22 results about "Forest classification" patented technology

The invention discloses an island shoreline identification method and system based on artificial intelligence, and relates to the technical field of shoreline change monitoring. According to the invention, through fusion modeling of the optical image, the SAR image and the topographic data, the limitation of a single data source is broken through; a vegetation coverage area and a non-vegetation shoreline can be accurately distinguished through the multispectral characteristics of the optical image, and the problem of monitoring blind areas in scenes such as cloud and mist and nighttime is solved through the all-weather penetrating power of the SAR image; a random forest classifier can effectively distinguish bedrock steep cliff, sandy sand beach and the like in combination with superpixel unit feature vectors constructed by topographic data, and accurate partitioning of a target island shoreline is completed; moreover, a dynamic global correction mechanism further eliminates space-time error accumulation, a standard deviation mean value is calculated through cross-season historical data, accidental errors such as typhoon and rainstorm are filtered, a future global correction coefficient is predicted in combination with a time sequence model, and the reliability of long-term monitoring is remarkably improved.

The invention discloses a terrain classification method based on a random forest algorithm, a server and a storage medium, and belongs to the technical field of terrain classification, and the method comprises the steps: obtaining DEM data and remote sensing image data of a research area, and extracting terrain features, spectral features, index features and texture features; designing a feature combination scheme, and generating an optimal feature combination data set as input data of terrain classification; calculating an optimal parameter combination of a random forest algorithm, and generating an optimal random forest classifier; establishing a terrain classification system suitable for the research area, and constructing a terrain classification training sample set and a verification sample set; performing terrain classification by using a random forest classifier and the optimal feature combination data set to obtain a terrain classification result of the research area; and calculating and evaluating the precision of a terrain classification result by utilizing a verification sample set and a Kappa coefficient evaluation method. By adopting the method, the terrain classification refinement degree and the terrain classification calculation efficiency and classification efficiency can be improved.

The invention relates to a landslide susceptibility evaluation method considering InSAR and a multi-stage optimization random forest model. According to the method, non-landslide points are selected through multi-factor control and spatial stratified sampling, feature factors are selected by using a random forest classifier in combination with a Pearson's correlation coefficient matrix and a VIF method, a random forest model is optimized based on a Bayesian algorithm, finally, the performance of the model is evaluated through multiple indexes, and susceptibility indexes are divided into five grades. Compared with the prior art, the method not only takes the earth surface deformation rate identified by InSAR as a characteristic factor, but also performs quantitative analysis on the characteristic factor and the susceptibility zoning result, verifies the consistency of the two factors, can effectively improve the precision of a landslide susceptibility evaluation model, provides important reference for landslide disaster prevention and reduction research, and has a wide application prospect. The method is especially suitable for areas with frequent geological disasters such as southwest regions in China.

The invention provides a crown vertical structure feature inversion method and a large-area forest classification method and system, and belongs to the field of forest ecological environment monitoring. The inversion method comprises the steps of collecting data of two different years in a long time sequence, speculating canopy height spatial distribution diagrams respectively, extracting pixels with increased canopy height as a height sample set, taking the rest as a feature sample set, and obtaining forest canopy height data; spectral features and spatial texture features are obtained; taking the year with forest disturbance as a growth and development starting point, and obtaining the canopy height, spectral reflectivity and spatial texture of each pixel from the growth and development starting point to the last year; respectively extracting annual average growth speed characteristics in the last year or when the spectral reflectivity of the canopy is saturated, and representing the vertical structure characteristics of the canopy by using the height characteristics and the growth speed characteristics of the canopy; according to the classification method, the horizontal space-time spectrum features and the vertical structure features are fused for forest classification. According to the invention, the accuracy and precision of large-area forest classification are improved.

The embodiment of the invention discloses a camellia oleifera forest classification method, device and equipment and a medium. The method comprises the steps that a remote sensing image and topographic data of a detection area are acquired, the remote sensing image comprises time sequence spectral data, and the time sequence spectral data represents spectral characteristics of the earth surface of the detection area; based on a pre-trained classification model, first classification information of the camellia oleifera forest in the remote sensing image is determined according to the remote sensing image, phenological characteristics and topographic data, the phenological characteristics represent the phenological stage of the camellia oleifera, and the phenological stage is determined based on the growth cycle of the camellia oleifera; the classification model is obtained by training a data set formed by combining a remote sensing image sample corresponding to the sampling area with topographic data and phenological characteristics; determining a target area in the remote sensing image based on the first classification information, determining a first classification parameter of the target area based on the time sequence spectrum data of the target area and the weight of the phenological feature, and determining second classification information of the target area based on the first classification parameter. According to the technical scheme provided by the invention, the classification capability and the classification accuracy of the camellia oleifera forest can be improved.

The invention discloses an intertidal zone identification method and related equipment, and belongs to the technical field of remote sensing identification, and the method comprises the steps: obtaining a remote sensing image of a target area; fitting different wavebands of the remote sensing image based on the pixels to obtain a time sequence fitting curve; based on the time sequence fitting curve, selecting a long time sequence training sample and a verification sample for training a random forest classification model; and performing classification identification on the remote sensing image by using the trained random forest classification model to obtain an intertidal zone identification result of the target area. According to the method, the long-time sequence samples are extracted on the basis of taking the pixels as units, the intertidal zone range is extracted by fully applying the information between the long-time sequence images, and the efficiency, precision and flexibility of extracting the intertidal zone information of the long-time sequence images are effectively improved.

The invention relates to a leaf area index time sequence processing method, which comprises the following steps of: inputting a time sequence remote sensing image and suburb forest and economic forest classification data of the same region; calculating a normalized differential vegetation index of the research area and carrying out time sequence sorting; calculating vegetation coverage data of the research area and carrying out time sequence sorting; synthesizing to obtain monthly FVC time sequence data of the suburb forest region and the economic forest region; calculating a suburb forest leaf area index LAI in the research area and performing time sequence sorting; calculating the economic forest leaf area index LAI of the research area and performing time sequence sorting; performing seasonal decomposition on the suburb forest time sequence data after adaptive filtering; performing seasonal decomposition on the economic forest time series data after adaptive filtering; merging the suburb forest time sequence data and the economic forest time sequence data; and outputting to obtain final optimized data. The invention further relates to a leaf area index time sequence processing system. According to the invention, a purification LAI time sequence product which clearly represents long-term trend, mutation and gradual change signals of respective vegetation canopy structures can be output.

This invention discloses a method and system for identifying planted and natural forests based on multi-source remote sensing and spatiotemporal joint optimization, belonging to the fields of remote sensing image processing, earth observation, and dynamic monitoring of forest ecology. Addressing the problems of high label noise and disordered temporal logic in existing technologies, this invention obtains high-purity training samples through multi-stage iterative purification by fusing CCDC mutation history and deep semantic features; obtains pixel-level confidence through independent initial classification by geographic partitioning; constructs an HMM model integrating ecological prior constraints and dynamic adaptive emission probabilities for Viterbi decoding; and performs constrained spatial restoration with cross-dimensional temporal security checks. This invention effectively suppresses temporal pseudo-variables and salt-and-pepper noise, achieving high-resolution thematic mapping year by year with no temporal logical contradictions and regular spatial patches, suitable for fine-grained forest classification and compliance monitoring of zero-deforestation laws in tropical and subtropical regions.

The application provides a forest canopy vertical structure feature inversion method, a large-area forest classification method and system, and belongs to the field of forest ecological environment monitoring. The inversion method collects two different year data in a long time sequence, respectively infers a canopy height spatial distribution map, extracts a pixel with increased canopy height as a height sample set, and extracts the rest as a feature sample set, to obtain forest canopy height data; then, spectral features and spatial texture features are obtained; the year when forest disturbance occurs is taken as a growth starting point, and the canopy height, spectral reflectance and spatial texture of each pixel from the growth starting point to the last year are obtained; the average growth speed feature of the last year or when the canopy spectral reflectance reaches saturation is extracted, and the canopy height feature and the growth speed feature are used to represent the canopy vertical structure feature; and the classification method fuses horizontal space-time spectrum features and vertical structure features to perform forest classification. The application improves the accuracy and precision of large-area forest classification.

This application presents a method for classifying land features along railway lines using remote sensing imagery based on morphological entropy fusion. The method includes: acquiring Sentinel-1 radar imagery, Sentinel-2 optical imagery, and a fused image of the target area from a GEE platform, and preprocessing the data; generating masks on the Sentinel-1, Sentinel-2, and fused images respectively, and calculating and extracting circumferential morphological entropy, radial morphological entropy, and joint morphological entropy; extracting polarization and texture features from the Sentinel-1 imagery, and spectral and exponential features from the Sentinel-2 imagery, and constructing corresponding composite features on the fused imagery; combining these features with morphological entropy features from corresponding data sources; training a random forest classification model based on each feature set, and using the trained model to classify four typical land features along the railway line: buildings, railways, vegetation, and highways. This method, by fully utilizing multi-source remote sensing information, effectively characterizes the structural complexity and irregularity of land feature outlines through morphological entropy, significantly improving the accuracy and reliability of the classification results.

The invention relates to an urban vehicle-mounted point cloud forest tree classification method based on RoPE operator local feature aggregation, and belongs to the technical field of photogrammetry and remote sensing. According to the method, the ROPE operator technology is adopted, the relative coordinates and the original features of all points in the neighborhood are coded to establish the weight relation, a model is helped to distinguish the fine features of the trunk, the coding operator is utilized to carry out secondary local feature dynamic aggregation, the local structure and context information of the forest are captured, and the problem that the classification precision of the trunk and the shielding forest is low is solved; meanwhile, a U-shaped encoder-decoder network architecture and a jump connection strategy are introduced, forest features are fused among different levels, and multi-scale forest information is obtained. Compared with other methods, the method can solve the problem that tree classification errors of tree trunks and shelters often occur in urban tree classification, and realizes accurate segmentation of complete tree information in a complex urban scene.

The invention discloses a memory fault maintenance system and method based on a random forest model, electronic equipment, a storage medium and a computer program product, and relates to the technical field of fault diagnos.The memory fault maintenance system comprises a multi-source data collection module used for collecting multi-source data of a memory in real time, the multi-source data comprises hardware signals, software logs and performance index data; the random forest classification unit is connected with the multi-source data acquisition module, comprises a trained random forest model and is used for carrying out fault identification on the multi-source data through the trained random forest model to obtain fault information of the memory; and the intelligent maintenance unit is used for determining a repair strategy according to the fault information and executing the repair strategy so as to carry out fault maintenance on the memory.

The present invention discloses a drought monitoring and early warning method, system, and equipment based on changes in characteristic water areas. First, a water body precision extraction technology based on random deep forest classification and a comprehensive threshold segmentation model is used to automatically and accurately extract surface water areas in the target area. Then, characteristic water areas are identified and separated based on prior knowledge and area thresholds, and the changes in the area of ​​characteristic water areas over multiple years are monitored in real time. Finally, drought early warnings and comprehensive assessments are performed based on the dynamic changes in the characteristic water areas. When the area of ​​characteristic water areas decreases significantly during the monitoring period, a conclusion is drawn that drought may occur based on the actual situation and a drought early warning is issued. The present invention directly uses surface water as an indicator and factor to characterize the drought state, which is simpler than other models and methods and can more intuitively reflect the development and evolution of droughts. The present invention performs real-time drought monitoring based on changes in the area of ​​characteristic water areas, and can more accurately and timely perform drought early warnings and comprehensive assessments.

This invention, entitled "A Method, System, and Storage Medium for Remote Sensing Estimation of Carbon Storage Based on Forest Heterogeneity," belongs to the field of carbon storage estimation technology. The technical problem it aims to solve is the poor accuracy of regional carbon storage estimation due to insufficient ground sample quantity and uneven spatial distribution. Key technical points include: S1, expanding the limited ground carbon storage samples using spaceborne lidar photon point cloud data to construct a ground-space fusion forest carbon storage measured sample library; S2, interpreting remote sensing data of the target area's forest ecosystem and classifying forest vegetation types, considering forest carbon sink heterogeneity and employing continuous change detection and classification algorithms for forest vegetation type classification; S3, constructing a carbon storage remote sensing estimation model coupling forest classification and multidimensional features; and S4, performing carbon storage estimation based on the carbon storage remote sensing estimation model.

The invention discloses a remote control manipulator system based on surface electromyogram signals, and relates to the field of remote control robot manipulators. The system comprises an sEMG collecting and processing module used for collecting multi-channel electromyographic signals at the forearms of a human body through a metal dry electrode array, conducting multi-stage amplification and filtering and converting the multi-channel electromyographic signals into digital signals; the wireless transmission and upper computer module is used for transmitting the processed digital signal sequence to an upper computer through a wireless serial port, performing preprocessing and feature extraction on the sEMG signal, and then identifying a predefined gesture category by using a random forest classification model; and the manipulator remote control module is used for remotely transmitting the gesture classification prediction result to the manipulator control unit in real time through the LoRa module so as to remotely control the manipulator to execute the corresponding action. The invention provides a visual and natural manipulator remote control scheme with strong anti-interference capability, and is particularly suitable for the fields of artificial limb control, dangerous environment operation and the like.

The invention provides an object-oriented natural forest and artificial forest classification method, and relates to the field of forest classification and recognition research, and the method comprises the steps: carrying out the superpixel segmentation of a preprocessed Sentinel-2 image, and obtaining an optimal segmentation scale; obtaining a multi-source remote sensing product; the method comprises the following steps: based on a Sentinel-2 image and a multi-source remote sensing product, taking an object as a basic unit, and extracting object-oriented multi-type features; performing feature optimization on the multi-type features; classifying the optimized features by applying a random forest algorithm and a support vector algorithm respectively, comparing the classification precision of the two methods, and determining an optimal classification model; obtaining a to-be-classified remote sensing image; and inputting a remote sensing image to be classified into the trained optimal classification model to obtain an optimal classification result of the natural forest and the artificial forest. According to the technical scheme, the problem that natural forests and artificial forests are difficult to classify is solved.

The invention discloses an electromagnetic wave signal random forest classification method and device, equipment and a medium. The method comprises the following steps: acquiring and preprocessing an original electromagnetic wave time domain discrete signal sequence to obtain preprocessed signal data; based on the preprocessed signal data, positive and negative pulse event data are extracted through an adaptive threshold pulse detection algorithm; the positive and negative pulse event data comprises a positive pulse number, a negative pulse number and a total pulse number; determining a polarity characteristic, an average frequency characteristic and an average pulse width characteristic of the preprocessed signal data according to the positive and negative pulse event data; forming a feature vector by the total pulse number, the polarity feature, the average frequency feature and the average pulse width feature; and inputting the feature vector into a pre-trained random forest classification model, and outputting a classification result of the electromagnetic wave signal.

One example method includes receiving a random forest classifier model that comprises a group of decision trees, wherein the random forest classifier model is created using a vertical federated framework, providing new observations, not included in a set of training observations, to a trained random forest classifier model, wherein the random forest classifier model is trained in the vertical federated framework, and wherein the training is performed using the set of training observations as input to the random forest classifier model, and generating, by the trained random forest classifier model, one or more diversity scores pertaining to the new observations.

The invention relates to a vegetation coverage time sequence processing method, which comprises the following steps of: inputting a time sequence remote sensing image and suburb forest and economic forest classification data of the same region; calculating a normalized differential vegetation index of the research area and carrying out time sequence sorting; calculating vegetation coverage data of the research area and carrying out time sequence sorting; synthesizing to obtain monthly FVC time sequence data of the suburb forest region and the economic forest region; adaptive filtering is carried out on the suburb forest time sequence data; carrying out seasonal decomposition based on a fixed period on the filtered suburb forest time sequence data; filtering the economic forest time series data; carrying out adaptive seasonal decomposition on the filtered economic forest time series data; combining the suburb forest time series data and the economic forest time series data obtained through decomposition; and outputting the merged time series data to obtain final optimized data. The invention also relates to a vegetation coverage time sequence processing system. According to the method, a reliable data basis can be provided for high-precision change detection and attribution analysis.

The application discloses a PUE prediction method and device of a data center and a storage medium, relates to the field of data center energy consumption, and is used for accurately and quickly predicting the PUE of the data center. The method comprises the following steps: inputting a to-be-predicted PUE sample at a first moment into a deep forest classifier for sample analysis, obtaining an analysis result, wherein the analysis result comprises N probability values, each probability value is used for representing the probability that the to-be-predicted PUE sample belongs to one of N preset categories, and N is a positive integer; and training an RDPG model based on the analysis result, a first historical PUE data set and environmental data at the first moment, obtaining a trained RDPG model, and the trained RDPG model is used for predicting the PUE value of the data center.

This application relates to the field of computer technology and provides a method, device, electronic device, and storage medium for identifying target merchants. It constructs a merchant network graph by using merchants in a transaction dataset as nodes and their common transaction objects as edges. The merchant network graph is then divided into multiple communities based on a modularity algorithm. Each community is input into a distributed representation learning model to obtain a graph structure feature vector for each community. This vector is then concatenated with the transaction features of the merchants to obtain a concatenated result. This concatenated result is then input into an isolated forest classification model to identify unknown target merchant communities. This application's embodiments can identify intermediary merchants and / or individuals involved in abnormal transactions without requiring any black tags, and can also identify unknown abnormal transactions, uncovering uncovered high-risk merchants, thereby further improving the management level of the transaction system.

This invention provides an environmental quality detection method and apparatus, relating to the fields of image data processing and environmental quality detection, to address problems such as low detection accuracy and limitations. The method includes: preprocessing a hyperspectral remote sensing image to obtain an enhanced image; performing dimensionality reduction on the enhanced image using principal component analysis to obtain a first feature map; calculating the attention score of the first feature map, which is used to highlight target features in the dimensionality-reduced feature map; extracting features from the dimensionality-reduced feature image based on the attention score to obtain an attention feature map; performing multi-scale convolution on the first feature map to obtain second feature maps at different scales, and then performing a weighted summation to obtain a third feature map; fusing the attention feature map and the third feature map, and inputting the resulting fourth feature map into a random forest classification model to predict the quality detection result, wherein the weight of the decision tree that can accurately classify minority class samples in the random forest classification model is greater than the weight of other decision trees.