36 results about "Forest vegetation" patented technology

The invention relates to the technical field of radar echo denoising, in particular to a radar echo signal filtering noise self-adaptive suppression method and a radar echo signal filtering noise self-adaptive suppression system. Acquiring radar station position and beam pointing data, and synchronously acquiring forest vegetation data and environment interference data of a beam pointing target area to form radar environment data; the radar environment data are identified through the multi-band decision model, and multi-modal waveform parameters are obtained; based on the multi-modal waveform parameters, radar equipment is controlled to perform periodic detection on the target area, and radar echo data are obtained; the method comprises the steps of identifying radar echo data and radar environment data, extracting dynamic spectrum characteristics in the radar echo data, performing clutter suppression processing on the dynamic spectrum characteristics by using the radar environment data, calculating the confidence probability that an early fire exists in a target area, and executing space-time consistency joint verification; according to the invention, fire early warning is carried out through the confidence probability of the fire.

The invention discloses an identification and treatment method for a water and soil loss priority treatment area of a small watershed slope in a hilly area. The method comprises the following steps: calculating a small watershed soil erosion modulus and a sediment connectivity index, and carrying out geospatial modeling to form a soil erosion and connectivity partition identification priority treatment area; determining the land utilization type and terrain status of water and soil loss, and dividing functional zones from the spatial angles of an upper slope, a middle slope and a lower slope; a water and soil conservation forest is built in a forest and grass vegetation area on the slope or an abandoned cultivated land above a reclamation prohibition slope and a water and soil loss treatment area on the barren mountain and slope according to the principle of suitable land and suitable trees, and water and soil loss is prevented and controlled through natural restoration; water and soil loss prevention and control measures of slope land reclamation-hydrological network optimization and node control-grassland soil fixation and fertility increasing are adopted in a slope water and soil loss treatment area below the reclamation forbidding gradient in the slope; the slope bottom is provided with a plant filter belt, a tail end grit chamber and an ecological ditch measure for desilting and pollution reduction; the problem that a water and soil conservation treatment system in a small watershed is low in treatment degree is solved.

The invention belongs to the technical field of ecological hydrology, and particularly discloses a forest underlying surface earth surface evapotranspiration estimation and component segmentation method based on kNDVI. According to the method, based on a maximum entropy increase model, under the constraint of a surface energy balance equation, the surface state parameters of a forest vegetation-free area are used for resolving to obtain the soil evaporation capacity of the bare soil; on the basis of a maximum entropy increase model, under the constraint of a canopy energy balance equation, the vegetation transpiration amount is obtained through calculation according to the canopy state parameters of the forest vegetation coverage area; calculating the canopy interception evaporation capacity according to the leaf area index and the rainfall capacity of the forest vegetation coverage area; and finally, based on the kNDVI index, carrying out weighted summation on the bare soil evaporation amount, the vegetation evaporation amount and the canopy interception evaporation amount in the forest to obtain the total evapotranspiration of the underlying surface of the forest. Compared with an existing evapotranspiration estimation model, the forest underlying surface earth surface evapotranspiration estimation method is lower in complexity and higher in estimation precision.

The application discloses an under-forest goose breeding ecological circulation planting and breeding system, which comprises a first iron fence, fixing columns, double doors, a door frame, an opening and closing mechanism, a water storage tank, a dropper, a water receiving tank, a fixing plate, a fence, a second iron fence and a fixing shaft, the first iron fence and the second iron fence respectively enclose two planting and breeding circles, and a plurality of fixing columns are evenly arranged on the inner sides of the first iron fence and the second iron fence respectively. The application adopts the mode of breeding geese under fruit trees, uses goose excrement as fertilizer for fruit trees, and provides food for geese with the under-forest vegetation, which is beneficial to saving breeding cost, simultaneously producing fruits and geese, and improving breeding economic benefits. Meanwhile, the double-circle design is adopted, which avoids excessive damage to the vegetation caused by excessive grazing, and is beneficial to realizing sustainable breeding. Through the setting of the opening and closing mechanism, the double doors are automatically opened and closed to the two sides, and manual opening and closing of the doors is not needed, thereby improving work efficiency.

The invention belongs to the technical field of forest fire monitoring and early warning, and particularly relates to an extreme climate prediction method based on a large warm period weighted FWI index and application. According to the specific technical scheme, the method comprises the following steps: acquiring climate data of a great warm period, and calculating FWI index components in each time period according to the climate data of the great warm period; the method comprises the following steps: acquiring forest vegetation data and topographic data by utilizing satellite remote sensing, performing space matching on the forest vegetation data and the topographic data, and calculating a vegetation-topographic collaborative weight factor; calculating a weighted FWI index based on the vegetation-terrain collaborative weight factor; and analyzing the relationship between the weighted FWI index and the occurrence frequency, intensity and duration of the extreme climate event by adopting a time sequence analysis method and a space analysis method, and visualizing the analysis result. The application blank of vegetation and topographic factors in large warm period FWI calculation is filled, so that the scientificity and accuracy of extreme climate prediction are improved.

The invention discloses a litter leaf area index determination method, the proposed litter leaf area index has the same meaning as a vegetation leaf area index, can participate in related operation as the vegetation leaf area index, and can bring litters into a vegetation investigation system; the index is the number of covering layers of leaves on the earth surface, and the measuring and calculating method comprises the steps that the total spreading area of litter leaves on the unit area of the earth surface is established, and a litter leaf area index estimation model of forest lands such as needle-leaved, broad-leaved and needle-broad mixed forests is established by establishing the power exponent relation between the total spreading area and the dry weight of the litters in the unit area. Therefore, litter unit area dry weight index meaning conversion is achieved, and the method has better adaptive capacity in forest land vegetation investigation.

The invention discloses an image data acquisition method and device based on an unattended intelligent airport, and relates to the technical field of image data processing.The method comprises the steps that forest image data are acquired based on the unattended intelligent airport, the unattended intelligent airport selects a forest area meeting preset mountain forest conditions, and the selected forest area is selected as a forest area; the unattended intelligent airport is used for taking off, landing and parking a predetermined type of unmanned aerial vehicle, and the predetermined mountain forest conditions comprise that the height fall of a mountain land is greater than a preset fall threshold value, the forest vegetation coverage rate is higher than a preset coverage rate threshold value, and the number of roads in a forest is lower than a preset number threshold value; and analyzing the preprocessed forest image data by adopting a pre-trained machine learning model, identifying a forest resource state, and displaying the forest resource state on a terminal screen. The technical problems that image resource analysis is difficult to carry out on an unattended intelligent airport system in a severe environment and the operation efficiency is low in the prior art are solved.

The utility model discloses an under-forest multifunctional movable poultry breeding device, including the henhouse, and the henhouse includes the egg production area, the habitat area, the feeding area and the movable cage, and the henhouse bottom is provided with the bottom plate, and the bottom plate lower surface is provided with a plurality of moving wheels, and the bottom plate top is provided with the mesh, and the mesh and the bottom plate between sliding insertion have the excrement receiving tray, and the habitat area is provided with a plurality of habitat racks, and the habitat area top is provided with the supporting plate, and the supporting plate upper surface is provided with a plurality of egg production nests, and the mesh upper surface of feeding area is provided with the trough and the water tank respectively, and the feeding area one end surface is provided with the sliding door. The utility model discloses through setting up the pull rod and the moving wheel, can realize flexible movement, avoid poultry long -term concentration in the same area activity, can replace the site in time, protect under-forest vegetation and soil structure, maintain the ecological balance, set up the excrement receiving tray, and the excrement receiving tray lets the staff can daily unified collection excrement, avoid excrement pollution breeding environment, and the cleaning process need not frequently enter the henhouse, and the time and energy are saved.

The present application relates to the technical field of remote sensing, in particular to a method for estimating total primary productivity of pantropical forest vegetation depending on leaf age, which comprises obtaining a weather basic data set and a leaf age-based vegetation leaf area index data set, grouping current leaves according to leaf age, including new leaves, mature leaves and old leaves, obtaining the maximum light energy utilization rate of all leaves, and outputting the final productivity through a remote sensing estimation model of total primary productivity. The present application considers the regulation of leaf phenology on canopy photosynthesis, distinguishes the photosynthetic capacity difference between leaf age components, and constructs a remote sensing estimation method suitable for total primary productivity of pantropical forest vegetation, so as to improve the simulation accuracy of total primary productivity of pantropical forest vegetation, and help to accurately monitor the evolution trend of photosynthesis of pantropical forest under climate change.

The invention relates to the technical field of image detection, in particular to a forest vegetation detection method and system based on an unmanned aerial vehicle, and the method comprises the steps: obtaining an RGB image of a forest region in real time through the unmanned aerial vehicle, and carrying out the preprocessing; constructing a comprehensive feature index of an image block in the image, wherein the comprehensive feature index is used for representing a comprehensive feature expression condition of the image block under different illumination influences; adaptively setting a contrast limit threshold value of each image block according to the comprehensive feature index, and performing local enhancement by adopting a CLAHE algorithm; rotten and dead wood detection model training is carried out on the enhanced RGB image so as to identify rotten and dead wood in the RGB image acquired by the unmanned aerial vehicle; and deploying the trained model to the unmanned aerial vehicle for real-time detection and position feedback. According to the method, the image is effectively enhanced according to the distinguishing features of the decayed dead wood in different blocks and the background surface soil in the image, and the accuracy of subsequent forest vegetation detection and recognition is improved.

Achieve greater efficiency and sophistication in forest management operations. [Solution] A program to be executed by a computer having a processor, which causes the processor to perform the following steps: acquiring status information indicating the state of the forest at a specified time from multiple information sources; organizing the acquired status information in chronological order using identification information that uniquely identifies specific areas within the forest geographically as a common key, and storing it in a database as historical information; analyzing time-series changes in the satellite data included in the historical information, and extracting areas of change where vegetation in the forest has changed; and comparing the extracted changed areas with ground record data included in the historical information, and calculating the priority of visiting the forest for on-site inspection based on whether or not there is ground record data corresponding to the changed areas.

This application relates to the field of computer vision technology and provides a method for detecting forest vegetation diseases and pests based on unmanned aerial vehicles (UAVs). The method includes: acquiring forest disease and pest images using a UAV; preprocessing the images to generate an initial input feature map; performing feature enhancement on the initial input feature map to obtain a first feature map; performing saliency gate calibration and feature encoding on the first feature map to generate a multi-scale fusion feature map; performing multi-level fusion on the multi-scale fusion feature map to obtain first feature maps of different scales; and performing reparameterized large kernel convolution on each first feature map to output a target feature map, which can locate small disease and pest areas. This application introduces targeted improvements to the YOLO11 model at each processing stage to generate an MDFR-YOLO model, thereby enhancing the network's ability to detect forest vegetation diseases and pests in complex backgrounds.

The present application relates to the technical field of forest detection, and specifically discloses a forest vegetation detection method and system based on a UAV, which collects visible light images, spectral images and radar data of a forest area through a UAV carrying a visible light camera, a spectral camera and a radar, constructs a three-dimensional model of a tree based on the radar data, determines tree crown data and tree species of the tree based on the visible light images and the spectral images, and determines the volume, biomass and carbon storage of the tree based on the three-dimensional model, the tree species and the tree crown data, so as to calculate the total volume, total biomass and total carbon storage of all trees in the target forest area, without manually inspecting and sampling tree data, thereby shortening the time consumption of forest detection, improving the detection efficiency, and calculating the volume, biomass and carbon storage of each tree through single-tree modeling combined with visible light images and spectral images, and then summing up the total volume, total biomass and total carbon storage of the entire target forest area, so that the detection result is more accurate.

The invention discloses a comprehensive evaluation method for a service function of a forest vegetation ecosystem, and relates to the field of forest ecology, and the method comprises the steps: obtaining and preprocessing multi-source data of an evaluation region, and generating a standardized multi-source data cube; based on the standardized multi-source data cube, calculating the material quality of multiple ecological system service functions of each evaluation unit in the evaluation area, and generating an ecological system service function material quality data set; performing time sequence analysis on the ecosystem service function material quality data set, and calculating to obtain a service function toughness index of each evaluation unit for each ecosystem service function; and extracting each ecological system service function from the standardized multi-source data cube to form resource consumption data. According to the method, the conversion from static evaluation to sustainable comprehensive evaluation covering stability and efficiency is realized, and the problem that the long-term sustainability of an ecological system service function cannot be effectively quantified in the prior art is solved.

The invention provides a tree carbon reserve monitoring method and device based on a laser radar and a storage medium. The method comprises the following steps: scanning each sample plot in N sample plots sampled in a monitoring area corresponding to forest vegetation through a laser radar, and acquiring first point cloud data corresponding to a tree; preprocessing the first point cloud data to obtain second point cloud data of the trees in each sample plot; based on the second point cloud data of each sample plot, performing individual tree segmentation and parameter extraction to obtain a first individual tree number and a first feature parameter corresponding to the individual tree; correcting the first characteristic parameter based on a pre-established correction model to obtain a corrected first characteristic parameter; and based on the first individual tree number of each sample plot and the corrected first characteristic parameter, determining the tree carbon reserve of the monitoring area through a preset carbon measurement model. Therefore, the monitoring efficiency and the monitoring accuracy of the tree carbon reserves can be improved.

The application discloses a vegetation reflectance spectrum measurement method and system for forest monitoring, acquires original spectrum collection signals of a target area of a forest and carries out canopy penetration attenuation correction to establish a benchmark reflectance spectrum set, identifies a forest canopy projection area and atmospheric scattering interference of the benchmark reflectance spectrum set to form a ground true reflectance spectrum, carries out red edge feature positioning and band ratio analysis based on the ground true reflectance spectrum to form a vegetation reflectance spectrum index, carries out multi-time spectrum difference detection on the vegetation reflectance spectrum index and removes phenology interference to determine a change confidence interval, screens abnormal spectra based on the change confidence interval and carries out forest health grade labeling after correction by a spectrum evolution coefficient to output forest coverage change measurement results. The application can effectively eliminate the interference of forest canopy shadow and atmospheric scattering, distinguish normal phenology fluctuation from real health abnormality, and provide reliable spectrum data support for forest vegetation state monitoring.

The embodiment of the invention relates to the technical field of plant management, and discloses an unmanned aerial vehicle-based vegetation AI identification management method, which comprises the following steps of: identifying wide-area remote sensing data to obtain all abnormal sub-areas and window detection areas in a target detection area; executing a second acquisition task on all abnormal sub-regions and window detection regions to obtain corresponding fine remote sensing data; determining vegetation state information of the abnormal sub-region according to the fine remote sensing data, determining a canopy shielding relationship in the window detection region according to the fine remote sensing data, and identifying and determining a potential under-forest region under canopy coverage; probabilistic inference is carried out on the vegetation state of the potential under-forest region through fusion of multi-source evidence, and an under-forest vegetation probability distribution diagram containing inferred vegetation categories and corresponding confidence coefficients is generated; and generating a comprehensive vegetation state report of the target detection area. And the overall identification management efficiency is improved through a layered acquisition mode.

The application discloses a kind of forest community biomass stability monitoring device and method based on ground-based laser radar in the technical field of forest community biomass stability research, comprising: mounting seat and the supporting leg of being arranged on mounting seat, laser radar and data processing equipment are arranged on the mounting seat, the mounting seat includes receiving box and the supporting plate of being slidably arranged in receiving box, the data processing equipment is arranged on supporting plate, the receiving box is detachably provided with the fixed sample stake for being arranged in fixed sample plot, and a plurality of hanging cards for marking the vegetation of fixed sample plot are sleeved on the fixed sample stake;The LiDAR three-dimensional structure point cloud data of forest vegetation of fixed sample plot is obtained by laser radar, the data processing equipment processes the obtained LiDAR three-dimensional structure point cloud data, and the forest community biomass stability is obtained, the dependence on staff is reduced, and the accuracy of forest community biomass stability monitoring is improved.

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.

This invention discloses a method for evaluating the effect of forest vegetation restoration, comprising: determining the study area for evaluating the effect of forest vegetation restoration; calculating the vegetation community index and ecological function index of the control area based on time series; calculating the extreme values ​​of fluctuation of the vegetation community index and ecological function index; calculating the influence weights of environmental factors and disturbance factors on vegetation community and ecological function; constructing a calculation model for calculating the forest vegetation restoration index based on the influence weights; uniformly selecting several sampling areas within the study area; calculating the forest vegetation restoration index of each sampling area using the calculation model; constructing a semi-variogram function; and evaluating the effect of forest vegetation restoration in the study area. This invention, starting from a four-dimensional index system of vegetation community-ecological function-environmental factors-disturbance factors, comprehensively evaluates the effect of forest vegetation restoration, obtains the spatial differences in the effect of forest vegetation restoration, and provides a basis for differentiated restoration management.

The invention discloses a composite operation method for breeding bees under a eucalyptus forest, which specifically comprises the following steps: step 1, carrying out habitat evaluation and zoning planning under the eucalyptus forest, and carrying out ecological environment evaluation and zoning planning on the basis of site conditions, forest stand structures, vegetation composition under the forest and microclimate characteristics of the eucalyptus forest; the under-forest space is divided into a core culture area, a transition buffer area and an ecological conservation area, and corresponding culture units are configured according to the biological characteristics of different bees; the invention relates to the technical field of agriculture and forestry. According to the composite operation method for breeding the bees under the eucalyptus forest, through systematic habitat evaluation and zoning planning, multi-layer nectar source plant construction, multi-bee-species collaborative breeding, dynamic management and resource recycling, the problem that a traditional under-forest bee breeding mode is not matched with the special habitat of the eucalyptus forest is solved; the health level of a bee colony and the utilization efficiency of a natural nectar source are remarkably improved, and multi-element synergy of pollination, biological control and bee product production is realized.

The invention relates to a mangrove forest vegetation carbon reserve estimation method based on multi-source data. The method comprises the following steps: extracting a ground surface characteristic parameter set of ground surface multi-source remote sensing data of a research area, and dividing mangrove forest ecological units according to the ground surface characteristic parameter set; acquiring measured data of the underground carbon library of the representative sample plot in the unit, training by combining with the parameter set to obtain an underground carbon distribution inference model, and inverting an underground carbon distribution proportion map; and the above-ground carbon reserve map and the above-ground carbon reserve map are superposed to calculate a mangrove forest vegetation all-carbon library spatial distribution map. By adopting the method, the problems of low precision and large underground carbon conversion deviation of traditional estimation can be solved, and spatial explicit quantitative accurate estimation of the mangrove forest vegetation all-carbon library is realized.

A forest vegetation carbon emission reduction remote sensing evaluation method and system, the method comprises the following steps: obtaining the land use change data of the target area year by year, extracting the existing range data of the newly increased forest as vector data; obtain remote sensing digital image data, calculate multiple vegetation indexes in different seasons, and extract pixel characteristics; obtain forest resource inventory data, combine the pixel characteristics, establish a tree species identification model, identify the tree species in the pixel characteristics, combine the vector data, determine the dominant tree species in the target area, obtain the tree species parameters, plot quantity and plot area of the newly increased forest in the target area; combine the tree species growth model, calculate the carbon layer removal amount of each plot, and calculate the carbon emission reduction amount of each plot, and complete the remote sensing evaluation of the carbon emission reduction amount of each plot of the target area. The present application adopts dynamic, multi-source long time sequence data fusion, combines tree species, and calculates carbon emission reduction amount of each plot, so that the evaluation result is more accurate.

This invention relates to the field of ecological environment monitoring and carbon cycle assessment technology, and discloses a forest carbon sink prediction method based on spatial heterogeneity decoupling and hierarchical distributed autoregressive meta-learning fusion. The method first integrates ground survey data, remote sensing temporal features, meteorological driving factors, topographic factors, and soil attribute data to construct a multi-source heterogeneous feature system. Then, it decouples and models different categories of features, and divides ecologically homogeneous clusters based on a self-organizing map neural network. For each ecological cluster, it constructs an independent nonlinear autoregressive model for distributed prediction. Based on this, a meta-learning fusion model is constructed, and the outputs of each distributed model are dynamically weighted and biased to obtain the predicted forest vegetation carbon density, and the forest vegetation carbon storage and annual carbon sink are calculated. This method can effectively handle spatial heterogeneity and temporal lag effects, constructs a progressive error convergence structure, and improves the accuracy, stability, and cross-regional generalization ability of forest carbon sink prediction.

The embodiment of the invention provides an interannual change determination method of forest biomass and a related product. The method is applied to the field of forest biomass calculation. The method comprises the following steps: acquiring ground observation time sequence characteristics and multi-source static characteristics; wherein the ground observation time sequence features indicate observation forest environment features and forest vegetation features related to observation time, and the multi-source static features indicate environment features related to the spatial position of the sample plot; processing the ground observation time sequence features and the multi-source static features by using a random forest regression algorithm to obtain interannual changes of forest biomass; wherein the random forest regression algorithm is a forest regression algorithm which is trained in a layered manner according to a time interval by using a training sample set; the training sample set comprises training data with a preset time limit. According to the method, through the random forest model, the ground observation time sequence features and the multi-source static features are processed, and the annual change of the large-scale forest biomass can be determined.

The invention relates to the technical field of artificial forest ecological management and soil carbon immobilization, in particular to a method and system for improving organic carbon in artificial forest soil based on multi-source data, and the method comprises the steps: collecting the multi-source data which comprises vegetation data, soil data, microorganism data and environment data; and preprocessing the multi-source data to obtain a structured data set. According to the method, multi-source data of a vegetation soil microbial environment is collected, and a structured data set is formed through cleaning, normalization and fusion, so that the problem that a soil organic carbon lifting mechanism cannot be systematically analyzed by depending on single data in the prior art is solved; a random forest and long-short-term memory network hybrid model is adopted to construct a prediction model, an optimal under-forest vegetation construction mode is output, the process from data acquisition to measure execution is periodically repeated, management measures are dynamically adjusted, and the problem that an existing fixed management mode is difficult to adapt to environmental changes, so that the carbon lifting effect fluctuates is solved.

The utility model belongs to the technical field of woodland waste treatment, and particularly relates to woodland grass and tree waste fermenting and crushing equipment which comprises a shell, a separating mechanism is fixedly connected in the shell, and a crushing mechanism is arranged on one side of the separating mechanism and fixedly connected in the shell. A shifting assembly for shifting the grass and tree wastes into the crushing mechanism is arranged in the separating mechanism and is positioned on one side, close to the crushing mechanism, of the separating mechanism; the separating mechanism comprises a water tank, water is contained in the water tank, the material stirring assembly is rotationally connected into the water tank, and the lower portion of the material stirring assembly sinks into the water. According to the grass and tree waste crushing device, sundries such as broken stones and soil blocks in grass and tree waste can be screened out, then the grass and tree waste is crushed, and the problem that the sundries such as the broken stones and the soil blocks affect the crushing efficiency of the crushing mechanism and even damage the crushing mechanism can be effectively solved.

The application provides a forest fire risk perception AI terminal, and relates to the field of forest fire monitoring and early warning.The forest fire risk perception AI terminal comprises a perception layer, wherein the perception layer comprises a multispectral perception module, a video perception module, a smoke perception module and a laser radar perception module; the multispectral perception module is used for acquiring spectral information of different wave bands, detecting the health condition of forest vegetation, the change of ground surface temperature and the thermal radiation characteristics of early fire, and is equipped with a plurality of detectors of different wave bands covering the wave band range of visible light, near infrared, medium infrared and far infrared; through connection with an early warning system and other related equipment and systems, real-time transmission and sharing of data are realized, fire risk information and thunder and lightning early warning information can be timely and effectively conveyed to relevant departments and personnel, the emergency response speed is improved, and the loss caused by forest fire is reduced.

The invention discloses a forest sample plot vertical layered vegetation coverage estimation method, which comprises the following steps of S10, acquiring laser radar point cloud data and optical image data of a target plot, and performing preprocessing; s20, vertically layering the laser radar point cloud data, respectively extracting the uppermost point cloud of each vegetation layer in a visible and sheltered region based on the first echo and the multiple echoes, and calculating the initial projection area of each layer according to the extracted uppermost point cloud; s30, fusing the optical features with the point cloud structure features, constructing a vegetation layer projection area correction model, and correcting the initial projection area of the vegetation layer by using the model to obtain a corrected projection area; and S40, establishing a quantitative conversion model of the two-dimensional FVC and the three-dimensional FVC based on a geometrical optical model and a clearance rate theory. According to the method, the dependence on point cloud classification precision is overcome, the estimation precision of the FVC of the under-forest vegetation, especially the herbaceous layer, is remarkably improved, and the method has the advantages of being clear in physical mechanism and high in adaptability.

The application discloses a cross-domain few-sample forest vegetation adaptive feature recognition and extraction system and method, which comprises a ViT basic model, an adaptive semantic embedding module, a boundary perception module, a phenological disturbance and consistency engine module, a context prototype anchoring remapping module and a gradual boundary optimization network module. The system extracts basic query features and support set features by using a ViT query model. The support set features are subjected to adaptive semantic embedding, prototype generation and phenological disturbance to form a complete prototype. The basic query features are combined with the complete prototype, subjected to context remapping and then sent to a gradual boundary optimization module. Finally, pixel-level forest segmentation masks and fine-grained category results are output. The adaptive semantic embedding module of the system dynamically optimizes the feature distance between categories, and fuses forest-specific features such as NDVI spectral index and canopy texture, so that the model focuses more on the core discriminative information in the limited samples, and effectively improves the stability and distinguishability of fine-grained recognition under the condition of small samples.