A method and system for extracting information from comprehensive survey of natural resources

Abstract

The present invention relates to a method and system for extracting information from a comprehensive survey of natural resources, and to the field of computer technology. The method comprises: obtaining surface information through multi-source remote sensing images, collecting environmental data in real time through a ground sensor network, and obtaining multi-source data; performing spatial alignment and standardization processing on the multi-source data to obtain processed multi-source data; performing data fusion on the processed multi-source data to obtain corresponding fused data; performing extraction and analysis on the fused data to obtain corresponding extracted information; performing data update on the extracted information through a recursive filtering algorithm to obtain an updated state estimate corresponding to the extracted information; adjusting the existing information extraction strategy according to the updated state estimate to obtain an updated information extraction strategy, and performing subsequent information extraction according to the updated information extraction strategy. The present invention can realize a more efficient and accurate operation of extracting information from a comprehensive survey of natural resources.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a method, system, electronic device and non-transitory computer-readable storage medium for extracting information from a comprehensive survey of natural resources. Background Art

[0002] Nowadays, information extraction from comprehensive natural resource surveys mainly relies on a combination of traditional manual surveys and remote sensing technology. Manual surveys usually include field sampling, measurement and recording, which can provide highly accurate and detailed data, but are time-consuming and labor-intensive and have limited coverage. Remote sensing technology uses satellite images, aerial photography, drones and other equipment to obtain surface information from the air over a large area. These image data can be quickly generated into resource distribution maps over a large area through image interpretation, classification and analysis.

[0003] However, manual surveys consume a lot of manpower and material resources, especially in areas with complex terrain or harsh climate, where the operation is difficult and dangerous. Secondly, remote sensing technology is limited by image resolution, lighting conditions and weather conditions, which may lead to incomplete or inaccurate data. In addition, image data processing and interpretation require a high level of professional skills, and the integration and analysis of different data sources are challenging. Real-time updating and dynamic monitoring of data are also difficult, and cannot reflect the rapid changes in natural resources in a timely manner.

[0004] Therefore, there is an urgent need to explore more efficient and accurate methods for extracting information from comprehensive natural resource surveys to improve the comprehensiveness, timeliness and accuracy of data acquisition. Summary of the invention

[0005] In response to the technical problems existing in the prior art, the present invention provides a natural resource comprehensive survey information extraction method, system, electronic device and non-transitory computer-readable storage medium that can realize more efficient and more accurate natural resource comprehensive survey information extraction operations.

[0006] The technical solution of the present invention to solve the above technical problems is as follows:

[0007] The present invention provides a method for extracting information from a comprehensive survey of natural resources, the method comprising:

[0008] Acquire surface information through multi-source remote sensing images, collect environmental data in real time through ground sensor networks, and obtain multi-source data;

[0009] Performing spatial alignment and standardization processing on the multi-source data to obtain processed multi-source data;

[0010] Performing data fusion on the processed multi-source data to obtain corresponding fused data;

[0011] Performing extraction analysis on the fused data to obtain corresponding extraction information;

[0012] updating the extracted information by a recursive filtering algorithm to obtain an updated state estimate corresponding to the extracted information;

[0013] According to the updated state estimation, the existing information extraction strategy is adjusted to obtain an updated information extraction strategy, and subsequent information extraction is performed according to the updated information extraction strategy.

[0014] Optionally, performing data fusion on the processed multi-source data to obtain corresponding fused data includes:

[0015] Get the original data value of each data source at the preset position;

[0016] Obtaining the confidence level of each of the data sources at the preset location;

[0017] Obtaining a time function of how each of the data sources changes over time;

[0018] Determining the fused data according to the original data value, the confidence level and the time function;

[0019] Optionally, the fused data is represented as:

[0020] ;

[0021] in, is the fused data, specifically the fused data value, is the original data value of the ith data source at position (x,y), is the weight coefficient of the i-th data source, is the confidence of the ith data source at position (x,y), is the time integration coefficient of data source i, is the time function of data source i, is the normalization coefficient, is a function of time The weight of is the jth time function, a, b, c, d are the upper and lower limits of the integral, N is the number of data sources, and M is the number of time functions.

[0022] Optionally, the extracting and analyzing the fused data to obtain corresponding extracted information includes:

[0023] Inputting the fused data into a deep learning model for linear processing to obtain a corresponding linear transformation result;

[0024] Obtaining an auxiliary function for representing a nonlinear transformation of the fused data;

[0025] The linear transformation result and the auxiliary function are processed by the output layer activation function of the deep learning model to obtain corresponding extraction information.

[0026] Optionally, the extracted information is expressed as:

[0027] ;

[0028] Among them, X is the input fusion data, and are the weight matrices of the first and second layers respectively, and are the biases of the first and second layers, respectively. (x)=max(0,x) is the activation function, is the activation function of the output layer, For auxiliary functions The coefficient of is the auxiliary function, and P is the number of auxiliary functions.

[0029] Optionally, updating the extracted information by a recursive filtering algorithm to obtain an updated state estimate corresponding to the extracted information includes:

[0030] According to the extracted information, obtaining an observation value corresponding to the current moment and a predicted state estimate of the resource state in the extracted information before the current moment;

[0031] Obtaining a nonlinear compensation function of the observation value;

[0032] An updated state estimate corresponding to the extracted information is obtained based on the observed value, the predicted state estimate and the nonlinear compensation function.

[0033] Optionally, the updated state estimate is expressed as:

[0034] ;

[0035] in, is the updated state estimate, is the predicted state estimate, is the Kalman gain, is the observation value at the current moment, H is the observation matrix, is the nonlinear compensation function The coefficient of is the nonlinear compensation function, and Q is the number of nonlinear compensation functions.

[0036] Optionally, adjusting the existing information extraction strategy according to the updated state estimation to obtain an updated information extraction strategy includes:

[0037] analyzing the changing trends and dynamic characteristics of the current natural resource status based on the updated status estimate;

[0038] Based on the change trends and dynamic characteristics, identify areas of significant change in the areas where natural resources are located;

[0039] The emphasis and frequency of data acquisition are adjusted according to the significant change areas.

[0040] Optionally, the multi-source remote sensing images include satellite images, aerial photography and drone images, and the data collected by the ground sensor network include temperature, humidity and soil moisture.

[0041] The present invention also provides a natural resources comprehensive survey information extraction system, the system comprising:

[0042] The data acquisition module is used to obtain surface information through multi-source remote sensing images and collect environmental data in real time through a ground sensor network to obtain multi-source data;

[0043] A data processing module, used for performing spatial alignment and standardization processing on the multi-source data to obtain processed multi-source data;

[0044] A data fusion module is used to perform data fusion on the processed multi-source data to obtain corresponding fused data;

[0045] A first extraction module, used to extract and analyze the fused data to obtain corresponding extraction information;

[0046] A state updating module, used to update the extracted information through a recursive filtering algorithm to obtain an updated state estimate corresponding to the extracted information;

[0047] The second extraction module is used to adjust the existing information extraction strategy according to the updated state estimation to obtain an updated information extraction strategy, and perform subsequent information extraction according to the updated information extraction strategy.

[0048] In addition, to achieve the above-mentioned purpose, the present invention also proposes an electronic device, comprising: a memory for storing computer software programs; a processor for reading and executing the computer software programs, thereby implementing a method for extracting information from a comprehensive survey of natural resources as described above.

[0049] In addition, to achieve the above-mentioned purpose, the present invention also proposes a non-transitory computer-readable storage medium, in which a computer software program is stored. When the computer software program is executed by a processor, it implements a method for extracting comprehensive natural resource survey information as described above.

[0050] The beneficial effects of the present invention are:

[0051] (1) The present invention combines multi-source remote sensing technology with ground sensor networks to quickly obtain surface information and real-time environmental data over a wide area, significantly improving the efficiency of data acquisition;

[0052] (2) The present invention adopts an improved weighted fusion algorithm to efficiently integrate multi-source data, comprehensively utilize the advantages of different data sources, and ensure the accuracy and comprehensiveness of the data after fusion;

[0053] (3) The present invention introduces a deep learning model to efficiently analyze and interpret the integrated information map, which can automatically extract and identify natural resource information, greatly improving the automation level and accuracy of information extraction;

[0054] (4) The present invention realizes dynamic monitoring of natural resources and real-time information updating through recursive filtering algorithm, which can timely reflect changes in resource status and provide accurate and timely data support for resource management;

[0055] (5) The present invention reduces the reliance on traditional manual surveys, reduces labor costs and operational risks, and significantly improves the safety and operability of survey work, especially in areas with complex terrain or harsh climates;

[0056] (6) The present invention uses visualization tools to display analysis and monitoring results, providing managers with intuitive and easy-to-understand decision support information, helping them to conduct scientific and reasonable resource management and decision-making, and improving the efficiency and effectiveness of resource management;

[0057] (7) The present invention has strong adaptability and expansibility, and can be adjusted and optimized according to different resource types and regional characteristics to meet the needs of comprehensive surveys of various natural resources.

[0058] Through multi-source data fusion and artificial intelligence technology, the present invention significantly improves the information acquisition efficiency, accuracy and dynamic monitoring capabilities of comprehensive natural resource surveys, provides strong support for scientific decision-making and resource management, and has broad application prospects and significant economic and social benefits. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 A scene diagram of a natural resources comprehensive survey information extraction method provided by the present invention;

[0060] Figure 2 A flowchart of a method for extracting information from a comprehensive survey of natural resources provided by the present invention;

[0061] Figure 3 A schematic diagram of the structure of a natural resources comprehensive survey information extraction system provided by the present invention;

[0062] Figure 4 A schematic diagram of the hardware structure of a possible electronic device provided by the present invention;

[0063] Figure 5 A schematic diagram of the hardware structure of a possible computer-readable storage medium provided by the present invention. DETAILED DESCRIPTION

[0064] 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 those skilled in the art without creative work are within the scope of protection of the present invention.

[0065] In the description of the present invention, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the present invention, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.

[0066] In the description of the present invention, the term "for example" is used to mean "used as an example, illustration or explanation". Any embodiment described as "for example" in the present invention is not necessarily to be construed as being more preferred or advantageous than other embodiments. The following description is given to enable any technician in the field to implement and use the present invention. In the following description, details are listed for the purpose of explanation. It should be understood that a person of ordinary skill in the art can recognize that the present invention can be implemented without using these specific details. In other examples, well-known structures and processes will not be elaborated in detail to avoid obscuring the description of the present invention with unnecessary details. Therefore, the present invention is not intended to be limited to the embodiments shown, but is consistent with the widest scope consistent with the principles and features disclosed in the present invention.

[0067] See also Figure 1 , Figure 1 A scene diagram of a natural resources comprehensive survey information extraction method provided by the present invention. Figure 1As shown, the terminal and the server are connected via a network, such as a wired or wireless network connection. The terminal may include but is not limited to portable terminals such as mobile phones and tablets installed with various network platform applications, as well as fixed terminals such as computers, query machines, and advertising machines. The server provides users with various business services, including service push servers, user recommendation servers, etc.

[0068] It should be noted that Figure 1 The scenario diagram of a method for extracting information from a comprehensive survey of natural resources shown is merely an example. The terminal, server, and application scenario described in the embodiment of the present invention are intended to more clearly illustrate the technical solution of the embodiment of the present invention, and do not constitute a limitation on the technical solution provided by the embodiment of the present invention. A person of ordinary skill in the art will appreciate that, with the evolution of the system and the emergence of new business scenarios, the technical solution provided by the embodiment of the present invention is equally applicable to similar technical problems.

[0069] Among them, the terminal can be used for:

[0070] Acquire surface information through multi-source remote sensing images, collect environmental data in real time through ground sensor networks, and obtain multi-source data;

[0071] Performing spatial alignment and standardization processing on the multi-source data to obtain processed multi-source data;

[0072] Performing data fusion on the processed multi-source data to obtain corresponding fused data;

[0073] Performing extraction analysis on the fused data to obtain corresponding extraction information;

[0074] updating the extracted information by a recursive filtering algorithm to obtain an updated state estimate corresponding to the extracted information;

[0075] According to the updated state estimation, the existing information extraction strategy is adjusted to obtain an updated information extraction strategy, and subsequent information extraction is performed according to the updated information extraction strategy.

[0076] See also Figure 2 , provides a flowchart of a natural resources comprehensive survey information extraction method of the present invention, comprising the following steps:

[0077] Step 201: Acquire surface information through multi-source remote sensing images, collect environmental data in real time through a ground sensor network, and obtain multi-source data.

[0078] In some embodiments, the multi-source remote sensing images may include satellite images, aerial photography, and drone images, and the data collected by the ground sensor network may include temperature, humidity, and soil moisture. In some embodiments, the collected environmental data may also include: vegetation coverage, land use type, water distribution, topography, and geomorphic features.

[0079] Step 202: perform spatial alignment and standardization processing on the multi-source data to obtain processed multi-source data.

[0080] Spatial alignment refers to spatially aligning data from different sensors and sources to ensure that they can be compared and analyzed in the same geographic coordinate system. This involves the following steps:

[0081] Coordinate conversion: Convert geographic coordinates from different data sources to a unified coordinate system. For example, convert the coordinate system of satellite images, drone images, and ground sensor data to the same geographic coordinate system, such as WGS84.

[0082] Geometric correction: Perform geometric correction on image data to correct geometric distortion caused by shooting angle, terrain undulation, etc. This ensures that the location of objects in the image data is consistent with the actual geographical location.

[0083] Registration: spatially register different images and sensor data so that they are aligned at the same geographic location. This is usually done by selecting geographic reference points (such as road intersections, rivers, and other significant features).

[0084] Standardization refers to the unification of the scale and format of multi-source data so that they can be analyzed under a unified framework. This involves the following steps:

[0085] Unit unification: Unify the units of different data sources. For example, convert temperature data to degrees Celsius and humidity data to percentage.

[0086] Scale normalization: Normalize the data ranges of different data sources, such as normalizing the data to between 0 and 1. This helps eliminate the dimensional differences between different data sources and makes the data comparable during analysis.

[0087] Data format conversion: Convert the data formats of different data sources to conform to a unified data format standard, such as unifying image data into GeoTIFF format and sensor data into CSV format.

[0088] Assume that there are the following multi-source data: Satellite imagery: provides large-scale surface coverage information. Drone imagery: provides small-scale high-resolution surface details. Ground sensor data: provides point-distributed temperature, humidity, and soil moisture data. In some embodiments, the coordinates of satellite images and drone images can be converted to a unified WGS84 coordinate system. The coordinates of the ground sensor data are also converted to the WGS84 coordinate system. Perform geometric correction on the satellite imagery and drone imagery to ensure that the location of the objects in the imagery is accurate. Select several geographic reference points for registration, such as road intersections, buildings, etc., to align the satellite imagery and drone imagery in space.

[0089] During the standardization process, the temperature data is first converted to degrees Celsius and the humidity data is converted to percentage. Then all data are normalized to between 0 and 1, for example, the temperature data is normalized by range and the humidity data is normalized by range. Finally, the image data is saved in GeoTIFF format and the sensor data is saved in CSV format.

[0090] Through the above spatial alignment and standardization processing, the present invention obtains processed multi-source data, which are spatially aligned, unit-unified, scale-normalized, and conform to a unified data format standard. The processed multi-source data can be further fused and analyzed to provide more accurate and comprehensive natural resource information.

[0091] Step 203: perform data fusion on the processed multi-source data to obtain corresponding fused data.

[0092] In some embodiments, step 203 may include:

[0093] Get the original data value of each data source at the preset position;

[0094] Obtaining the confidence level of each of the data sources at the preset location;

[0095] Obtaining a time function of how each of the data sources changes over time;

[0096] Determining the fused data according to the original data value, the confidence level and the time function;

[0097] In some embodiments, the fused data is represented as:

[0098] ;

[0099] in, is the fused data, specifically the fused data value, is the original data value of the ith data source at position (x,y), is the weight coefficient of the i-th data source, is the confidence of the ith data source at position (x,y), is the time integration coefficient of data source i, is the time function of data source i, is the normalization coefficient, is a function of time The weight of is the jth time function, a, b, c, d are the upper and lower limits of the integral, N is the number of data sources, and M is the number of time functions.

[0100] In the specific implementation, It is the data value after fusion, which is the comprehensive data we ultimately hope to obtain;

[0101] is the original data value of the ith data source at the (x, y) position, indicating the observation value of different data sources at the same position; is the weight coefficient of the i-th data source, which is used to indicate the importance of the data source in the fusion process. Different data sources are assigned different weights according to their reliability and accuracy; is the confidence of the ith data source at the (x, y) position, indicating the credibility of the data at a specific position. The higher the confidence, the greater the impact of the data source on the final result. is the time integration coefficient of data source i, which is used to consider the cumulative changes of the data source within a certain time range and represents the time importance of data source i; It is the time function of data source i, which indicates the change of data source i over time t. By integrating this function, we can get the cumulative impact of the data source within the time range. is the normalization coefficient, which is used to normalize the denominator so that the value of the denominator is suitable for properly balancing and standardizing the terms in the numerator; is a function of time The weights are used to indicate the importance of different time functions in the data fusion process; is the jth time function, which represents another set of time-varying characteristics. The integration of this function can reflect the impact of different time periods on the fusion results. a, b, c, d are the upper and lower limits of the integration, and a and b are applicable to The integral of The time range of the integration, c and d apply to The integral of The time range of the integration; N is the number of data sources, indicating how many different data sources are involved in the fusion; M is the number of time functions, indicating how many different time functions are involved in the fusion.

[0102] In the specific implementation, represents the weighted sum of each data source in space and time. First, the spatial data of each data source is According to its weight and confidence , so that the impact of spatial data and temporal changes on the final results can be comprehensively considered.

[0103] The denominator of the formula The weight and confidence are normalized. First, the weight of each data source is and confidence Sum, then add another set of time functions The weighted integral value of can ensure the balance and stability of the whole formula.

[0104] Finally, the fused data value can be obtained through the ratio of the numerator and the denominator This value comprehensively considers the spatial and temporal information of different data sources and can more accurately reflect the actual situation.

[0105] Step 204: extract and analyze the fused data to obtain corresponding extraction information.

[0106] In some embodiments, step 204 may include:

[0107] Inputting the fused data into a deep learning model for linear processing to obtain a corresponding linear transformation result;

[0108] Obtaining an auxiliary function for representing a nonlinear transformation of the fused data;

[0109] The linear transformation result and the auxiliary function are processed by the output layer activation function of the deep learning model to obtain corresponding extraction information.

[0110] In some embodiments, the extracted information is represented as:

[0111] ;

[0112] Among them, X is the input fusion data, and are the weight matrices of the first and second layers respectively, and are the biases of the first and second layers, respectively. (x)=max(0,x) is the activation function, is the activation function of the output layer, For auxiliary functions The coefficient of is the auxiliary function, and P is the number of auxiliary functions.

[0113] In the specific implementation, X is the input fusion data, which is the comprehensive data after the fusion of multi-source data in the previous step, and is used as the input of the deep learning model; and The weight matrices of the first and second layers respectively contain the trainable parameters of the model and are used to perform linear transformation on the input data; and The biases of the first and second layers are respectively. These biases are also trainable parameters of the model and are used to adjust the results after linear transformation; (x)=max(0,x) is the activation function. The ReLU function changes the input value less than 0 to 0 and keeps the input value greater than 0 unchanged, which helps the model learn complex nonlinear relationships. The activation function of the output layer (such as Softmax) is used to convert the output of the model into a probability distribution or other desired output form; For auxiliary functions The coefficients of are trainable parameters used to weight the outputs of different auxiliary functions; is an auxiliary function, which represents the nonlinear transformation of the input data. The auxiliary function is used to introduce additional nonlinear features to enhance the expressive power of the model. P is the number of auxiliary functions, which indicates how many auxiliary functions are introduced to perform different forms of nonlinear transformation on the input data.

[0114] In the specific implementation, the input fusion data X undergoes a linear transformation in the first layer. , and get the intermediate result. The result is transformed nonlinearly through the ReLU activation function, and the formula is , the introduction of the ReLU function helps the model capture the nonlinear characteristics of the data. Then, the result after ReLU activation is input to the second layer for linear transformation At the same time, introduce P auxiliary functions , perform different forms of nonlinear transformations on the input data X. The output of each auxiliary function is passed through the coefficient The weighted formula is Finally, the linear transformation result of the second layer is added to the weighted sum of the auxiliary function and then passed through the activation function of the output layer. Processing to get the final output .

[0115] The overall expression of this formula is a two-layer neural network model, in which each layer processes the input data through linear transformation and nonlinear activation function. At the same time, the auxiliary function is introduced and its coefficient It is used to enhance the nonlinear expression ability of the model so that the model can better capture complex features. This design helps to improve the accuracy and reliability of information extraction.

[0116] Step 205: Update the extracted information using a recursive filtering algorithm to obtain an updated state estimate corresponding to the extracted information.

[0117] In some embodiments, step 205 may include:

[0118] According to the extracted information, obtaining an observation value corresponding to the current moment and a predicted state estimate of the resource state in the extracted information before the current moment;

[0119] Obtaining a nonlinear compensation function of the observation value;

[0120] An updated state estimate corresponding to the extracted information is obtained based on the observed value, the predicted state estimate and the nonlinear compensation function.

[0121] In some embodiments, the updated state estimate is expressed as:

[0122] ;

[0123] in, is the updated state estimate, is the predicted state estimate, is the Kalman gain, is the observation value at the current moment, H is the observation matrix, is the nonlinear compensation function The coefficient of is the nonlinear compensation function, and Q is the number of nonlinear compensation functions.

[0124] In the specific implementation, is the updated state estimate, which is the state estimate obtained by correcting the current observation value and the previous prediction value at time k; is the predicted state estimate. Before time k (i.e., time ) The predicted value of the state; is the Kalman gain, which is used to adjust the error between prediction and observation. The gain coefficient is calculated based on the error covariance matrix and the observation noise matrix; is the observed value at the current moment, which represents the actual observed data of the state at time k; H is the observation matrix, which is used to map the state to the observation space and defines the relationship between the observed value and the state; is the nonlinear compensation function The coefficients are used to weight the outputs of different nonlinear compensation functions; is a nonlinear compensation function, which represents the nonlinear transformation of the observed value and is used to introduce nonlinear compensation for the observed value to enhance the accuracy of state estimation. Q is the number of nonlinear compensation functions, which indicates how many different nonlinear compensation functions are introduced to transform the observed value.

[0125] In the specific implementation, the predicted state estimate at the current time k is obtained through the system's dynamic model and the previous state estimate , which is an estimate of the current state based on prior information and the system dynamic equations.

[0126] In some embodiments, the state estimation of the present invention can be the forest coverage rate and water resource distribution of a certain area at a certain time, such as water level, water quality and water quantity. It can also be the land use type distribution of a certain area at a certain time, such as farmland, city, wetland, etc., the air quality index (AQI) of a certain area at a certain time, including pollutant concentrations such as PM2.5, PM10, NO2, etc., the biodiversity index of a certain area at a certain time, including species richness and uniformity, and the glacier and snow cover area of ​​a certain area at a certain time.

[0127] Kalman Gain The calculation is based on the relationship between the prediction error and the observation error, adjusting the difference between the predicted state estimate and the actual observed value. Part of the observations With predicted observations The error between them is amplified or reduced and then corrected to the predicted state estimate.

[0128] In order to further improve the accuracy of state estimation, Q nonlinear compensation functions are introduced These functions apply to the observed values Perform different forms of nonlinear transformations through coefficients After weighting, it is added to the state estimation, and the formula is .

[0129] The final updated state estimate is estimated by predicting the state Add the Kalman gain correction and nonlinear compensation term As a result, not only the linear error correction is taken into account, but also the adaptability of the model to complex and nonlinear phenomena is enhanced through nonlinear compensation.

[0130] The role of the present invention in multi-source data fusion and dynamic monitoring is to estimate the state more accurately by combining linear and nonlinear methods. The Kalman filter provides the optimal estimate of the linear system, and the introduced nonlinear compensation function further enhances the model's ability to handle complex and nonlinear data, making the state estimation more accurate and reliable. This method is particularly useful in comprehensive surveys of natural resources, and can dynamically adjust and optimize information extraction strategies to improve the comprehensiveness, timeliness and accuracy of data acquisition.

[0131] Step 206: According to the updated state estimation, the existing information extraction strategy is adjusted to obtain an updated information extraction strategy, and subsequent information extraction is performed according to the updated information extraction strategy.

[0132] In some embodiments, step 206 may include:

[0133] analyzing the changing trends and dynamic characteristics of the current natural resource status based on the updated status estimate;

[0134] Based on the change trends and dynamic characteristics, identify areas of significant change in the areas where natural resources are located;

[0135] The emphasis and frequency of data acquisition are adjusted according to the significant change areas.

[0136] In some embodiments, the updated state estimate is the best estimate of the natural resource status at the current moment obtained by multi-source data fusion, spatial alignment and standardization, combined with Kalman filter or other estimation methods.

[0137] The analysis of changing trends and dynamic characteristics mainly includes:

[0138] Time series analysis: Analyze the estimated values ​​of the state over a period of time to identify the changing trend of the natural resource status. For example, by analyzing the time series data of forest cover, determine whether the forest area is increasing or decreasing.

[0139] Spatial analysis: Combine spatial distribution information to identify the dynamic characteristics of natural resources in different regions. For example, by analyzing water resource distribution data, we can identify which areas have large water level fluctuations.

[0140] Specific steps:

[0141] Data collection and organization: Collect state estimation data over a period of time to form a time series.

[0142] Trend analysis: Use statistical analysis methods, such as moving average and regression analysis, to identify long-term trends in natural resource conditions.

[0143] Dynamic characteristic analysis: Use change detection algorithms, such as differential analysis and change vector analysis, to identify the dynamic change characteristics of natural resources.

[0144] In some embodiments, a significant change area refers to an area identified in the analysis where the natural resource conditions have changed significantly, which may be due to natural causes (such as climate change, natural disasters) or human causes (such as land use changes, resource development).

[0145] Steps to identify areas of significant change:

[0146] Change detection: Use change detection algorithms (such as thresholding, classification, change vector analysis, etc.) to identify areas with significant changes in the time series.

[0147] Significance judgment: Determine which changes are significant based on the preset change threshold or statistical significance level.

[0148] Region marking: Spatial marking of areas with significant changes to generate a distribution map of the changed areas.

[0149] In some embodiments, the focus and frequency of data acquisition are adjusted to ensure more intensive and detailed monitoring of areas with significant changes, thereby capturing the dynamics of changes in a timely manner and improving the efficiency and effectiveness of resource management.

[0150] Specific adjustment steps:

[0151] Determination of key areas: Based on the distribution map of areas with significant changes, determine the areas that require key monitoring.

[0152] Adjustment of data acquisition frequency: Increase the frequency of data acquisition for areas with significant changes. For example, an area that originally acquired data once a month can be adjusted to acquire data once a week.

[0153] Resource allocation optimization: Optimize the allocation of sensor, drone and satellite resources according to key monitoring areas to ensure the quality and timeliness of monitoring data in key areas.

[0154] For example, taking forest coverage monitoring as an example, state estimation can estimate the forest coverage of each area at the current moment by integrating satellite images, drone images and ground sensor data. . Trend and dynamic characteristics analysis can compare forest cover estimates from the past few months to identify areas where forest cover is increasing or decreasing, and analyze their changing trends and dynamic characteristics. Identification of areas with significant changes can use change detection algorithms to identify areas where forest cover has changed significantly and mark them spatially. Data acquisition adjustments can increase the frequency of monitoring areas with significant changes in forest cover, for example, from once a month to once a week, while optimizing drone flight routes and satellite image acquisition plans to ensure that data updates in these areas are more timely and accurate.

[0155] The updated state estimation can dynamically analyze the changing trends and characteristics of natural resource conditions, identify areas of significant changes, and adjust the focus and frequency of data acquisition accordingly. This process not only improves the timeliness and accuracy of data, but also enhances the monitoring and management capabilities of natural resource changes, ensuring effective resource protection and rational use in key areas.

[0156] See also Figure 3 , Figure 3 A schematic diagram of the structure of a natural resources comprehensive survey information extraction system provided by the present invention.

[0157] like Figure 3 As shown, a natural resources comprehensive survey information extraction system proposed in an embodiment of the present invention includes:

[0158] The data acquisition module 301 is used to acquire surface information through multi-source remote sensing images and collect environmental data in real time through a ground sensor network to obtain multi-source data;

[0159] The data processing module 302 is used to perform spatial alignment and standardization processing on the multi-source data to obtain processed multi-source data;

[0160] A data fusion module 303 is used to perform data fusion on the processed multi-source data to obtain corresponding fused data;

[0161] A first extraction module 304 is used to extract and analyze the fused data to obtain corresponding extraction information;

[0162] A state updating module 305 is used to update the extracted information through a recursive filtering algorithm to obtain an updated state estimate corresponding to the extracted information;

[0163] The second extraction module 306 is used to adjust the existing information extraction strategy according to the updated state estimation, obtain an updated information extraction strategy, and perform subsequent information extraction according to the updated information extraction strategy.

[0164] See also Figure 4 , Figure 4Schematic diagram of an electronic device provided by an embodiment of the present invention. Figure 4 As shown, an embodiment of the present invention provides an electronic device 400, including a memory 410, a processor 420, and a computer program 411 stored in the memory 410 and executable on the processor 420. When the processor 420 executes the computer program 411, the following steps are implemented:

[0165] Acquire surface information through multi-source remote sensing images, collect environmental data in real time through ground sensor networks, and obtain multi-source data;

[0166] Performing spatial alignment and standardization processing on the multi-source data to obtain processed multi-source data;

[0167] Performing data fusion on the processed multi-source data to obtain corresponding fused data;

[0168] Performing extraction analysis on the fused data to obtain corresponding extraction information;

[0169] updating the extracted information by a recursive filtering algorithm to obtain an updated state estimate corresponding to the extracted information;

[0170] According to the updated state estimation, the existing information extraction strategy is adjusted to obtain an updated information extraction strategy, and subsequent information extraction is performed according to the updated information extraction strategy.

[0171] See also Figure 5 , Figure 5 A schematic diagram of an embodiment of a computer-readable storage medium provided in an embodiment of the present invention. Figure 5 As shown, this embodiment provides a computer-readable storage medium 500, on which a computer program 411 is stored. When the computer program 411 is executed by a processor, the following steps are implemented:

[0172] Acquire surface information through multi-source remote sensing images, collect environmental data in real time through ground sensor networks, and obtain multi-source data;

[0173] Performing spatial alignment and standardization processing on the multi-source data to obtain processed multi-source data;

[0174] Performing data fusion on the processed multi-source data to obtain corresponding fused data;

[0175] Performing extraction analysis on the fused data to obtain corresponding extraction information;

[0176] updating the extracted information by a recursive filtering algorithm to obtain an updated state estimate corresponding to the extracted information;

[0177] According to the updated state estimation, the existing information extraction strategy is adjusted to obtain an updated information extraction strategy, and subsequent information extraction is performed according to the updated information extraction strategy.

[0178] It should be noted that in the above embodiments, the description of each embodiment has its own emphasis, and for parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0179] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0180] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded computer, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A system that specifies the functions of a box or boxes.

[0181] These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory produce a product including an instruction system, which is implemented in the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0182] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0183] Although the preferred embodiments of the present invention have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0184] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A method for extracting information from a comprehensive survey of natural resources, characterized in that: The method comprises: Acquire surface information through multi-source remote sensing images, collect environmental data in real time through ground sensor networks, and obtain multi-source data; Performing spatial alignment and standardization processing on the multi-source data to obtain processed multi-source data; Performing data fusion on the processed multi-source data to obtain corresponding fused data; Performing extraction analysis on the fused data to obtain corresponding extraction information; updating the extracted information by a recursive filtering algorithm to obtain an updated state estimate corresponding to the extracted information; According to the updated state estimation, adjusting the existing information extraction strategy to obtain an updated information extraction strategy, and performing subsequent information extraction according to the updated information extraction strategy; The fused data is represented as: in, is the fused data, specifically the fused data value, is the original data value of the ith data source at position (x,y), is the weight coefficient of the i-th data source, is the confidence of the ith data source at position (x,y), is the time integration coefficient of data source i, is the time function of data source i, is the normalization coefficient, is a function of time The weight of is the jth time function, a, b, c, d are the upper and lower limits of the integral, N is the number of data sources, and M is the number of time functions; The extracting and analyzing the fused data to obtain corresponding extracted information includes: Inputting the fused data into a deep learning model for linear processing to obtain a corresponding linear transformation result; Obtaining an auxiliary function for representing a nonlinear transformation of the fused data; The linear transformation result and the auxiliary function are processed by the output layer activation function of the deep learning model to obtain corresponding extraction information.

2. The method for extracting information from a comprehensive survey of natural resources according to claim 1, characterized in that: The step of fusing the processed multi-source data to obtain corresponding fused data includes: Get the original data value of each data source at the preset position; Obtaining the confidence level of each of the data sources at the preset location; Obtaining a time function of how each of the data sources changes over time; The fused data is determined according to the original data value, the confidence level and the time function.

3. The method for extracting information from a comprehensive survey of natural resources according to claim 1, characterized in that: The extracted information is represented as: Among them, X is the input fusion data, and are the weight matrices for the first and second layers respectively, and are the biases of the first and second layers, respectively. (x)=max(0,x) is the activation function, is the activation function of the output layer, For auxiliary functions The coefficient of is the auxiliary function, and P is the number of auxiliary functions.

4. The method for extracting information from a comprehensive survey of natural resources according to claim 3, characterized in that: The step of updating the extracted information by a recursive filtering algorithm to obtain an updated state estimate corresponding to the extracted information includes: According to the extracted information, obtaining an observation value corresponding to the current moment and a predicted state estimate of the resource state in the extracted information before the current moment; Obtaining a nonlinear compensation function of the observation value; An updated state estimate corresponding to the extracted information is obtained based on the observed value, the predicted state estimate and the nonlinear compensation function.

5. The method for extracting information from a comprehensive survey of natural resources according to claim 4, characterized in that: The updated state estimate is expressed as: in, is the updated state estimate, is the predicted state estimate, is the Kalman gain, is the observation value at the current moment, H is the observation matrix, is the nonlinear compensation function The coefficient of is the nonlinear compensation function, and Q is the number of nonlinear compensation functions.

6. The method for extracting information from a comprehensive survey of natural resources according to claim 5, characterized in that: The step of adjusting the existing information extraction strategy according to the updated state estimation to obtain an updated information extraction strategy includes: analyzing the changing trends and dynamic characteristics of the current natural resource status based on the updated status estimate; Based on the change trends and dynamic characteristics, identify areas of significant change in the areas where natural resources are located; The emphasis and frequency of data acquisition are adjusted according to the significant change areas.

7. The method for extracting information from a comprehensive survey of natural resources according to claim 6, characterized in that: The multi-source remote sensing images include satellite images, aerial photography and drone images, and the data collected by the ground sensor network include temperature, humidity and soil moisture.

8. A natural resources comprehensive survey information extraction system, characterized in that: The system comprises: The data acquisition module is used to obtain surface information through multi-source remote sensing images and collect environmental data in real time through a ground sensor network to obtain multi-source data; A data processing module, used for performing spatial alignment and standardization processing on the multi-source data to obtain processed multi-source data; A data fusion module is used to perform data fusion on the processed multi-source data to obtain corresponding fused data; A first extraction module, used to extract and analyze the fused data to obtain corresponding extraction information; A state updating module, used to update the extracted information through a recursive filtering algorithm to obtain an updated state estimate corresponding to the extracted information; A second extraction module, configured to adjust the existing information extraction strategy according to the updated state estimation, obtain an updated information extraction strategy, and perform subsequent information extraction according to the updated information extraction strategy; The fused data is represented as: in, is the fused data, specifically the fused data value, is the original data value of the ith data source at position (x,y), is the weight coefficient of the i-th data source, is the confidence of the ith data source at position (x,y), is the time integration coefficient of data source i, is the time function of data source i, is the normalization coefficient, is a function of time The weight of is the jth time function, a, b, c, d are the upper and lower limits of the integral, N is the number of data sources, and M is the number of time functions; The extracting and analyzing the fused data to obtain corresponding extracted information includes: Inputting the fused data into a deep learning model for linear processing to obtain a corresponding linear transformation result; Obtaining an auxiliary function for representing a nonlinear transformation of the fused data; The linear transformation result and the auxiliary function are processed by the output layer activation function of the deep learning model to obtain corresponding extraction information.

Images