Ecological environment degradation evaluation system and method, storage medium and electronic equipment

Abstract

The invention relates to the technical field of ecological environment evaluation, in particular to an ecological environment degradation evaluation system and method, a storage medium and electronic equipment. The ecological environment degradation evaluation system comprises a data acquisition module; the feature extraction module comprises a first extraction network and a second extraction network, the first extraction network is used for extracting space-spectrum fusion feature data of the hyperspectral remote sensing image time series data, and the second extraction network is used for extracting time-space fusion feature data of the hyperspectral remote sensing image time series data; the canopy density calculation module is used for calculating canopy density grade time sequence data by adopting a multivariable regression model according to the space-spectrum fusion feature data, the time-space fusion feature data and the environment monitoring time sequence data; the degradation evaluation module is used for performing degradation reason analysis according to the canopy density grade time series data, the hyperspectral remote sensing image time series data and the environment monitoring time series data; and a report generation module. The precision of ecological environment degradation evaluation can be improved.

Description

Technical Field

[0001] The present disclosure relates to the technical field of ecological environment assessment, and in particular to an ecological environment degradation assessment system, an ecological environment degradation assessment method, a storage medium and an electronic device. Background Art

[0002] With the intensification of global ecological and environmental changes, the problem of regional ecosystem degradation is becoming increasingly serious. Especially under the dual effects of large-scale human activities and natural factors, the speed and scope of ecological degradation continue to expand. Ecosystem degradation not only directly affects biodiversity, soil quality and water conservation, but may also cause serious environmental disasters such as soil erosion, desertification, forest fires, etc., further threatening the human living environment and economic sustainable development.

[0003] Traditional ecosystem degradation analysis methods mainly rely on manual surveys and ground monitoring. Although these methods can provide certain degradation assessments, they face problems such as poor timeliness, limited coverage, and low data accuracy.

[0004] It should be noted that the information disclosed in the above background technology section is only used to enhance the understanding of the background of the present disclosure, and therefore may include information that does not constitute the prior art known to ordinary technicians in the field. Summary of the invention

[0005] The purpose of the present disclosure is to provide an ecological environment degradation assessment system, an ecological environment degradation assessment method, a storage medium and an electronic device, aiming to solve the problem of poor accuracy in ecological environment degradation assessment.

[0006] Other features and advantages of the present disclosure will become apparent from the following detailed description, or may be learned in part by the practice of the present disclosure.

[0007] According to one aspect of the present disclosure, an ecological environment degradation assessment system is provided, characterized in that it includes: a data acquisition module for collecting hyperspectral remote sensing image time series data and environmental monitoring time series data; a feature extraction module, including a first extraction network and a second extraction network, the first extraction network is used to extract the space-spectrum fusion feature data of the hyperspectral remote sensing image time series data, and the second extraction network is used to extract the time-space fusion feature data of the hyperspectral remote sensing image time series data; a canopy closure calculation module is used to calculate the canopy closure level time series data using a multivariate regression model based on the space-spectrum fusion feature data, the time-space fusion feature data and the environmental monitoring time series data; a degradation assessment module is used to analyze the causes of degradation based on the canopy closure level time series data, the hyperspectral remote sensing image time series data and the environmental monitoring time series data; the degradation causes include natural factors and human factors; a report generation module is used to generate an ecological environment degradation assessment report based on the data analysis process and data analysis results in the degradation assessment module.

[0008] Optionally, the first extraction network includes: an input layer, which extracts a hyperspectral remote sensing image at a target time point based on the hyperspectral remote sensing image time series data; a collaborative feature extraction module, which collaboratively extracts spatial feature data and spectral feature data of the hyperspectral remote sensing image; and a fully connected layer, which fuses the spatial feature data and the spectral feature data to obtain spatial-spectral fusion feature data at the target time point.

[0009] Optionally, the second extraction network includes: an input layer, which receives the time series data of the hyperspectral remote sensing image; an LSTM module, which extracts the time series feature data in the time series data of the hyperspectral remote sensing image through memory units and a gating mechanism; a spatial convolution module, which extracts the spatial feature data in the time series data of the hyperspectral remote sensing image through a spatial convolution operation; and a fully connected layer, which fuses the time series feature data and the spatial feature data to obtain the time-space fusion feature data.

[0010] Optionally, the degradation cause analysis module includes: a natural factor analysis module, which is used to calculate the correlation between natural factors and canopy closure levels based on the canopy closure level time series data and the environmental monitoring time series data; a human factor analysis module, which is used to evaluate the impact of human factors on ecological environment degradation based on the canopy closure level time series data and the hyperspectral remote sensing image time series data; and a proportion analysis module, which evaluates the proportion of the impact of natural factors and human factors on ecological environment degradation respectively according to the analysis results of the natural factor analysis module and the human factor analysis module.

[0011] According to a second aspect of the present disclosure, an ecological environment degradation assessment method is provided, including: collecting hyperspectral remote sensing image time series data and environmental monitoring time series data; extracting spatial-spectral fusion feature data of the hyperspectral remote sensing image time series data based on a first extraction network, and extracting temporal-spatial fusion feature data of the hyperspectral remote sensing image time series data based on a second extraction network; calculating canopy closure level time series data using a multivariate regression model based on the spatial-spectral fusion feature data, the temporal-spatial fusion feature data and the environmental monitoring time series data; performing degradation cause analysis based on the canopy closure level time series data, the hyperspectral remote sensing image time series data and the environmental monitoring time series data to obtain data analysis results; and generating an ecological environment degradation assessment report based on the data analysis results.

[0012] According to a third aspect of the present disclosure, a computer-readable storage medium is provided, on which a computer program is stored. When the program is executed by a processor, the method for assessing ecological environmental degradation as in the above-mentioned embodiment is implemented.

[0013] According to the fourth aspect of the embodiments of the present disclosure, an electronic device is provided, characterized in that it includes: one or more processors; a storage device for storing one or more programs, when the one or more programs are executed by the one or more processors, the one or more processors implement the ecological environment degradation assessment method as in the above-mentioned embodiments.

[0014] The exemplary embodiments of the present disclosure may have some or all of the following beneficial effects: In the technical solutions provided in some embodiments of the present disclosure, on the one hand, the present disclosure uses hyperspectral remote sensing image time series data and environmental monitoring time series data as the data basis for evaluation, and extracts spatial-spectral fusion feature data and time-space fusion feature data through a deep learning network for multi-source data fusion, which is not only applicable to regional ecological environment monitoring and evaluation over a large range and long time series, but also can improve the quality and effectiveness of the data, and improve the accuracy and reliability of ecological environment degradation evaluation; on the other hand, through the results of the deep learning model and the multi-dimensional analysis framework combining hyperspectral remote sensing image time series data and environmental monitoring time series data, it is possible to accurately conduct a comprehensive analysis of natural and human factors in environmental degradation, which is conducive to further formulating corresponding strategies based on the analysis results to alleviate the crisis of ecological environment degradation and provide a scientific basis for ecological protection.

[0015] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The drawings herein are incorporated into the specification and constitute a part of the specification, showing embodiments consistent with the present disclosure, and together with the specification, are used to explain the principles of the present disclosure. Obviously, the drawings described below are only some embodiments of the present disclosure, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. In the drawings: Figure 1 A schematic diagram schematically illustrates the composition of an ecological environment degradation assessment system in an exemplary embodiment of the present disclosure; Figure 2 A schematic diagram schematically illustrates a flow chart of an ecological environment degradation assessment method in an exemplary embodiment of the present disclosure; Figure 3 The structure diagram of a computer system of an electronic device in an exemplary embodiment of the present disclosure is schematically shown. DETAILED DESCRIPTION

[0017] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in a variety of forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided so that this disclosure will be more comprehensive and complete and will fully convey the concept of the example embodiments to those skilled in the art.

[0018] In addition, the described features, structures or characteristics may be combined in one or more embodiments in any suitable manner. In the following description, many specific details are provided to provide a full understanding of the embodiments of the present disclosure. However, those skilled in the art will appreciate that the technical solutions of the present disclosure may be practiced without one or more of the specific details, or other methods, components, devices, steps, etc. may be adopted. In other cases, known methods, devices, implementations or operations are not shown or described in detail to avoid blurring the various aspects of the present disclosure.

[0019] The block diagrams shown in the accompanying drawings are merely functional entities and do not necessarily correspond to physically independent entities. That is, these functional entities may be implemented in software form, or in one or more hardware modules or integrated circuits, or in different networks and / or processor devices and / or microcontroller devices.

[0020] The flowcharts shown in the accompanying drawings are only exemplary and do not necessarily include all the contents and operations / steps, nor must they be executed in the order described. For example, some operations / steps can be decomposed, and some operations / steps can be combined or partially combined, so the actual execution order may change according to actual conditions.

[0021] With the intensification of global ecological and environmental changes, the problem of regional ecosystem degradation is becoming increasingly serious. Especially under the dual effects of large-scale human activities and natural factors, the speed and scope of ecological degradation continue to expand. Ecosystem degradation not only directly affects biodiversity, soil quality and water conservation, but may also cause serious environmental disasters such as soil erosion, desertification, forest fires, etc., further threatening the human living environment and economic sustainable development.

[0022] Traditional ecosystem degradation analysis methods mainly rely on manual surveys and ground monitoring. Although these methods can provide certain degradation assessments, they face problems such as poor timeliness, limited coverage, and low data accuracy.

[0023] At present, although there are some ecological monitoring methods based on deep learning, the following problems still exist: the integration and processing of multi-source data by deep learning models are still immature, and it is difficult to fully utilize the spatiotemporal correlation between historical data and real-time data; the existing methods lack a comprehensive analysis of the impact of natural and human factors, resulting in inaccurate determination of the causes of degradation; the analysis of large areas still relies on a single data source or method, and fails to achieve a comprehensive assessment from multiple dimensions and angles.

[0024] Therefore, this paper proposes a new regional intelligent ecological environment degradation assessment method driven by multi-source data based on deep learning and hyperspectral remote sensing, that is, by constructing a deep learning model, including multi-scale convolutional neural networks and spatiotemporal long and short-term memory networks, and combining hyperspectral data to conduct large-scale, long-term series and historical retrospective degradation analysis, and infer the causes of degradation and their impact, thereby realizing intelligent analysis of regional ecosystem degradation. It aims to solve the limitations of existing methods, improve the accuracy and adaptability of ecological degradation analysis, and provide more reliable data support for ecological protection and resource management.

[0025] The implementation details of the technical solution of the embodiment of the present disclosure are elaborated in detail below.

[0026] Figure 1 The following is a schematic diagram showing the composition of an ecological environment degradation assessment system in an exemplary embodiment of the present disclosure. Figure 1 As shown, the ecological environment degradation assessment system 100 includes: The data acquisition module 101 is used to acquire time series data of hyperspectral remote sensing images and time series data of environmental monitoring; The feature extraction module 102 includes a first extraction network and a second extraction network, wherein the first extraction network is used to extract spatial-spectral fusion feature data of the hyperspectral remote sensing image time series data, and the second extraction network is used to extract temporal-spatial fusion feature data of the hyperspectral remote sensing image time series data; A canopy closure calculation module 103 is used to calculate canopy closure level time series data using a multivariate regression model according to the space-spectrum fusion feature data, the time-space fusion feature data and the environmental monitoring time series data; The degradation assessment module 104 is used to analyze the degradation causes according to the canopy density level time series data, the hyperspectral remote sensing image time series data and the environmental monitoring time series data; the degradation causes include natural factors and human factors; The report generation module 105 is used to generate an ecological environment degradation assessment report according to the data analysis process and data analysis results in the degradation assessment module.

[0027] In the technical solutions provided in some embodiments of the present disclosure, on the one hand, the present disclosure uses hyperspectral remote sensing image time series data and environmental monitoring time series data as the data basis for evaluation, and extracts spatial-spectral fusion feature data and time-space fusion feature data through a deep learning network for multi-source data fusion, which is not only applicable to regional ecological environment monitoring and evaluation over a large range and long time series, but also can improve the quality and effectiveness of the data, and improve the accuracy and reliability of ecological environment degradation evaluation; on the other hand, through the results of the deep learning model and the multi-dimensional analysis framework combining hyperspectral remote sensing image time series data and environmental monitoring time series data, it is possible to accurately conduct a comprehensive analysis of natural and human factors in environmental degradation, which is conducive to further formulating corresponding strategies based on the analysis results to alleviate the crisis of ecological environment degradation and provide a scientific basis for ecological protection.

[0028] Below, each module of the ecological environment degradation assessment system in this example implementation will be described in more detail with reference to the accompanying drawings and embodiments.

[0029] In one embodiment of the present disclosure, the data acquisition module 101 is used to acquire hyperspectral remote sensing image time series data and environmental monitoring time series data.

[0030] Specifically, the data acquisition module 101 is used to perform long-term series monitoring of the target geographical area to obtain hyperspectral remote sensing image time series data composed of hyperspectral remote sensing images that vary with time, and environmental monitoring time series data composed of environmental monitoring data that vary with time.

[0031] Among them, hyperspectral remote sensing images can be collected by remote sensing satellites. Each band in the image provides a detailed description of the material characteristics of the ground objects. And because hyperspectral images contain multiple continuous spectral band information, they can provide sufficient spectral resolution to identify and distinguish the fine differences in surface materials. At the same time, hyperspectral images not only provide spectral information, but also contain spatial distribution information of ground objects.

[0032] Environmental monitoring data is the environmental information targeted by ecological environment degradation assessment, which can be collected through environmental surveys using corresponding collection instruments. For example, it can be one or more ecological environment data contents in forestry, agriculture, grassland and wetland in the target geographical area at a certain point in time, thereby achieving a comprehensive degradation analysis of the ecological environment and improving the accuracy of data analysis results.

[0033] In one embodiment of the present disclosure, the feature extraction module 102 includes a first extraction network and a second extraction network, which are respectively used to extract space-spectrum fusion feature data and time-space fusion feature data of the hyperspectral remote sensing image time series data.

[0034] Among them, the first extraction network is a multi-scale convolutional neural network (MS-CNN) model, and the structure of MS-CNN includes: an input layer, which extracts the hyperspectral remote sensing image at the target time point based on the hyperspectral remote sensing image time series data; a collaborative feature extraction module, which collaboratively extracts the spatial feature data and spectral feature data of the hyperspectral remote sensing image; a fully connected layer, which fuses the spatial feature data and the spectral feature data to obtain the spatial-spectral fusion feature data at the target time point.

[0035] CNN, the full name of Convolutional Neural Network, is a deep learning model that is particularly suitable for processing image data. The working principle of CNN is mainly based on convolution operations and feature extraction. In the convolution layer, the convolution kernel slides on the input image, and extracts local features in the image through weighted summation and bias operations in the local area. These features are nonlinearly transformed through activation functions to enhance the expressive power of the model. Then, the pooling layer downsamples the output of the convolution layer to reduce the dimension and amount of calculation of the data while retaining important features. Finally, the fully connected layer further processes and classifies the output of the pooling layer.

[0036] The "MS" in MS-CNN stands for Multi-Scale, which reflects the core characteristics and design goals of the MS-CNN network, which is to detect objects at different scales to improve detection accuracy and robustness.

[0037] Specifically, in order to solve the problem that traditional CNN models cannot fully mine the multi-dimensional features in the data, the MS-CNN model of the present application adopts multi-scale convolution operations to simultaneously extract image features at multiple scales, while adapting to remote sensing data at different resolutions and scales.

[0038] In addition, the MS-CNN model of the present application introduces a collaborative feature extraction module based on the traditional CNN model to achieve joint optimization of spatial information and spectral information, thereby improving the accuracy of ecological environment degradation detection.

[0039] The hyperspectral remote sensing image at the target time point is extracted as ,in H is the image height; W is the image width; C is the number of spectral bands. Input it into the trained MS-CNN model to obtain a feature map; that is, ,in F It is the feature dimension of the model output.

[0040] For each pixel in the hyperspectral remote sensing image, its spatial characteristics are considered at the same time and spectral characteristics , can be extracted collaboratively through the following formula:

[0041] Then, the spatial features and spectral features are fused by weighted summation to obtain spatial-spectral fusion feature data:

[0042] Where α is the weight that controls the contribution of spatial and spectral features.

[0043] Among them, in the process of extracting spatial features S, multi-scale convolution operations can be used. Through convolution kernels of different scales, multi-scale convolution operations are performed on the input hyperspectral image. The feature maps extracted at each scale correspond to the spatial information under different receptive fields. Small-scale convolution can capture local details, such as texture and edges, and large-scale convolution can extract global context, such as regional structure and shape. Multi-scale features are fused through feature concatenation, and the final output spatial feature map FF integrates spatial information of different granularities, enhancing the model's adaptability to changes in target scale.

[0044] At every scale s The convolution operation can be expressed as:

[0045] In the formula, For scale s Convolution operation on ; is the convolution weight; is the bias term; is the number of convolution kernels; It is a convolution operation to extract spatial features.

[0046] Then use the feature splicing method to achieve the fusion of multiple convolutional layer output features, specifically:

[0047] Where: The convolution results at different scales are concatenated to obtain the final spatial features. S .

[0048] The second extraction network is a spatiotemporal long short-term memory network (ST-LSTM) model. The structure of ST-LSTM includes: an input layer, which receives the time series data of the hyperspectral remote sensing image; an LSTM module, which extracts the time series feature data in the time series data of the hyperspectral remote sensing image through memory units and gating mechanisms; a spatial convolution module, which extracts the spatial feature data in the time series data of the hyperspectral remote sensing image through spatial convolution operations; and a fully connected layer, which fuses the time series feature data and the spatial feature data to obtain the time-space fusion feature data.

[0049] The ST in ST-LSTM stands for "Spatial-Temporal". In ST-LSTM, ST emphasizes the ability of the network structure to process spatiotemporal data. Spatial features refer to the distribution and relationship of data in two-dimensional or three-dimensional space, while temporal features describe the change of data over time. ST-LSTM combines spatial and temporal features, enabling it to perform well in processing data with spatiotemporal dependencies.

[0050] LSTM stands for "Long Short-Term Memory" neural network. It is a special recurrent neural network (RNN) structure that aims to solve the gradient vanishing and gradient exploding problems encountered by traditional RNN when processing long sequence data. LSTM achieves effective storage and updating of information by introducing structural units such as input gate, forget gate and output gate, so as to capture long-term dependencies in long sequence data.

[0051] Specifically, the ST-LSTM model combines the time series modeling capability and spatial information capture capability of the LSTM network to monitor changes in the temporal dimension of ecosystem degradation. It can also integrate spatial information and model spatiotemporal relationships at the same time, making it very suitable for monitoring dynamic changes in ecological degradation.

[0052] Assume that at each time step t, the input hyperspectral remote sensing image time series data , the hyperspectral remote sensing image time series data is converted into a hyperspectral remote sensing image sequence represented as .

[0053] The basic operation of LSTM can be expressed as:

[0054]

[0055]

[0056]

[0057]

[0058] in: is the input gate; For the Gate of Forgetfulness, is the output gate, for status; is the hidden state at the current moment, i.e., the time series feature data; is the sigmoid activation function, tanh is the hyperbolic tangent function; W and b are the weight matrix and bias term respectively.

[0059] In addition, in order to enhance the capture of spatial information of the data, the ST-LSTM model adds a spatial convolution module (SCM). After the spatial convolution operation, the spatial feature data of the hyperspectral remote sensing image can be extracted, which is expressed by the following formula:

[0060] The fully connected layer of ST-LSTM fuses the temporal feature data and the spatial feature data to obtain the time-space fusion feature data, which is expressed by the following formula:

[0061] Where: α is the weight; S t It is the spatial feature data extracted by the spatial convolution module; h t It is the time series feature data output by LSTM.

[0062] In one embodiment of the present disclosure, the canopy closure calculation module 103 is used to calculate canopy closure level time series data.

[0063] Specifically, a multivariate regression model is used to calculate the canopy closure level of the target geographical area at each period as the canopy closure level time series data. The mathematical model is as follows:

[0064] Where: C tis the canopy closure level at time t; is the independent variable related to the canopy closure level; and is the regression coefficient; is the error term.

[0065] According to the distribution of canopy density levels in different periods, the degradation rate is calculated by difference, and the trend of changes in each period is compared. The details are as follows:

[0066] Where: For the t The degradation rate at the moment; For the t Canopy density level at the time; For the t -1 The canopy level at time.

[0067] For the long-term ecological degradation trend, the time series model is used for fitting and the overall trend of degradation is calculated, which is as follows:

[0068] Where: is the total degradation degree; T is the time span of the data.

[0069] In one embodiment of the present disclosure, the degradation assessment module 104 is used to perform degradation cause analysis. The degradation cause analysis module includes: a natural factor analysis module, which is used to calculate the correlation between natural factors and canopy closure levels based on the canopy closure level time series data and the environmental monitoring time series data; a human factor analysis module, which is used to evaluate the impact of human factors on ecological environment degradation based on the canopy closure level time series data and the hyperspectral remote sensing image time series data; and a proportion analysis module, which evaluates the impact proportions of natural factors and human factors on ecological environment degradation based on the analysis results of the natural factor analysis module and the human factor analysis module.

[0070] Specifically, when analyzing the causes of ecological environment degradation, it is necessary to consider both natural and human factors.

[0071] When analyzing natural factors, a deep learning model was used to conduct correlation analysis between each natural factor and the changes in canopy density level, and then the correlation coefficient was used to evaluate the impact of natural factors on the changes in canopy density level.

[0072] The analysis is carried out by taking meteorological data as an example of natural factors. The environmental monitoring time series data includes meteorological data of different periods (referred to as meteorological data for short), which is correlated with the canopy closure level time series data (referred to as canopy closure data for short) to obtain the correlation coefficient between meteorological data and canopy closure data. R , the specific calculation process is as follows:

[0073] Where: X For meteorological data; Y is the canopy closure data; cov ( X , Y ) is the covariance; Var ( X )and Var ( Y ) are their respective variances.

[0074] Similarly, we can also analyze the impact of other natural factors on changes in canopy density levels, such as topography, hydrology, soil, biology, etc., and calculate the correlation coefficients between these environmental content data and canopy density data.

[0075] In the analysis of human factors, the impact of human factors such as regional development and building addition is evaluated by comparing hyperspectral remote sensing images of different time periods, and the changes are quantified using image difference technology. Image difference is expressed as:

[0076] Where: and They are t and t-1 Hyperspectral remote sensing image at the moment.

[0077] By comparing the correlation between natural factors and human factors, the weighted average method is used to comprehensively analyze and calculate the impact ratio of natural and human factors, specifically:

[0078]

[0079] Where: The proportion of natural factors; The proportion of human factors; is the weight of each natural factor; is the impact value of each natural factor.

[0080] In one embodiment of the present disclosure, the report generation module 105 is used to generate an ecological environment degradation assessment report.

[0081] The analysis report is automatically generated by the software, and the output results include text, images (such as PDF, JPG format, etc.), and related mathematical formulas. The image is automatically generated by the computer based on the output of the degradation assessment module, showing the ecological environment degradation process and related information such as the analysis of the causes of degradation. The ecological environment degradation process, for example, shows changes in canopy density, meteorological data comparison, satellite image changes, etc. This disclosure is only for illustrative purposes.

[0082] It should be noted that, although several modules or units of the device for action execution are mentioned in the above detailed description, this division is not mandatory. In fact, according to the embodiments of the present disclosure, the features and functions of two or more modules or units described above can be embodied in one module or unit. On the contrary, the features and functions of one module or unit described above can be further divided into multiple modules or units to be embodied.

[0083] The present disclosure also provides an ecological environment degradation assessment method, which is completed using the above-mentioned ecological environment degradation assessment system. Figure 2 The following is a schematic diagram of a flow chart of an ecological environment degradation assessment method in an exemplary embodiment of the present disclosure. Figure 2 The specific steps of the ecological environment degradation assessment method are as follows: Start to conduct intelligent analysis of ecological environment degradation assessment; Multivariate data retrieval and acquisition, acquisition of hyperspectral remote sensing data and environmental monitoring time series data (such as tree species, soil type, vegetation coverage, NDVI value, etc.); Data spatial feature extraction: Use the MS-CNN model to extract spatial features of images at multiple scales, and achieve joint optimization of spatial and spectral information to improve data accuracy; Extract data temporal features and use the ST-LSTM model to monitor changes in the temporal dimension of ecosystem degradation. At the same time, build a spatiotemporal relationship model to monitor the dynamic changes of ecological degradation. The canopy closure level distribution, based on the results of the above feature extraction, combined with multi-source data, uses a multivariate regression model to analyze and determine the canopy closure level distribution in each period; Analysis of canopy density change and degradation, using differentials to calculate degradation rates and comparing trends over time; Analysis and inference of the causes of degradation: Based on the environmental monitoring time series data, the causes of ecological degradation are analyzed and inferred; Analysis of the proportion of degradation causes, using the weighted average method to comprehensively analyze and calculate the impact ratio of natural and human factors; Finally, the report is automatically generated.

[0084] The above-mentioned ecological environment degradation assessment method has been described in detail when introducing the functions of each module in the ecological environment degradation assessment system, so it will not be elaborated here.

[0085] Based on the above methods, the deep learning models of multi-scale convolutional neural network (MS-CNN) and spatiotemporal long short-term memory network (ST-LSTM) can be used to extract spatial features and temporal change information from hyperspectral remote sensing images. At the same time, by constructing a multi-dimensional analysis framework of multiple sets of environmental data and remote sensing data, a comprehensive analysis of natural and human factors is carried out. Using multivariate regression and time series analysis techniques, the types and causes of ecological degradation in the region can be accurately identified, providing a scientific basis for ecological protection.

[0086] By combining deep learning with remote sensing technology, efficient and accurate monitoring of ecosystem degradation can be achieved. Especially in complex ecological environments, large-scale, long-term sequence data can be automatically extracted and analyzed, greatly improving the accuracy and efficiency of analysis, and accurately assessing regional ecological degradation. Intelligent data processing technology is used to improve the accuracy and reliability of degradation analysis results through model optimization and multi-source data fusion. Automated report output not only improves analysis efficiency, but also provides a scientific basis for fields such as ecological protection and environmental monitoring.

[0087] In an exemplary embodiment of the present disclosure, a storage medium capable of implementing the above method is also provided. It can adopt a portable compact disk read-only memory (CD-ROM) and include program code, and can be run on a terminal device, such as a mobile phone. However, the program product of the present disclosure is not limited to this. In this document, a readable storage medium can be any tangible medium containing or storing a program, which can be used by or in combination with an instruction execution system, an apparatus or a device.

[0088] In an exemplary embodiment of the present disclosure, an electronic device capable of implementing the above method is also provided. Figure 3 The structure diagram of a computer system of an electronic device in an exemplary embodiment of the present disclosure is schematically shown.

[0089] It should be noted that Figure 3 The computer system 300 of the electronic device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.

[0090] like Figure 3As shown, the computer system 300 includes a central processing unit (CPU) 301, which can perform various appropriate actions and processes according to the program stored in the read-only memory (ROM) 302 or the program loaded from the storage part 308 to the random access memory (RAM) 303. Various programs and data required for system operation are also stored in the RAM 303. The CPU 301, the ROM 302 and the RAM 303 are connected to each other through a bus 304. The input / output (I / O) interface 305 is also connected to the bus 304.

[0091] The following components are connected to the I / O interface 305: an input section 306 including a keyboard, a mouse, etc.; an output section 307 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker; a storage section 308 including a hard disk, etc.; and a communication section 309 including a network interface card such as a LAN (Local Area Network) card, a modem, etc. The communication section 309 performs communication processing via a network such as the Internet. A drive 310 is also connected to the I / O interface 305 as needed. A removable medium 311, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 310 as needed so that a computer program read therefrom is installed into the storage section 308 as needed.

[0092] In particular, according to an embodiment of the present disclosure, the process described below with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present disclosure includes a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program contains program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through the communication part 309, and / or installed from a removable medium 311. When the computer program is executed by the central processing unit (CPU) 301, various functions defined in the system of the present disclosure are executed.

[0093] It should be noted that the computer-readable medium shown in the embodiment of the present disclosure may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. 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 of computer-readable storage media may include, but are not limited to: 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), a 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 the present disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program, which may be used by or in combination with an instruction execution system, device or device. In the present disclosure, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a 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 computer-readable storage media, which may send, propagate, or transmit programs for use by or in conjunction with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wired, etc., or any suitable combination of the above.

[0094] The flow charts and block diagrams in the accompanying drawings illustrate the possible architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each box in the flow chart or block diagram can represent a module, a program segment, or a part of a code, and the above-mentioned module, program segment, or a part of a code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flow chart, and the combination of the boxes in the block diagram or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0095] The units involved in the embodiments described in the present disclosure may be implemented by software or hardware, and the units described may also be arranged in a processor. The names of these units do not, in some cases, constitute limitations on the units themselves.

[0096] As another aspect, the present disclosure further provides a computer-readable medium, which may be included in the electronic device described in the above embodiment; or may exist independently without being assembled into the electronic device. The above computer-readable medium carries one or more programs, and when the above one or more programs are executed by an electronic device, the electronic device implements the method described in the above embodiment.

[0097] It should be noted that, although several modules or units of the device for action execution are mentioned in the above detailed description, this division is not mandatory. In fact, according to the embodiments of the present disclosure, the features and functions of two or more modules or units described above can be embodied in one module or unit. On the contrary, the features and functions of one module or unit described above can be further divided into multiple modules or units to be embodied.

[0098] Through the description of the above implementation, it is easy for those skilled in the art to understand that the example implementation described here can be implemented by software, or by software combined with necessary hardware. Therefore, the technical solution according to the implementation of the present disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, and includes several instructions to enable a computing device (which can be a personal computer, a server, a touch terminal, or a network device, etc.) to execute the method according to the implementation of the present disclosure.

[0099] Those skilled in the art will readily appreciate other embodiments of the present disclosure after considering the specification and practicing the invention disclosed herein. The present disclosure is intended to cover any variations, uses or adaptations of the present disclosure that follow the general principles of the present disclosure and include common knowledge or customary technical means in the art that are not disclosed in the present disclosure.

[0100] It should be understood that the present disclosure is not limited to the exact structures that have been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.

Claims

1. An ecological environment degradation assessment system, characterized in that: include: Data acquisition module, used to collect time series data of hyperspectral remote sensing images and environmental monitoring time series data; A feature extraction module includes a first extraction network and a second extraction network, wherein the first extraction network is used to extract spatial-spectral fusion feature data of the hyperspectral remote sensing image time series data, and the second extraction network is used to extract temporal-spatial fusion feature data of the hyperspectral remote sensing image time series data; A canopy closure calculation module, used to calculate canopy closure level time series data using a multivariate regression model based on the space-spectrum fusion feature data, the time-space fusion feature data and the environmental monitoring time series data; A degradation assessment module, used for analyzing the causes of degradation according to the canopy closure level time series data, the hyperspectral remote sensing image time series data and the environmental monitoring time series data; the causes of degradation include natural factors and human factors; The report generation module is used to generate an ecological environment degradation assessment report based on the data analysis process and data analysis results in the degradation assessment module.

2. The ecological environment degradation assessment system according to claim 1, characterized in that: The first extraction network comprises: An input layer extracts a hyperspectral remote sensing image at a target time point based on the hyperspectral remote sensing image time series data; A collaborative feature extraction module, collaboratively extracting spatial feature data and spectral feature data of the hyperspectral remote sensing image; The fully connected layer fuses the spatial feature data and the spectral feature data to obtain the spatial-spectral fusion feature data at the target time point.

3. The ecological environment degradation assessment system according to claim 1, characterized in that: The second extraction network comprises: An input layer receives the hyperspectral remote sensing image time series data; An LSTM module extracts time series feature data from the time series data of the hyperspectral remote sensing image through a memory unit and a gating mechanism; A spatial convolution module, which extracts spatial feature data from the hyperspectral remote sensing image time series data through a spatial convolution operation; The fully connected layer fuses the temporal feature data and the spatial feature data to obtain the time-space fused feature data.

4. The ecological environment degradation assessment system according to claim 1, characterized in that: The degradation cause analysis module includes: A natural factor analysis module, used to calculate the correlation between natural factors and canopy closure levels based on the canopy closure level time series data and the environmental monitoring time series data; A human factor analysis module, used to evaluate the impact of human factors on ecological environment degradation based on the canopy density level time series data and the hyperspectral remote sensing image time series data; The proportion analysis module evaluates the influence proportions of natural factors and human factors on ecological environment degradation respectively according to the analysis results of the natural factors analysis module and the human factors analysis module.

5. A method for assessing ecological environment degradation, characterized in that: include: Collect hyperspectral remote sensing image time series data and environmental monitoring time series data; Extracting spatial-spectral fusion feature data of the hyperspectral remote sensing image time series data based on a first extraction network, and extracting temporal-spatial fusion feature data of the hyperspectral remote sensing image time series data based on a second extraction network; Calculate the canopy closure level time series data using a multivariate regression model based on the space-spectrum fusion feature data, the time-space fusion feature data and the environmental monitoring time series data; Performing degradation cause analysis based on the canopy density level time series data, the hyperspectral remote sensing image time series data, and the environmental monitoring time series data to obtain data analysis results; An ecological environment degradation assessment report is generated based on the data analysis results.

6. The ecological environment degradation assessment method according to claim 5 is characterized in that: The step of extracting spatial-spectral fusion feature data of the hyperspectral remote sensing image time series data based on the first extraction network includes: Extracting a hyperspectral remote sensing image at a target time point based on the hyperspectral remote sensing image time series data; Collaboratively extracting spatial feature data and spectral feature data of the hyperspectral remote sensing image; The spatial feature data and the spectral feature data are fused to obtain spatial-spectral fusion feature data at the target time point.

7. The ecological environment degradation assessment method according to claim 5, characterized in that: The step of extracting the time-space fusion feature data of the hyperspectral remote sensing image time series data based on the second extraction network includes: Extracting time series feature data from the hyperspectral remote sensing image time series data through a memory unit and a gating mechanism; and Extracting spatial feature data from the hyperspectral remote sensing image time series data through a spatial convolution operation; The temporal feature data and the spatial feature data are fused to obtain the time-space fused feature data.

8. The ecological environment degradation assessment method according to claim 5, characterized in that: The data analysis results obtained by performing degradation cause analysis based on the canopy density level time series data, the hyperspectral remote sensing image time series data and the environmental monitoring time series data include: Calculating the correlation between natural factors and canopy closure levels according to the canopy closure level time series data and the environmental monitoring time series data; and Evaluate the impact of human factors on ecological environment degradation based on the canopy density level time series data and the hyperspectral remote sensing image time series data; Based on the correlation and the impact value, the impact ratios of natural factors and human factors on ecological environment degradation are evaluated to obtain the data analysis result.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the ecological environment degradation assessment method as described in any one of claims 5 to 8 is implemented.

10. An electronic device, characterized in that: include: one or more processors; A storage device for storing one or more programs. When the one or more computer programs are executed by the one or more processors, the one or more processors implement the ecological environment degradation assessment method as described in any one of claims 5 to 8.

Images