquisition time points. represents the interval between two adjacent time points, and j represents an imaginary unit; Through the Fourier coefficients, we further calculate the power spectral density of the environmental data. The calculation expression is: ; In the formula, Indicates frequency The power density at By accumulating the power spectral density in the entire frequency range, the total fluctuation energy of the environmental data is obtained. ; By analyzing the power spectral density and total fluctuation energy, the interference factor is calculated through the frequency-weighted power spectrum. The calculation expression is: ； In the formula, represents the interference factor of the ith sub-region, and E represents the maximum number of frequencies.

6. The method for identifying inefficient land use based on big data analysis according to claim 1, characterized in that: The determining whether the current sub-region is low-efficiency land specifically includes: Get the sub-region anomaly coefficient within each sub-region and interference factors ; For each sub-region, the sub-region anomaly coefficient and interference factors Perform normalization and calculate the land use score in each sub-area; The land use score is used to determine whether each sub-region is low-efficiency land; It is determined whether the land use score in each sub-region is greater than or equal to a preset threshold. If not, the current sub-region is low-efficiency land.

7. The method for identifying inefficient land use based on big data analysis according to claim 1, characterized in that: The determination of whether the land in the sub-region needs to be replanned specifically includes: Feedback the identified inefficient land use sub-areas to land managers by generating detailed GIS maps to indicate the location, size and influencing factors of inefficient land use; For sub-areas with high interference levels, measures are taken to reduce PM2.5 concentrations, PM10 concentrations, COD concentrations and ammonia nitrogen concentrations, and the interference factors and land use scores are recalculated. If the land use score is less than the preset threshold, the current area will be replanned.

8. The inefficient land identification system based on big data analysis is characterized by: The method for identifying inefficient land use based on big data analysis as claimed in any one of claims 1 to 7 comprises: The target area determination module identifies the land cover types, including forest land, farmland land and urban built-up area land, based on satellite remote sensing image technology, and obtains farmland land as the target area for monitoring. A data collection module, the data collection module is used to obtain crop data in each sub-area, including: crop yield, output value and species; Obtain environmental status data in each sub-area, including: air quality data and water quality data; An abnormality evaluation module, which is used to analyze the crop data in each sub-region and evaluate the abnormality of crop production in each sub-region; An interference degree assessment module, wherein the interference degree assessment module is used to analyze the environmental status data and assess the interference degree of the environmental status data on the crops; A comprehensive analysis module, which is used to comprehensively analyze the abnormality of crop production in each sub-region and the interference degree of the environmental status data of each sub-region on the crops; An inefficient land identification module, wherein the inefficient land identification module identifies whether each sub-area is inefficient land according to the analysis result of the comprehensive analysis module; The low-efficiency land optimization module extracts the low-efficiency land sub-areas according to the judgment result, feeds back the low-efficiency land sub-areas to the manager, and judges whether the land in the sub-areas needs to be replanned.

Images