A method and system for detecting forest phenology and environmental disturbance

Abstract

The present invention discloses a method and system for detecting forest phenology and environmental disturbance, including: obtaining remote sensing time series image data of forest area; constructing a time series full convolutional neural network model based on convolutional neural network and time attention mechanism according to the remote sensing time series image data of forest area; constructing a sample data set based on the remote sensing time series image data of forest area, training the time series full convolutional neural network model, and obtaining a forest phenology and environmental disturbance detection model; inputting the forest area remote sensing time series image data into the forest area phenology and environmental disturbance detection model, and obtaining the forest area phenology and environmental disturbance detection result. The technical solution provided by the present invention can use the forest area remote sensing time series image data with high spatial resolution in the time series to construct a time series full convolutional neural network model, identify the changes of forest phenology and environmental disturbance, and locate the disturbance area and time.

Description

Technical Field

[0001] The present invention relates to the field of forest area dynamic change detection, and in particular to a forest area phenology and environmental disturbance detection method. Background Art

[0002] Forest phenology refers to the periodic phenomena of organisms (such as plants and animals) in forests that change with the seasons. This includes the germination, flowering, fruiting, and leaf fall of plants, as well as the migration and reproduction of animals. The changes in forest phenology are affected by many factors, including natural factors such as climate, topography, and soil, as well as human factors such as human activities.

[0003] Environmental disturbances in forest areas refer to various factors that interfere with forest ecosystems, including natural and human factors. These disturbances may lead to changes in the structure, function and dynamics of forest ecosystems.

[0004] There is a close relationship between forest phenology and environmental disturbances. On the one hand, environmental disturbances can affect changes in forest phenology. For example, climate change may cause changes in the growth and development cycle of plants, which in turn affects the structure and function of forest ecosystems. On the other hand, changes in forest phenology may also have a feedback effect on environmental disturbances. For example, changes in forest vegetation may affect soil moisture and nutrient conditions, which in turn affects environmental issues such as hydrological processes and soil erosion.

[0005] Dynamic change detection of forest phenology and environmental disturbance can timely understand the quantity and quality of forest resources, grasp the laws and trends of forest resource growth and decline, analyze the natural, economic and social objective conditions that affect and restrict forest growth, establish or update forest resource archives, and is of great significance for revealing ecosystem environmental changes, vegetation restoration and reconstruction layout, etc. However, existing technologies rarely involve detection of forest phenology and environmental disturbance. Summary of the invention

[0006] The present invention provides a method for detecting forest phenology and environmental disturbance, which can utilize forest remote sensing time series image data to construct a time series fully convolutional neural network model to identify changes in forest phenology and environmental disturbance.

[0007] To achieve the above object, the embodiments of the present invention adopt the following technical solutions:

[0008] A method for detecting forest phenology and environmental disturbances comprises the following steps:

[0009] Obtain remote sensing time series image data of forest areas;

[0010] According to the remote sensing time series image data of forest areas, a time series full convolutional neural network model is constructed based on convolutional neural network and temporal attention mechanism;

[0011] Based on the remote sensing time series image data of forest areas, a sample data set was constructed, and the time series fully convolutional neural network model was trained to obtain the forest phenology and environmental disturbance detection model;

[0012] The forest area remote sensing time series image data is input into the forest area phenology and environmental disturbance detection model to obtain the forest area phenology and environmental disturbance detection results.

[0013] According to one aspect of the present invention, the forest phenology and environmental disturbance detection method further includes: preprocessing the forest remote sensing time series image data.

[0014] According to one aspect of the present invention, the preprocessing of forest area remote sensing time series image data includes: radiation and atmosphere correction, cloud and shadow removal, and geometric correction.

[0015] According to one aspect of the present invention, the method of constructing a time series full convolutional neural network model based on a convolutional neural network and a temporal attention mechanism includes:

[0016] A hierarchical deep convolutional network architecture is used as a spatial encoder. By setting convolutional layers, ReLU activation functions and batch normalization layers step by step and combining them with step convolution modules, the feature extraction is deepened layer by layer to extract the spatial features of the image and obtain feature maps stacked along the time axis and processed by spatial encoding.

[0017] According to one aspect of the present invention, the time series full convolutional neural network model based on the convolutional neural network and the time attention mechanism further includes:

[0018] The temporal attention encoder is integrated and the temporal attention mechanism is applied to assign different weights to the feature maps at each time point in the sequence, so as to achieve efficient encoding of the dataset in the temporal dimension and identify the temporal dynamic characteristics of the forest area.

[0019] According to one aspect of the present invention, the time series full convolutional neural network model based on the convolutional neural network and the time attention mechanism further includes:

[0020] The model is built based on U-Net, and the spatial resolution of the feature map is gradually improved by using the layered strided transposed convolution technology. The skip connection technology and attention mechanism are introduced to fuse the feature map with the temporal dynamic features to reconstruct the spatial details of the original remote sensing time series image data.

[0021] According to one aspect of the present invention, when constructing a sample data set, data enhancement is performed on image data, and the sample data set is constructed based on a time window sliding technology.

[0022] According to one aspect of the present invention, the training of the time series fully convolutional neural network model includes:

[0023] The cross entropy loss function is selected to handle pixel-level classification tasks, and the Dice loss is introduced to improve the recognition ability of small objects and balance the problem of uneven categories;

[0024] Use Adam optimizer to improve training efficiency and model convergence speed;

[0025] Implement a learning rate decay strategy to achieve fast learning in the early stages of training and maintain stable convergence in the later stages;

[0026] An early stopping mechanism is used to prevent overfitting.

[0027] According to one aspect of the present invention, the training of the time series fully convolutional neural network model further comprises:

[0028] Comprehensively evaluate the model based on multiple indicators on the validation set, including but not limited to precision, recall, and F1 score;

[0029] Based on the comprehensive evaluation results, the network model is fine-tuned.

[0030] According to one aspect of the present invention, the forest phenology and environmental disturbance detection method further comprises the following steps:

[0031] Based on the detection results of forest phenology and environmental disturbance, a spatiotemporal distribution map of forest phenology and environmental disturbance was constructed.

[0032] A forest phenology and environmental disturbance detection system, based on the forest phenology and environmental disturbance detection method as described above, comprises:

[0033] A data acquisition module is used to acquire remote sensing time series image data of forest areas;

[0034] The model building module is used to build a time series full convolutional neural network model based on the convolutional neural network and the temporal attention mechanism according to the forest area remote sensing time series image data;

[0035] The training module is used to construct a sample data set based on the remote sensing time series image data of the forest area, train the time series full convolutional neural network model, and obtain the forest phenology and environmental disturbance detection model;

[0036] The detection module is used to input the forest area remote sensing time series image data into the forest area phenology and environmental disturbance detection model to obtain the forest area phenology and environmental disturbance detection results.

[0037] Advantages of the present invention:

[0038] The present invention provides a method for detecting forest phenology and environmental disturbance, which can utilize forest remote sensing time series image data with high spatial resolution in a temporal sequence to construct a time series fully convolutional neural network model, identify changes in forest phenology and environmental disturbance, and locate disturbance areas and times.

[0039] This method first collects remote sensing time series data and optimizes it through a series of preprocessing steps to ensure data quality and availability; then, based on the preprocessed remote sensing time series data, a fully convolutional neural network model optimized for time series data analysis is constructed to map in detail the periodic phenological characteristics of vegetation coverage over time and the spatiotemporal distribution of environmental disturbance events, accurately identify the time and location of environmental disturbance events in the vegetation coverage area, and achieve efficient and accurate monitoring of vegetation disturbance information.

[0040] This method can also conduct a comprehensive evaluation of the network model based on multiple indicators, and fine-tune and improve the network model according to the evaluation results; according to the detection results of forest phenology and environmental disturbance, the data can be integrated to construct the spatiotemporal distribution map of forest phenology and environmental disturbance, providing important information for environmental monitoring and decision-making analysis. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0042] Figure 1 This is a flow chart of a method for detecting forest phenology and environmental disturbance according to Embodiment 1 of the present invention;

[0043] Figure 2 It is a flow chart of a method for detecting forest phenology and environmental disturbance according to the present invention;

[0044] Figure 3 This is a flow chart of a method for detecting forest phenology and environmental disturbances as described in Example 2 of the present invention. DETAILED DESCRIPTION

[0045] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0046] Embodiment 1

[0047] like Figure 1and 2 As shown, a method for detecting forest phenology and environmental disturbance includes the following steps:

[0048] S1: Obtain remote sensing time series image data of forest areas.

[0049] This method mainly uses Gaofen (GF) high-time series data to carry out research, and multispectral remote sensing data such as Sentinel-2 (S2) and Landsat satellites as auxiliary data sources to participate in the research. Among them, Gaofen series satellites (such as GF-1 / 6) provide high-temporal and spatial resolution remote sensing images, which are helpful for accurately monitoring surface changes; S2 provides 10 to 20 meters of high-resolution multispectral images, which are suitable for capturing vegetation cover changes; Landsat series provides long-term continuous observation records, which are suitable for analyzing long-term trends.

[0050] Specifically, multispectral remote sensing time series data must meet the following requirements: spatial resolution is better than 15 meters, and better than 10 meters in key monitoring areas; to ensure that multispectral remote sensing data fully reflect the changes in each stage within the key growing season and capture key phenological change points, based on data availability, data are acquired every two weeks to one month over a one-year span to form a continuous time series with consistent radiation and spatial alignment; in the selection of multispectral sensors, priority is given to using data acquired by the same sensor at different time points. If conditions limit the use of different multispectral sensors, sensors with similar band characteristics should be selected as much as possible to ensure the comparability of the obtained data; cloud cover area is less than 5%.

[0051] Since the test areas are distributed in the north and south of my country, different geographical features, cloud coverage and surface characteristics directly affect the selection of data sources. Therefore, we will carefully select appropriate remote sensing data according to the specific conditions of each test area to ensure that the collected data can accurately reflect the actual conditions of each place and effectively respond to the challenges brought by climate factors, ensuring the accuracy of research data and the practicality of results.

[0052] The forest phenology and environmental disturbance detection method also includes: preprocessing the forest remote sensing time series image data.

[0053] The preprocessing of forest area remote sensing time series image data includes: radiation and atmosphere correction, cloud and shadow removal, and geometric correction.

[0054] Preprocessing aims to improve data quality and provide a solid foundation for subsequent analysis.

[0055] S2: Based on the remote sensing time series image data of the forest area, a time series fully convolutional neural network model is constructed based on convolutional neural network and temporal attention mechanism.

[0056] The time series fully convolutional neural network model is dedicated to improving the recognition accuracy of vegetation cover changes and environmental disturbance events, and emphasizes the ability to locate disturbances in space and time. The model consists of three main parts: first, the spatial encoding module uses deep convolutional layers to extract the spatial features of the image; second, the temporal encoding module introduces an attention mechanism to identify key change moments in the time series and enhance sensitivity to temporal changes; finally, the spatial decoding module is responsible for accurately reconstructing the spatial details of the semantic segmentation results while maintaining high spatial resolution.

[0057] The time series full convolutional neural network model based on the convolutional neural network and the time attention mechanism is constructed as follows:

[0058] S21: Use the layered deep convolutional network architecture as the spatial encoder. By setting the convolution layer, ReLU activation function and batch normalization layer step by step, combined with the strided convolution module, the feature extraction is deepened layer by layer to extract the spatial features of the image and obtain the feature map stacked along the time axis and processed by spatial encoding.

[0059] A layered deep convolutional network architecture is used as a spatial encoder to extract spatial features of remote sensing images. By setting convolutional layers, ReLU activation functions, and batch normalization layers step by step, combined with the step convolution module, feature extraction is deepened layer by layer, effectively reducing feature dimensions and compressing remote sensing data information, providing effective features for the temporal encoding and spatial decoding stages.

[0060] In view of the inherent complexity of time series data, a strategy of stacking feature maps processed by spatial encoding along the time axis is adopted, so that the model can comprehensively utilize the information of the entire time series in subsequent analysis, effectively identify and capture various vegetation cover change patterns from seasonal changes to sudden environmental disturbances, enhance the model's ability to analyze remote sensing image time series data, and provide reliable data support for accurately identifying and locating vegetation changes and environmental disturbances.

[0061] S22: Integrate the temporal attention encoder and apply the temporal attention mechanism to assign different weights to the feature maps at each time point in the sequence, achieve efficient encoding of the dataset in the temporal dimension, and identify the temporal dynamic characteristics of the forest area.

[0062] Temporal encoding is a key step in step S4, which aims to accurately capture and analyze the dynamic changes of remote sensing time series data. By integrating the temporal attention encoder and applying the temporal attention mechanism, the analysis of feature sequences is optimized. This mechanism enhances the model's ability to identify key time points by assigning different weights to the feature graphs of each time point in the sequence, while achieving effective compression of redundant time information and improving processing efficiency.

[0063] In the implementation, the method of combining the temporal attention encoder with the multi-scale analysis strategy is adopted to achieve efficient encoding of the time dimension on remote sensing datasets with uniform spatial resolution characteristics. This strategy aims to significantly improve the model's ability to identify and analyze the intrinsic changes in time series data by comprehensively utilizing various time scales, comprehensively capture periodic phenological changes and short-term environmental disturbances in time series data, and optimize the model's effectiveness in accurately identifying and locating vegetation cover changes and environmental disturbance events.

[0064] S23: Build a model based on U-Net, use layered strided transposed convolution technology to gradually improve the spatial resolution of feature maps, introduce skip connection technology and attention mechanism, fuse feature maps with temporal dynamic features, and reconstruct the spatial details of the original remote sensing time series image data.

[0065] The spatial decoding stage adopts a design that is symmetrical to the spatial encoding stage. By using the layered strided transposed convolution technology to gradually improve the spatial resolution of the feature map, the spatial details of the original remote sensing time series data can be accurately reconstructed. At the same time, by introducing the jump connection technology, the deep features extracted in the spatial encoding stage and the temporal dynamic features identified in the temporal encoding stage are effectively integrated, the model's ability to capture spatial details and the comprehensive expression of temporal information is optimized, and the recognition accuracy of vegetation cover changes and environmental disturbances in spatiotemporal distribution information is enhanced.

[0066] This step builds a model based on U-Net. U-Net is chosen because its symmetrical design and feature fusion capabilities have been proven in medical image processing and remote sensing image analysis. This step customizes the U-Net architecture, including optimizing the transposed convolution layer to better restore high-resolution spatial information, and adjusting the jump connection strategy to promote the fusion of spatial and temporal dimension features. In addition, considering the characteristics of remote sensing time series data, the attention mechanism is introduced to enhance the model's ability to recognize key spatiotemporal change features, and improve the model's ability to identify and accurately locate vegetation cover changes and environmental disturbance events.

[0067] S3: Based on the remote sensing time series image data of the forest area, a sample data set is constructed, and the time series fully convolutional neural network model is trained to obtain the forest phenology and environmental disturbance detection model.

[0068] When constructing the sample data set, the image data is enhanced and the sample data set is constructed based on the time window sliding technology.

[0069] The sample data construction first generates a sample set reflecting the change spots based on the existing remote sensing images and spot data. The calibrated samples include the vegetation cover type before and after and the change type samples.

[0070] Using two phases of remote sensing images from the same sensor in adjacent quarters, vegetation cover remote sensing image change calibration is performed by province or region based on remote sensing image features and vegetation cover changes. Based on the vegetation cover type reflected by the attributes of fixed sample plots or forest map patches, the remote sensing image features of the location are used for comparative analysis, and representative plots with relatively consistent remote sensing image features are selected. The vegetation cover type before and after, remote sensing data source, remote sensing data acquisition time, spatial resolution and other attribute information are calibrated according to a 512m×512m square vector frame.

[0071] Based on the established remote sensing images and data labels of each vegetation cover type, combined with the remote sensing images and forest and grass resource data of the previous and next two periods, and referring to the attribute information of forest and grass resource spots, the characteristic change areas of the remote sensing images of the previous and next two periods are compared and determined, and the change type and the corresponding remote sensing data source, remote sensing data acquisition time, spatial resolution and other attribute information are calibrated according to the 512m×512m square vector frame. Specific change types include but are not limited to construction project use, logging, reclamation, disaster events, and ecological protection and restoration measures.

[0072] This step aims to establish a large-scale, diverse and spatially distributed sample database to support iterative training and parameter optimization of deep learning models. The goal is to increase the training accuracy of the model to more than 85%. Models that do not meet this accuracy standard will be inspected in detail, supplemented with labels for interpretation signs and change types, and training parameters will be adjusted and iterative training will be repeated to achieve higher accuracy.

[0073] By carefully dividing the pre-processed time series remote sensing data, we create training and validation sets that reflect the diversity of vegetation coverage and environmental changes. At the same time, we use data enhancement methods such as random rotation, flipping, and cropping to enhance the model's generalization ability for diverse surface features. The training samples constructed based on the time window sliding technology not only capture the vegetation coverage information at a single moment, but also integrate the change trend on the time axis, providing the model with dynamic change information on vegetation coverage.

[0074] The training of the time series fully convolutional neural network model includes:

[0075] The cross entropy loss function is selected to handle pixel-level classification tasks, and the Dice loss is introduced to improve the recognition ability of small objects and balance the problem of uneven categories;

[0076] Use Adam optimizer to improve training efficiency and model convergence speed;

[0077] Implement learning rate decay to achieve fast learning in the early stages of training and maintain stable convergence in the later stages;

[0078] An early stopping mechanism is used to prevent overfitting.

[0079] Model training is the key to achieving accurate semantic segmentation based on time series remote sensing data, which includes multiple links such as loss function design, optimizer selection and learning rate management.

[0080] The cross entropy loss function is selected to handle pixel-level classification tasks, and the Dice loss is introduced to improve the recognition ability of small targets and balance the problem of uneven categories. The Adam optimizer is selected to improve training efficiency and model convergence speed. By implementing learning rate decay, the model can achieve rapid learning in the early stage of training and maintain stable convergence in the later stage. The early stopping mechanism can prevent overfitting of the model.

[0081] S4: Input the forest area remote sensing time series image data into the forest area phenology and environmental disturbance detection model to obtain the forest area phenology and environmental disturbance detection results.

[0082] This method focuses on two types of change detection in forest areas. First, according to the vegetation coverage type classification system of the "National Technical Regulations for Comprehensive Monitoring of Forest and Grass Ecology", combined with the specific vegetation coverage types of the project technology integration demonstration base, the change patches within the vegetation coverage are extracted; second, according to the remote sensing interpretation and verification change types in the "National Technical Regulations for Comprehensive Monitoring of Forest and Grass Ecology", the change types of the change patches obtained in the previous step are extracted, including construction project use, logging, reclamation, disasters, and ecological protection and restoration measures.

[0083] The goal of the detection model output of this method is to generate a feature map with the same resolution as the input image, in which each pixel is carefully classified according to its perturbed state. This classification operation is implemented through the spatial decoder inside the model, which maps the high-dimensional features to a predefined category space through a 1x1 convolutional layer to achieve accurate recognition of the surface state.

[0084] With the help of the model's accurate classification of each pixel, this method focuses on determining the specific time and location of the disturbance event.

[0085] The beneficial effect of this embodiment is that this embodiment proposes a method that combines remote sensing time series data with a fully convolutional neural network to automatically identify the phenological characteristics and disturbance information of vegetation cover types. This method first collects remote sensing time series data and optimizes it through a series of preprocessing steps to ensure data quality and availability; then, based on the preprocessed remote sensing time series data, a fully convolutional neural network model optimized for time series data analysis is constructed to map in detail the periodic phenological characteristics of vegetation coverage over time and the spatiotemporal distribution of environmental disturbance events, accurately identify the time and location of environmental disturbance events in the vegetation coverage area, and achieve efficient and accurate monitoring of vegetation disturbance information.

[0086] Embodiment 2

[0087] like Figure 2 and3 As shown, a method for detecting forest phenology and environmental disturbance includes the following steps:

[0088] S1: Obtain remote sensing time series image data of forest areas.

[0089] This method mainly uses Gaofen (GF) high-time series data to carry out research, and multispectral remote sensing data such as Sentinel-2 (S2) and Landsat satellites as auxiliary data sources to participate in the research. Among them, Gaofen series satellites (such as GF-1 / 6) provide high-temporal and spatial resolution remote sensing images, which are helpful for accurately monitoring surface changes; S2 provides 10 to 20 meters of high-resolution multispectral images, which are suitable for capturing vegetation cover changes; Landsat series provides long-term continuous observation records, which are suitable for analyzing long-term trends.

[0090] Specifically, multispectral remote sensing time series data must meet the following requirements: spatial resolution is better than 15 meters, and better than 10 meters in key monitoring areas; to ensure that multispectral remote sensing data fully reflect the changes in each stage within the key growing season and capture key phenological change points, based on data availability, data are acquired every two weeks to one month over a one-year span to form a continuous time series with consistent radiation and spatial alignment; in the selection of multispectral sensors, priority is given to using data acquired by the same sensor at different time points. If conditions limit the use of different multispectral sensors, sensors with similar band characteristics should be selected as much as possible to ensure the comparability of the obtained data; cloud cover area is less than 5%.

[0091] Since the test areas are distributed in the north and south of my country, different geographical features, cloud coverage and surface characteristics directly affect the selection of data sources. Therefore, we will carefully select appropriate remote sensing data according to the specific conditions of each test area to ensure that the collected data can accurately reflect the actual conditions of each place and effectively respond to the challenges brought by climate factors, ensuring the accuracy of research data and the practicality of results.

[0092] The forest phenology and environmental disturbance detection method also includes: preprocessing the forest remote sensing time series image data.

[0093] The preprocessing of forest area remote sensing time series image data includes: radiation and atmosphere correction, cloud and shadow removal, and geometric correction.

[0094] Preprocessing aims to improve data quality and provide a solid foundation for subsequent analysis.

[0095] S2: Based on the remote sensing time series image data of the forest area, a time series fully convolutional neural network model is constructed based on convolutional neural network and temporal attention mechanism.

[0096] The time series fully convolutional neural network model is dedicated to improving the recognition accuracy of vegetation cover changes and environmental disturbance events, and emphasizes the ability to locate disturbances in space and time. The model consists of three main parts: first, the spatial encoding module uses deep convolutional layers to extract the spatial features of the image; second, the temporal encoding module introduces an attention mechanism to identify key change moments in the time series and enhance sensitivity to temporal changes; finally, the spatial decoding module is responsible for accurately reconstructing the spatial details of the semantic segmentation results while maintaining high spatial resolution.

[0097] The time series full convolutional neural network model based on the convolutional neural network and the time attention mechanism is constructed as follows:

[0098] S21: Use the layered deep convolutional network architecture as the spatial encoder. By setting the convolution layer, ReLU activation function and batch normalization layer step by step, combined with the strided convolution module, the feature extraction is deepened layer by layer to extract the spatial features of the image and obtain the feature map stacked along the time axis and processed by spatial encoding.

[0099] A layered deep convolutional network architecture is used as a spatial encoder to extract spatial features of remote sensing images. By setting convolutional layers, ReLU activation functions, and batch normalization layers step by step, combined with the step convolution module, feature extraction is deepened layer by layer, effectively reducing feature dimensions and compressing remote sensing data information, providing effective features for the temporal encoding and spatial decoding stages.

[0100] In view of the inherent complexity of time series data, a strategy of stacking feature maps processed by spatial encoding along the time axis is adopted, so that the model can comprehensively utilize the information of the entire time series in subsequent analysis, effectively identify and capture various vegetation cover change patterns from seasonal changes to sudden environmental disturbances, enhance the model's ability to analyze remote sensing image time series data, and provide reliable data support for accurately identifying and locating vegetation changes and environmental disturbances.

[0101] S22: Integrate the temporal attention encoder and apply the temporal attention mechanism to assign different weights to the feature maps at each time point in the sequence, achieve efficient encoding of the dataset in the temporal dimension, and identify the temporal dynamic characteristics of the forest area.

[0102] Temporal encoding is a key step in step S4, which aims to accurately capture and analyze the dynamic changes of remote sensing time series data. By integrating the temporal attention encoder and applying the temporal attention mechanism, the analysis of feature sequences is optimized. This mechanism enhances the model's ability to identify key time points by assigning different weights to the feature graphs of each time point in the sequence, while achieving effective compression of redundant time information and improving processing efficiency.

[0103] In the implementation, the method of combining the temporal attention encoder with the multi-scale analysis strategy is adopted to achieve efficient encoding of the time dimension on remote sensing datasets with uniform spatial resolution characteristics. This strategy aims to significantly improve the model's ability to identify and analyze the intrinsic changes in time series data by comprehensively utilizing various time scales, comprehensively capture periodic phenological changes and short-term environmental disturbances in time series data, and optimize the model's effectiveness in accurately identifying and locating vegetation cover changes and environmental disturbance events.

[0104] S23: Build a model based on U-Net, use layered strided transposed convolution technology to gradually improve the spatial resolution of feature maps, introduce skip connection technology and attention mechanism, fuse feature maps with temporal dynamic features, and reconstruct the spatial details of the original remote sensing time series image data.

[0105] The spatial decoding stage adopts a design that is symmetrical to the spatial encoding stage. By using the layered strided transposed convolution technology to gradually improve the spatial resolution of the feature map, the spatial details of the original remote sensing time series data can be accurately reconstructed. At the same time, by introducing the jump connection technology, the deep features extracted in the spatial encoding stage and the temporal dynamic features identified in the temporal encoding stage are effectively integrated, the model's ability to capture spatial details and the comprehensive expression of temporal information is optimized, and the recognition accuracy of vegetation cover changes and environmental disturbances in spatiotemporal distribution information is enhanced.

[0106] This step builds a model based on U-Net. U-Net is chosen because its symmetrical design and feature fusion capabilities have been proven in medical image processing and remote sensing image analysis. This step customizes the U-Net architecture, including optimizing the transposed convolution layer to better restore high-resolution spatial information, and adjusting the jump connection strategy to promote the fusion of spatial and temporal dimension features. In addition, considering the characteristics of remote sensing time series data, the attention mechanism is introduced to enhance the model's ability to recognize key spatiotemporal change features, and improve the model's ability to identify and accurately locate vegetation cover changes and environmental disturbance events.

[0107] S3: Based on the remote sensing time series image data of the forest area, a sample data set is constructed, and the time series fully convolutional neural network model is trained to obtain the forest phenology and environmental disturbance detection model.

[0108] When constructing the sample data set, the image data is enhanced and the sample data set is constructed based on the time window sliding technology.

[0109] The sample data construction first generates a sample set reflecting the change spots based on the existing remote sensing images and spot data. The calibrated samples include the vegetation cover type before and after and the change type samples.

[0110] Using two phases of remote sensing images from the same sensor in adjacent quarters, vegetation cover remote sensing image change calibration is performed by province or region based on remote sensing image features and vegetation cover changes. Based on the vegetation cover type reflected by the attributes of fixed sample plots or forest map patches, the remote sensing image features of the location are used for comparative analysis, and representative plots with relatively consistent remote sensing image features are selected. The vegetation cover type before and after, remote sensing data source, remote sensing data acquisition time, spatial resolution and other attribute information are calibrated according to a 512m×512m square vector frame.

[0111] Based on the established remote sensing images and data labels of each vegetation cover type, combined with the remote sensing images and forest and grass resource data of the previous and next two periods, and referring to the attribute information of forest and grass resource spots, the characteristic change areas of the remote sensing images of the previous and next two periods are compared and determined, and the change type and the corresponding remote sensing data source, remote sensing data acquisition time, spatial resolution and other attribute information are calibrated according to the 512m×512m square vector frame. Specific change types include but are not limited to construction project use, logging, reclamation, disaster events, and ecological protection and restoration measures.

[0112] This step aims to establish a large-scale, diverse and spatially distributed sample database to support iterative training and parameter optimization of deep learning models. The goal is to increase the training accuracy of the model to more than 85%. Models that do not meet this accuracy standard will be inspected in detail, supplemented with labels for interpretation signs and change types, and training parameters will be adjusted and iterative training will be repeated to achieve higher accuracy.

[0113] By carefully dividing the pre-processed time series remote sensing data, we create training and validation sets that reflect the diversity of vegetation coverage and environmental changes. At the same time, we use data enhancement methods such as random rotation, flipping, and cropping to enhance the model's generalization ability for diverse surface features. The training samples constructed based on the time window sliding technology not only capture the vegetation coverage information at a single moment, but also integrate the change trend on the time axis, providing the model with dynamic change information on vegetation coverage.

[0114] The training of the time series fully convolutional neural network model includes:

[0115] The cross entropy loss function is selected to handle pixel-level classification tasks, and the Dice loss is introduced to improve the recognition ability of small objects and balance the problem of uneven categories;

[0116] Use Adam optimizer to improve training efficiency and model convergence speed;

[0117] Implement learning rate decay to achieve fast learning in the early stages of training and maintain stable convergence in the later stages;

[0118] Use early stopping mechanism to prevent overfitting;

[0119] Comprehensively evaluate the model based on multiple indicators on the validation set, including but not limited to precision, recall, and F1 score;

[0120] Based on the comprehensive evaluation results, the network model is fine-tuned.

[0121] Model training is the key to achieving accurate semantic segmentation based on time series remote sensing data, which includes multiple links such as loss function design, optimizer selection and learning rate management.

[0122] The cross entropy loss function is selected to handle pixel-level classification tasks, and the Dice loss is introduced to improve the recognition ability of small targets and balance the problem of uneven categories. The Adam optimizer is selected to improve training efficiency and model convergence speed. By implementing learning rate decay, the model can achieve rapid learning in the early stage of training and maintain stable convergence in the later stage. The early stopping mechanism can prevent overfitting of the model.

[0123] Through comprehensive evaluation based on key indicators such as precision, recall rate and F1 score on the validation set, the efficiency and accuracy of the model in high-precision semantic segmentation tasks are ensured.

[0124] During the performance evaluation, the segmentation prediction results of the model were compared with the field verification data to further explore the performance of the model in capturing seasonal changes and identifying disturbance events. Based on the evaluation results, this study fine-tuned the network architecture, refined the temporal attention mechanism, and adjusted the training strategy to enhance the model's ability to process remote sensing time series data and improve the accuracy of semantic segmentation, thereby enhancing the model's ability to reveal the spatial and temporal distribution characteristics of disturbances.

[0125] S4: Input the forest area remote sensing time series image data into the forest area phenology and environmental disturbance detection model to obtain the forest area phenology and environmental disturbance detection results.

[0126] This method focuses on two types of change detection in forest areas. First, according to the vegetation coverage type classification system of the "National Technical Regulations for Comprehensive Monitoring of Forest and Grass Ecology", combined with the specific vegetation coverage types of the project technology integration demonstration base, the change patches within the vegetation coverage are extracted; second, according to the remote sensing interpretation and verification change types in the "National Technical Regulations for Comprehensive Monitoring of Forest and Grass Ecology", the change types of the change patches obtained in the previous step are extracted, including construction project use, logging, reclamation, disasters, and ecological protection and restoration measures.

[0127] The goal of the detection model output of this method is to generate a feature map with the same resolution as the input image, in which each pixel is carefully classified according to its perturbed state. This classification operation is implemented through the spatial decoder inside the model, which maps the high-dimensional features to a predefined category space through a 1x1 convolutional layer to achieve accurate recognition of the surface state.

[0128] With the help of the model's accurate classification of each pixel, this method focuses on determining the specific time and location of the disturbance event.

[0129] S5: Based on the detection results of forest phenology and environmental disturbance, construct the spatiotemporal distribution map of forest phenology and environmental disturbance.

[0130] After obtaining the forest phenology and environmental disturbance detection results using the forest phenology and environmental disturbance detection model, these data are integrated to construct a spatiotemporal distribution map showing the geographical distribution of the disturbance and its time process, providing important information for environmental monitoring and decision-making analysis.

[0131] The beneficial effects of this embodiment are: this method can also conduct a comprehensive evaluation of the network model based on multiple indicators, and fine-tune and improve the network model according to the evaluation results; according to the forest phenology and environmental disturbance detection results, the data is integrated to construct the spatiotemporal distribution map of the forest phenology and environmental disturbance, which provides important information for environmental monitoring and decision-making analysis.

[0132] Embodiment 3

[0133] A forest phenology and environmental disturbance detection system, based on the forest phenology and environmental disturbance detection method as described in Embodiment 1 or 2, comprises:

[0134] A data acquisition module is used to acquire remote sensing time series image data of forest areas;

[0135] The model building module is used to build a time series full convolutional neural network model based on the convolutional neural network and the temporal attention mechanism according to the forest area remote sensing time series image data;

[0136] The training module is used to construct a sample data set based on the remote sensing time series image data of the forest area, train the time series full convolutional neural network model, and obtain the forest phenology and environmental disturbance detection model;

[0137] The detection module is used to input the forest area remote sensing time series image data into the forest area phenology and environmental disturbance detection model to obtain the forest area phenology and environmental disturbance detection results.

[0138] Preferably, the system further comprises:

[0139] A preprocessing module is used to preprocess the remote sensing time series image data of forest areas;

[0140] The distribution map construction module is used to construct the spatiotemporal distribution map of forest phenology and environmental disturbance based on the detection results of forest phenology and environmental disturbance.

[0141] Embodiment 4

[0142] A computer program which, when executed, implements the steps of the forest phenology and environmental disturbance detection method as described in Example 1 or 2.

[0143] Embodiment 5

[0144] A readable storage medium stores a computer program as described in Example 4, and when the computer program is executed, the steps of the forest phenology and environmental disturbance detection method as described in Example 1 or 2 are implemented.

[0145] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by any technician familiar with the art within the technical scope disclosed in the present invention should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be based on the protection scope of the claims.

Claims

1. A method for detecting forest phenology and environmental disturbance, characterized in that: The following steps are involved: Obtain remote sensing time series image data of forest areas; According to the remote sensing time series image data of forest areas, a time series full convolutional neural network model is constructed based on convolutional neural network and temporal attention mechanism; Based on the remote sensing time series image data of forest areas, a sample data set was constructed, and the time series fully convolutional neural network model was trained to obtain the forest phenology and environmental disturbance detection model; Input the forest area remote sensing time series image data into the forest area phenology and environmental disturbance detection model to obtain the forest area phenology and environmental disturbance detection results; The method of constructing a time series full convolutional neural network model based on a convolutional neural network and a temporal attention mechanism includes: Using the layered deep convolutional network architecture as the spatial encoder, by setting the convolution layer, ReLU activation function and batch normalization layer step by step, combined with the stride convolution module, the feature extraction is deepened layer by layer, the spatial features of the image are extracted, and the feature map stacked along the time axis and processed by spatial encoding is obtained; Integrate the temporal attention encoder and apply the temporal attention mechanism to assign different weights to the feature maps at each time point in the sequence, achieve efficient encoding of the dataset in the temporal dimension, and identify the temporal dynamic characteristics of the forest area; The model is built based on U-Net, and the spatial resolution of the feature map is gradually improved by using the layered strided transposed convolution technology. The skip connection technology and attention mechanism are introduced to fuse the feature map with the temporal dynamic features to reconstruct the spatial details of the original remote sensing time series image data. The training of the time series fully convolutional neural network model includes: The cross entropy loss function is selected to handle pixel-level classification tasks, and the Dice loss is introduced to improve the recognition ability of small objects and balance the problem of uneven categories; Use Adam optimizer to improve training efficiency and model convergence speed; Implement a learning rate decay strategy to achieve fast learning in the early stages of training and maintain stable convergence in the later stages; Use early stopping mechanism to prevent overfitting; Comprehensively evaluate the model based on multiple indicators on the validation set, including precision, recall, and F1 score; Based on the comprehensive evaluation results, the network model is fine-tuned.

2. The method for detecting forest phenology and environmental disturbance according to claim 1, characterized in that: The forest phenology and environmental disturbance detection method also includes: preprocessing the forest remote sensing time series image data.

3. The method for detecting forest phenology and environmental disturbance according to claim 2, characterized in that: The preprocessing of forest area remote sensing time series image data includes: radiation and atmosphere correction, cloud and shadow removal, and geometric correction.

4. The method for detecting forest phenology and environmental disturbance according to claim 1, characterized in that: When constructing the sample data set, the image data is enhanced and the sample data set is constructed based on the time window sliding technology.

5. The method for detecting forest phenology and environmental disturbance according to claim 1, characterized in that: The forest phenology and environmental disturbance detection method also includes the following steps: Based on the detection results of forest phenology and environmental disturbance, a spatiotemporal distribution map of forest phenology and environmental disturbance was constructed.

6. A forest phenology and environmental disturbance detection system, characterized in that: The method for detecting forest phenology and environmental disturbance according to any one of claims 1 to 5 comprises: A data acquisition module is used to acquire remote sensing time series image data of forest areas; The model building module is used to build a time series full convolutional neural network model based on the convolutional neural network and the temporal attention mechanism according to the forest area remote sensing time series image data; The training module is used to construct a sample data set based on the remote sensing time series image data of the forest area, train the time series full convolutional neural network model, and obtain the forest phenology and environmental disturbance detection model; The detection module is used to input the forest area remote sensing time series image data into the forest area phenology and environmental disturbance detection model to obtain the forest area phenology and environmental disturbance detection results; The method of constructing a time series full convolutional neural network model based on a convolutional neural network and a temporal attention mechanism includes: Using the layered deep convolutional network architecture as the spatial encoder, by setting the convolution layer, ReLU activation function and batch normalization layer step by step, combined with the stride convolution module, the feature extraction is deepened layer by layer, the spatial features of the image are extracted, and the feature map stacked along the time axis and processed by spatial encoding is obtained; Integrate the temporal attention encoder and apply the temporal attention mechanism to assign different weights to the feature maps at each time point in the sequence, achieve efficient encoding of the dataset in the temporal dimension, and identify the temporal dynamic characteristics of the forest area; The model is built based on U-Net, and the spatial resolution of the feature map is gradually improved by using the layered strided transposed convolution technology. The skip connection technology and attention mechanism are introduced to fuse the feature map with the temporal dynamic features to reconstruct the spatial details of the original remote sensing time series image data. The training of the time series fully convolutional neural network model includes: The cross entropy loss function is selected to handle pixel-level classification tasks, and the Dice loss is introduced to improve the recognition ability of small objects and balance the problem of uneven categories; Use Adam optimizer to improve training efficiency and model convergence speed; Implement a learning rate decay strategy to achieve fast learning in the early stages of training and maintain stable convergence in the later stages; Use early stopping mechanism to prevent overfitting; Comprehensively evaluate the model based on multiple indicators on the validation set, including precision, recall, and F1 score; Based on the comprehensive evaluation results, the network model is fine-tuned.

Images