AI-based ecological environment monitoring service system

Abstract

The invention relates to an ecological environment monitoring service system based on AI, and belongs to the technical field of artificial intelligence. Wherein the data acquisition module is used for acquiring real-time environment data through a sensor group, adding a timestamp into time sequence data, storing according to an area number to obtain an environment data sequence, acquiring an image set by using an unmanned aerial vehicle, and acquiring image data by using satellite remote sensing; the sequence data preprocessing unit is used for obtaining abnormal values from the pollution index sequence by using a clustering algorithm, removing the abnormal values, filling missing values by linear interpolation, reducing dimensions by discrete cosine transform and then normalizing; the image data preprocessing unit is used for processing the image set and the remote sensing image and extracting elevation information from the corrected remote sensing data to construct DEM data; the data analysis module is used for calculating topographic data according to DEM data, obtaining vegetation density in combination with a normalized vegetation index, and calculating species, population density, a pollution prediction value and a fire risk prediction value through a model; and the response module is used for transmitting the key data to a human-computer interaction interface. And intelligent monitoring of the ecological environment is realized.

Description

Technical Field

[0001] The present invention belongs to the field of artificial intelligence technology, and specifically relates to an ecological environment monitoring service system based on AI. Background Art

[0002] There are many pain points in the traditional field of ecological environment monitoring. Early monitoring methods relied on regular manual inspections, which were not only inefficient, but also had a limited monitoring range and could not grasp the environmental dynamics of large monitoring areas in real time. For example, for remote mountainous areas or vast forest areas, manual inspections are difficult to fully cover and may miss some environmental change information.

[0003] In terms of data collection, the previous sensor technology was single and could only obtain a few key indicators, such as simple temperature and humidity data, and could not comprehensively and accurately monitor pollution indicators. Moreover, the data collected by the traditional monitoring system lacked time dimension identification, making it difficult to conduct time series analysis and effectively predict environmental change trends.

[0004] In the past, it was difficult to extract high-precision key information such as terrain and vegetation from complex images by relying on simple image processing methods. For example, there are large errors in obtaining accurate elevation information and vegetation coverage data from satellite remote sensing images, which affects the comprehensive assessment of the ecological environment.

[0005] Traditional methods for analyzing monitoring data mainly rely on manual experience and judgment, and lack intelligent data analysis models. This results in insufficient depth in mining environmental data, making it impossible to accurately predict forest fire risks and pollution diffusion trends in a timely manner, and making it difficult to provide strong support for environmental protection decision-making. Summary of the invention

[0006] In order to solve the above problems existing in the prior art, the present invention provides an AI-based ecological environment monitoring service system, comprising: a data acquisition unit, a sequence data preprocessing unit, an image data preprocessing unit, a data analysis module, and a response module; The data acquisition module is used to collect real-time environmental data of the monitoring area through the sensor group, obtain time series environmental data by adding timestamps according to the real-time environmental data, obtain the monitoring area number, store the time series environmental data according to the monitoring area number to obtain an environmental data sequence, collect image sets through drones, and collect remote sensing image data through satellite remote sensing technology; the environmental data sequence includes a pollution index sequence; The sequence data preprocessing unit is used to calculate sequence outliers according to the pollution index sequence through a clustering algorithm, remove the sequence outliers to obtain a removed sequence, fill missing values ​​according to the removed sequence through a linear interpolation algorithm to obtain a filled sequence, perform dimensionality reduction reduction according to the filled sequence through discrete cosine transform to obtain a reduced sequence, and normalize according to the reduced sequence to obtain a normalized sequence; The image data preprocessing unit is used to perform image preprocessing according to the image set through image processing software to obtain image preprocessing data, perform remote sensing image preprocessing according to the remote sensing image data to obtain corrected remote sensing data, directly extract elevation information by analyzing pixel spectral information according to the corrected remote sensing data, and obtain DEM data by constructing according to the elevation information; The data analysis module is used to calculate terrain data based on the DEM data by a fitting plane algorithm, obtain a normalized vegetation index, calculate vegetation density data based on the terrain data and the normalized vegetation index, calculate species and population density based on the image preprocessing data through a statistical model, calculate pollution prediction values ​​based on the normalized sequence through a prediction model, and calculate forest fire risk prediction values ​​based on the time series environmental data, the terrain data, and the vegetation density data through a fire risk prediction model; The response module is used to transmit the pollution prediction value, the forest fire risk prediction value, the species type, and the population density to a human-computer interaction interface.

[0007] Specifically, the real-time environmental data includes meteorological elements, hydrological elements, noise elements, electromagnetic radiation elements, atmospheric chemicals, water chemical indicators, and soil chemical properties.

[0008] Specifically, the clustering algorithm includes the following specific calculation steps: Preset an objective function, calculate an objective function value sequence according to the pollution index sequence through the objective function, preset a target threshold, traverse the objective function value sequence one by one to obtain a comparison function value, and judge whether the comparison function value is greater than the target threshold. If so, the comparison function value is an abnormal value of the sequence; if not, the comparison function value is a normal value; The objective function expression is: , in, J m (U,V) is the objective function value, U is the membership matrix, V is the cluster center matrix, i, j, k are variable parameters, x i is the firsti elements, v j is the cluster center matrix j elements, v k is the cluster center matrix k elements, n is the number of elements in the pollution indicator sequence, c is the number of clusters, m is the weighted index, ||*|| is the Euclidean distance.

[0009] Specifically, the image preprocessing includes image stitching and image correction; the remote sensing image preprocessing includes radiation calibration, atmospheric correction, and geometric correction.

[0010] Specifically, the specific calculation steps of the statistical model include: Obtain an extended image by random cropping according to the image preprocessing data, preset a rotation angle, rotate the extended image according to the rotation angle to obtain a rotated image, and flip the rotated image with the horizontal midline and the vertical midline of the image as symmetry axes to obtain a flipped image; Denoising the flipped image by Gaussian filtering to obtain denoised data, performing data enhancement on the denoised data to obtain enhanced data, and resizing the enhanced data by interpolation to obtain adjusted data; An image feature sequence is calculated through a CNN model based on the adjustment data, the species type is calculated through an LSTM model based on the image feature sequence, the image preprocessing data is labeled according to the species type to obtain a labeled image, the number of objects is obtained based on the labeled image, the area of ​​the monitoring area is obtained, and the population density data is calculated based on the number of objects and the area of ​​the monitoring area.

[0011] Specifically, the specific calculation steps of the prediction model include: Calculate sequence feature data according to the normalized sequence through a nonlinear dimensionality reduction algorithm, preset a sliding window, intercept the sequence feature data according to the sliding window to obtain a predicted feature sequence, and calculate the pollution prediction value according to the predicted feature sequence through a SA-BiLSTM model; The nonlinear dimensionality reduction algorithm expression is: , in, T i is the sequence feature data, i, j is a variable parameter, η is the learning rate, n is the number of normalized sequence elements,y i is the first i elements, y j is the first j elements, t is the number of iterations, α （t） is a dynamic term. Specifically, the fire risk prediction model includes the following specific calculation steps: Fire prediction related parameters are calculated through principal component analysis based on the time series environmental data, the terrain data, and the vegetation density data. Missing values ​​are filled in through mean interpolation method based on the fire prediction related parameters to obtain filled data. Discrete data are obtained through data binning based on the filled data. Encoded data are calculated through one-hot encoding based on the discretized data. Normalized data are obtained through normalization based on the encoded data. The forest fire risk prediction value is calculated through BP neural network based on the normalized data.

[0012] An electronic device includes a memory, a processor, and a computer program stored in the memory and capable of running on the processor. When the processor executes the program, an AI-based ecological environment monitoring service system as described above is implemented.

[0013] A storage medium containing computer executable instructions, which, when executed by a computer processor, are used to execute the AI-based ecological environment monitoring service system as described above.

[0014] The beneficial effects of the present invention are: (1) Efficient data collection and management: The sensor group collects environmental data in real time and adds timestamps to form an environmental data sequence. Combined with drone and satellite remote sensing technology, it realizes the comprehensive collection of multi-source data, greatly improving the efficiency and coverage of data collection. At the same time, the data is classified, stored and numbered to facilitate subsequent data processing and analysis.

[0015] (2) Accurate data preprocessing: Clustering algorithms are used to identify and remove outliers in the pollution indicator sequence, and linear interpolation algorithms are used to fill in missing values. Discrete cosine transform is then used for dimensionality reduction and normalization to ensure data accuracy and availability, providing a high-quality data foundation for subsequent analysis.

[0016] (3) In-depth image information extraction: Using image processing software and remote sensing image preprocessing technology, high-precision elevation information is extracted from the image set and DEM data is calculated, which can accurately reflect the topographic features. At the same time, the normalized vegetation index is obtained by analyzing the pixel spectral information, and then the vegetation density data is calculated, providing rich image information for ecological environment assessment.

[0017] (4) Intelligent data analysis and prediction: Based on the calculation of terrain data based on the plane fitting algorithm, combined with statistical models and prediction models, intelligent calculation and prediction of species types, population density, pollution prediction values ​​and forest fire risk prediction values ​​are achieved. This makes it possible to predict environmental risks in advance and provide a scientific decision-making basis for ecological environmental protection and governance.

[0018] (5) Convenient information interaction: The response module transmits key monitoring and prediction results to the human-computer interaction interface, which makes it convenient for relevant personnel to intuitively understand the ecological and environmental conditions and take corresponding protection and governance measures in a timely manner, thereby improving the practicality and interactivity of ecological and environmental monitoring services. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to facilitate understanding by those skilled in the art, the present invention is further described below with reference to the accompanying drawings.

[0020] Picture 1 This is a structural diagram of an AI-based ecological environment monitoring service system of the present invention. DETAILED DESCRIPTION

[0021] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the specific implementation methods, structures, features and effects of the present invention are described in detail below in conjunction with the accompanying drawings and preferred embodiments.

[0022] See also Picture 1 , an AI-based ecological environment monitoring service system, comprising: a data acquisition unit, a sequence data preprocessing unit, an image data preprocessing unit, a data analysis module, and a response module; The data acquisition module is used to collect real-time environmental data of the monitoring area through the sensor group, obtain time series environmental data by adding timestamps according to the real-time environmental data, obtain the monitoring area number, store the time series environmental data according to the monitoring area number to obtain an environmental data sequence, collect image sets through drones, and collect remote sensing image data through satellite remote sensing technology; the environmental data sequence includes a pollution index sequence; The sequence data preprocessing unit is used to calculate sequence outliers according to the pollution index sequence through a clustering algorithm, remove the sequence outliers to obtain a removed sequence, fill missing values ​​according to the removed sequence through a linear interpolation algorithm to obtain a filled sequence, perform dimensionality reduction reduction according to the filled sequence through discrete cosine transform to obtain a reduced sequence, and normalize according to the reduced sequence to obtain a normalized sequence; The image data preprocessing unit is used to perform image preprocessing according to the image set through image processing software to obtain image preprocessing data, perform remote sensing image preprocessing according to the remote sensing image data to obtain corrected remote sensing data, directly extract elevation information by analyzing pixel spectral information according to the corrected remote sensing data, and obtain DEM data by constructing according to the elevation information; The data analysis module is used to calculate terrain data based on the DEM data by a fitting plane algorithm, obtain a normalized vegetation index, calculate vegetation density data based on the terrain data and the normalized vegetation index, calculate species and population density based on the image preprocessing data through a statistical model, calculate pollution prediction values ​​based on the normalized sequence through a prediction model, and calculate forest fire risk prediction values ​​based on the time series environmental data, the terrain data, and the vegetation density data through a fire risk prediction model; The response module is used to transmit the pollution prediction value, the forest fire risk prediction value, the species type, and the population density to a human-computer interaction interface.

[0023] In this embodiment, DEM data is obtained by constructing it through GIS software according to the elevation information.

[0024] Specifically, the real-time environmental data includes meteorological elements, hydrological elements, noise elements, electromagnetic radiation elements, atmospheric chemicals, water chemical indicators, and soil chemical properties.

[0025] In this embodiment, the meteorological elements include temperature, humidity, wind direction, wind speed, air pressure, and precipitation; the hydrological elements include water level, flow rate, flow velocity, and water temperature; the noise elements include regional environmental noise and traffic noise; the electromagnetic radiation elements include electric field strength and magnetic field strength; the atmospheric chemical properties include gaseous pollutant content and particulate matter content; the water chemical indicators include water pH, dissolved oxygen, nutrients, heavy metals, and organic pollutants; and the soil chemical properties include soil pH, nutrient content, and heavy metal content.

[0026] Specifically, the clustering algorithm includes the following specific calculation steps: Preset an objective function, calculate an objective function value sequence according to the pollution index sequence through the objective function, preset a target threshold, traverse the objective function value sequence one by one to obtain a comparison function value, and judge whether the comparison function value is greater than the target threshold. If so, the comparison function value is an abnormal value of the sequence; if not, the comparison function value is a normal value; The objective function expression is: , in, J m (U,V) is the objective function value, U is the membership matrix, V is the cluster center matrix, i, j, k are variable parameters, x i is the first i elements, v j is the cluster center matrix j elements, v k is the cluster center matrix k elements, n is the number of elements in the pollution indicator sequence, c is the number of clusters, m is the weighted index, ||*|| is the Euclidean distance.

[0027] Specifically, the environmental data sequence also includes a non-pollution indicator sequence.

[0028] In this embodiment, the non-pollution indicator sequence includes a meteorological element sequence and a hydrological element sequence; the pollution indicator sequence includes a noise element sequence, an electromagnetic radiation element sequence, a water chemical indicator sequence, and a soil chemical property sequence; and the terrain data includes slope, aspect, and altitude.

[0029] Specifically, the image preprocessing includes image stitching and image correction; the remote sensing image preprocessing includes radiation calibration, atmospheric correction, and geometric correction.

[0030] Specifically, the specific calculation steps of the statistical model include: Obtain an extended image by random cropping according to the image preprocessing data, preset a rotation angle, rotate the extended image according to the rotation angle to obtain a rotated image, and flip the rotated image with the horizontal midline and the vertical midline of the image as symmetry axes to obtain a flipped image; Denoising the flipped image by Gaussian filtering to obtain denoised data, performing data enhancement on the denoised data to obtain enhanced data, and resizing the enhanced data by interpolation to obtain adjusted data; An image feature sequence is calculated through a CNN model based on the adjustment data, the species type is calculated through an LSTM model based on the image feature sequence, the image preprocessing data is labeled according to the species type to obtain a labeled image, the number of objects is obtained based on the labeled image, the area of ​​the monitoring area is obtained, and the population density data is calculated based on the number of objects and the area of ​​the monitoring area.

[0031] Specifically, the specific calculation steps of the prediction model include: Calculate sequence feature data according to the normalized sequence through a nonlinear dimensionality reduction algorithm, preset a sliding window, intercept the sequence feature data according to the sliding window to obtain a predicted feature sequence, and calculate the pollution prediction value according to the predicted feature sequence through a SA-BiLSTM model; The nonlinear dimensionality reduction algorithm expression is: , in, T i is the sequence feature data, i, j is a variable parameter, η is the learning rate, n is the number of normalized sequence elements, y i is the first i elements, y j is the first j elements, t is the number of iterations, α （t） It is a dynamic item.

[0032] Specifically, the fire risk prediction model includes the following specific calculation steps: Fire prediction related parameters are calculated through principal component analysis based on the time series environmental data, the terrain data, and the vegetation density data. Missing values ​​are filled in through mean interpolation method based on the fire prediction related parameters to obtain filled data. Discrete data are obtained through data binning based on the filled data. Encoded data are calculated through one-hot encoding based on the discretized data. Normalized data are obtained through normalization based on the encoded data. The forest fire risk prediction value is calculated through BP neural network based on the normalized data.

[0033] An electronic device includes a memory, a processor, and a computer program stored in the memory and capable of running on the processor. When the processor executes the program, an AI-based ecological environment monitoring service system as described above is implemented.

[0034] A storage medium containing computer executable instructions, which, when executed by a computer processor, are used to execute the AI-based ecological environment monitoring service system as described above.

[0035] The computer storage medium of the embodiment of the present invention may adopt any combination of one or more computer-readable media. The computer-readable medium may be a computer-readable signal medium or a computer-readable storage medium. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples (non-exhaustive list) of computer-readable storage media include: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this document, a computer-readable storage medium may be any tangible medium containing or storing a program that may be used by or in combination with an instruction execution system, device or device.

[0036] Computer-readable signal media may include data signals propagated in baseband or as part of a carrier wave, which carry computer-readable program code. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer-readable signal media may also be any computer-readable medium other than a computer-readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0037] The program code included on the computer readable medium can be transmitted with any appropriate medium, including but not limited to wireless, electric wire, optical cable, RF, etc., or any suitable combination of the above. The computer program code for performing the operation of the present invention can be written in one or more programming languages ​​or their combinations, and the programming language includes object-oriented programming languages-such as Java, Smalltalk, C++, and also includes conventional procedural programming languages-such as "C" language or similar programming languages. The program code can be executed completely on the user's computer, partially on the user's computer, as an independent software package, partially on the user's computer and partially on the remote computer, or completely on the remote computer or server. In the case of a remote computer, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computer (for example, using an Internet service provider to connect through the Internet).

[0038] The above description is only a preferred embodiment of the present invention and does not limit the present invention in any form. Although the present invention has been disclosed as a preferred embodiment as above, it is not used to limit the present invention. Any technical personnel in this field can make some changes or modify the technical contents disclosed above into equivalent embodiments without departing from the scope of the technical solution of the present invention. However, any simple modification, equivalent change and modification made to the above embodiments according to the technical essence of the present invention without departing from the content of the technical solution of the present invention still fall within the scope of the technical solution of the present invention.

Claims

1. An AI-based ecological environment monitoring service system, characterized in that: include: Data acquisition unit, sequence data preprocessing unit, image data preprocessing unit, data analysis module, response module; The data acquisition module is used to collect real-time environmental data of the monitoring area through the sensor group, obtain time series environmental data by adding timestamps according to the real-time environmental data, obtain the monitoring area number, store the time series environmental data according to the monitoring area number to obtain an environmental data sequence, collect image sets through drones, and collect remote sensing image data through satellite remote sensing technology; the environmental data sequence includes a pollution index sequence; The sequence data preprocessing unit is used to calculate sequence outliers according to the pollution index sequence through a clustering algorithm, remove the sequence outliers to obtain a removed sequence, fill missing values ​​according to the removed sequence through a linear interpolation algorithm to obtain a filled sequence, perform dimensionality reduction reduction according to the filled sequence through discrete cosine transform to obtain a reduced sequence, and normalize according to the reduced sequence to obtain a normalized sequence; The image data preprocessing unit is used to perform image preprocessing according to the image set through image processing software to obtain image preprocessing data, perform remote sensing image preprocessing according to the remote sensing image data to obtain corrected remote sensing data, directly extract elevation information by analyzing pixel spectral information according to the corrected remote sensing data, and obtain DEM data by constructing according to the elevation information; The data analysis module is used to calculate terrain data based on the DEM data by a fitting plane algorithm, obtain a normalized vegetation index, calculate vegetation density data based on the terrain data and the normalized vegetation index, calculate species and population density based on the image preprocessing data through a statistical model, calculate pollution prediction values ​​based on the normalized sequence through a prediction model, and calculate forest fire risk prediction values ​​based on the time series environmental data, the terrain data, and the vegetation density data through a fire risk prediction model; The response module is used to transmit the pollution prediction value, the forest fire risk prediction value, the species type, and the population density to a human-computer interaction interface.

2. The AI-based ecological environment monitoring service system according to claim 1 is characterized in that: The real-time environmental data includes meteorological elements, hydrological elements, noise elements, electromagnetic radiation elements, atmospheric chemical substances, water chemical indicators, and soil chemical properties.

3. The AI-based ecological environment monitoring service system according to claim 1 is characterized in that: The specific calculation steps of the clustering algorithm include: Preset an objective function, calculate an objective function value sequence according to the pollution index sequence through the objective function, preset a target threshold, traverse the objective function value sequence one by one to obtain a comparison function value, and judge whether the comparison function value is greater than the target threshold. If so, the comparison function value is an abnormal value of the sequence; if not, the comparison function value is a normal value; The objective function expression is: ， in, J m (U,V) is the objective function value, U is the membership matrix, V is the cluster center matrix, i, j, k are variable parameters, x i is the first i elements, v j is the cluster center matrix j elements, v k is the cluster center matrix k elements, n is the number of elements in the pollution indicator sequence, c is the number of clusters, m is the weighted index, ||*|| is the Euclidean distance.

4. The AI-based ecological environment monitoring service system according to claim 1 is characterized in that: The image preprocessing includes image stitching and image correction; the remote sensing image preprocessing includes radiation calibration, atmospheric correction and geometric correction.

5. The AI-based ecological environment monitoring service system according to claim 1 is characterized in that: The specific calculation steps of the statistical model include: Obtain an extended image by random cropping according to the image preprocessing data, preset a rotation angle, rotate the extended image according to the rotation angle to obtain a rotated image, and flip the rotated image with the horizontal midline and the vertical midline of the image as symmetry axes to obtain a flipped image; Denoising the flipped image by Gaussian filtering to obtain denoised data, performing data enhancement on the denoised data to obtain enhanced data, and resizing the enhanced data by interpolation to obtain adjusted data; An image feature sequence is calculated through a CNN model based on the adjustment data, the species type is calculated through an LSTM model based on the image feature sequence, the image preprocessing data is labeled according to the species type to obtain a labeled image, the number of objects is obtained based on the labeled image, the area of ​​the monitoring area is obtained, and the population density data is calculated based on the number of objects and the area of ​​the monitoring area.

6. The AI-based ecological environment monitoring service system according to claim 1 is characterized in that: The specific calculation steps of the prediction model include: Calculate sequence feature data according to the normalized sequence through a nonlinear dimensionality reduction algorithm, preset a sliding window, intercept the sequence feature data according to the sliding window to obtain a predicted feature sequence, and calculate the pollution prediction value according to the predicted feature sequence through a SA-BiLSTM model; The nonlinear dimensionality reduction algorithm expression is: ， in, T i is the sequence feature data, i, j is a variable parameter, η is the learning rate, n is the number of normalized sequence elements, y i is the first i elements, y j is the first j elements, t is the number of iterations, α （t） It is a dynamic item.

7. The AI-based ecological environment monitoring service system according to claim 1 is characterized in that: The specific calculation steps of the fire risk prediction model include: Fire prediction related parameters are calculated through principal component analysis based on the time series environmental data, the terrain data, and the vegetation density data. Missing values ​​are filled in through mean interpolation method based on the fire prediction related parameters to obtain filled data. Discrete data are obtained through data binning based on the filled data. Encoded data are calculated through one-hot encoding based on the discretized data. Normalized data are obtained through normalization based on the encoded data. The forest fire risk prediction value is calculated through BP neural network based on the normalized data.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and capable of running on the processor, characterized in that: When the processor executes the program, the AI-based ecological environment monitoring service system as described in any one of claims 1-7 is implemented.

9. A storage medium containing computer executable instructions, characterized in that: The computer executable instructions, when executed by a computer processor, are used to execute the AI-based ecological environment monitoring service system as described in any one of claims 1-7.

Images