An intelligent data processing method and system based on forestry ecological monitoring
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- JUNCHUANG (HANGZHOU) SPACE TECHNOLOGY CO LTD
- Filing Date
- 2026-04-09
- Publication Date
- 2026-06-09
Smart Images
Figure CN122176582A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the technical field of forestry ecological monitoring, and in particular to an intelligent data processing method and system based on forestry ecological monitoring. Background Technology
[0002] Forestry ecological monitoring is a comprehensive technical and management work that involves the long-term, systematic observation, assessment, and early warning of the structure, function, and health status of forestry ecosystems such as forests, wetlands, and deserts.
[0003] In the process of forestry ecological monitoring, personnel usually go to the forestry area to conduct on-site surveys, then manually control drones to take images of the forestry from the air, and finally combine the data from the on-site surveys with the images to generate relevant reports on forestry trees and upload them.
[0004] In the process of forestry ecological monitoring, the data processing efficiency is reduced when conducting manual field surveys and data analysis due to the large amount of surveyed data and images and the long processing time. Summary of the Invention
[0005] To improve the data processing efficiency of forestry ecological monitoring, this invention provides an intelligent data processing method and system based on forestry ecological monitoring.
[0006] In a first aspect, the present invention provides an intelligent data processing method based on forestry ecological monitoring, employing the following technical solution: A smart data processing method based on forestry ecological monitoring includes: S10: Responds to monitoring signals and preset detection range to control the UAV to collect image detection information and surrounding environment information; S11: Obtain tree image information by combining image detection information with preset tree features; S12: Identify tree type and tree shape from tree image information; S13: Combine tree type, tree shape and surrounding environment information to establish a tree detection model; S14: Obtain tree fall information based on the detected tree model; S15: Generate a tree fall report based on tree fall information and upload it.
[0007] By adopting the above technical solution, tree fall reports are obtained and uploaded by analyzing image detection information and surrounding environmental information. This enables the control of drones to automatically collect images of trees and their surrounding environment, and automatically analyze and upload future tree fall situations, thereby improving the data processing efficiency of forestry ecological monitoring.
[0008] Optionally, methods for obtaining tree fall information include: S20: Simulate the center of gravity of the tree from the detected tree model; S21: Obtain the tree's stress parameters based on the tree's center of gravity and surrounding environment information; S22: Identify abnormal information based on tree image information and preset abnormal features; S23: Combine anomaly information with tree stress parameters to simulate tree toppling information from the detected tree model.
[0009] By adopting the above technical solution and analyzing the detected tree model to obtain tree falling information, it is possible to more accurately simulate the stress state and falling risk of trees in the real environment, thereby improving the accuracy and reliability of tree falling information simulation.
[0010] Optional methods for verifying tree fall information include: S30: Determine whether the tree image information contains preset hole features; S31: No adjustment is made when the tree image information does not contain hole features; S32: When the tree image information contains cavity features, the scanning information of the tree image information is collected; S33: Forming the internal shape of a tree based on scanned information; S34: Update tree fall information based on the tree's internal shape, tree type, and surrounding environment information.
[0011] By adopting the above technical solution, tree image information and cavity features are analyzed to update tree fall information, thereby enabling timely detection of internal damage caused by biological nesting and other reasons. This reduces the problem that external images cannot assess internal structural defects, thus improving the comprehensiveness and accuracy of tree fall risk assessment.
[0012] Optional methods for updating tree fall information include: S40: Obtain the detected biota by combining information about the surrounding environment and the detection range; S41: Obtain cavity parameters based on tree image information and cavity features; S42: Identify labeled biomes from detected biomes based on tree type and cavity parameters; S43: Combine the marked biota with tree type to obtain the baseline internal shape; S44: Update tree fall information based on the baseline internal shape and the tree's internal shape.
[0013] By adopting the above technical solution, the tree fall information is updated by analyzing the internal shape of the tree, the tree type, and the surrounding environment. This allows for accurate identification of the causes of defects inside the tree cavity and analysis of the possibility of future defects, thereby improving the accuracy of tree fall information assessment.
[0014] Alternatively, methods for updating tree fall information also include: S50: Detection location ratio is obtained based on the internal shape of the tree and the detected tree model; S51: Obtain the mark reference shape based on the detection position ratio and the reference internal shape; S52: Compare the consistency between the marked reference shape and the internal shape of the tree to determine the type of event to be detected; S53: Simulate future hollowing trends from the detected tree model by combining the detected event type, tree shape, and labeled biomes; S54: Update tree fall information based on future hollowing trends and detected event types.
[0015] By adopting the above technical solutions, the tree's internal shape and the tree model are analyzed to update the tree's falling information, thereby enabling the analysis of future biological behavior inside the tree and improving the foresight and early warning capabilities of forestry ecological monitoring.
[0016] Optionally, the validation method for detecting event types also includes: S60: Obtaining labeled biological characteristics based on labeled biological populations; S61: Update tree image information and combine it with labeled biological features to obtain the target organism; S62: Obtain the biological change frequency based on the update status of the target organism and tree image information; S63: Update the scan information to obtain the changed information; S64: Combine change information with biological change frequency to update the detected event type.
[0017] By adopting the above technical solution, the detection event type is updated by analyzing the marked biological community, tree image information, and scanning information. This allows for real-time calibration of the judgment of biological activity stages, thereby improving the accuracy and timeliness of the event type determination.
[0018] Optionally, methods for updating tree-falling information based on future hollowing trends and detected event types include: S70: Obtain the duration of future changes based on the type of detected event; S71: Combine future hollowing trends with tree detection models to obtain the trend of center of gravity change; S72: Obtain the total duration of change based on the trend of the center of gravity change and the future duration of the change; S73: Obtain historical environmental information based on the detection range; S74: Combine the total duration of change, surrounding environmental information, and historical environmental information to obtain future environmental information; S75: Update tree fall information using future environmental information and total duration of change.
[0019] By adopting the above technical solution, and by analyzing future hollowing trends and detection event types to update tree falling information, it is possible to combine the future damage to the tree's interior by biological processes to further predict the future falling of the tree, thereby improving the accuracy of tree falling information.
[0020] Optional methods for verifying tree fall reports include: S80: Retrieve the marked image information surrounding the tree image information from the image detection information; S81: Identifying marked items based on marked image information; S82: Obtain the force information applied to the marker based on the information of the marked object and the surrounding environment; S83: Update the tree detection model and tree fall report by combining the marked force information.
[0021] By adopting the above technical solution, the supporting force or impact force exerted by the target object on the target tree is analyzed by identifying the marked objects around the target tree to update the detection tree model and tree fall report, thereby improving the accuracy of mechanical simulation, reducing the bias caused by isolated analysis of a single tree, and further improving the accuracy of tree fall risk assessment.
[0022] Optional methods for verifying tree fall reports include: S90: Retrieve the direction of tree fall from the tree fall report; S91: The extent of the fall impact is determined based on the direction and shape of the fallen trees; S92: Obtain target image information by marking image information and the extent of the impact of the dumping; S93: Combine target image information to retrieve target tree reports; S94: Update the target tree report based on the detected tree model and the extent of the fall impact, and add the target tree report to the tree fall report.
[0023] By adopting the above technical solution, the impact range of tree fall can be determined by predicting the direction of tree fall, the surrounding trees that may be affected within the range can be identified, and the existing target tree reports can be retrieved and the risk status of the tree fall reports on the impact on other trees can be updated, thereby improving the accuracy of tree fall risk assessment.
[0024] Secondly, this application provides an intelligent data processing system based on forestry ecological monitoring, which adopts the following technical solution: An intelligent data processing system based on forestry ecological monitoring includes: The acquisition module is used to acquire image detection information and surrounding environment information; A memory for storing a program for an intelligent data processing method based on forestry ecological monitoring; The processor is used to load and execute programs stored in memory.
[0025] In summary, this application includes at least one of the following beneficial technical effects: 1. By analyzing image detection information and surrounding environmental information to obtain and upload tree fall reports, it is possible to control drones to automatically collect images of trees and their surrounding environment, and automatically analyze and upload future tree fall situations, thereby improving the data processing efficiency of forestry ecological monitoring. 2. By analyzing the internal shape of trees, tree type, and surrounding environmental information to update tree fall information, the causes of defects inside tree cavities can be accurately identified and the future occurrence of internal defects can be analyzed, thereby improving the accuracy of tree fall information assessment. 3. By determining the impact range of the predicted tree falling direction, identifying the surrounding trees that may be affected within that range, retrieving existing target tree reports, and updating the risk status of the tree falling reports on the impact on other trees, the accuracy of tree falling risk assessment can be improved. Attached Figure Description
[0026] Figure 1 This is a flowchart of an intelligent data processing method based on forestry ecological monitoring according to an embodiment of the present invention; Detailed Implementation
[0027] The present invention will now be described in further detail with reference to the accompanying drawings and embodiments.
[0028] Reference Figure 1 This application discloses an intelligent data processing method based on forestry ecological monitoring, comprising the following steps: S10: Responds to monitoring signals and preset detection range to control the UAV to collect image detection information and surrounding environment information.
[0029] The monitoring signal is a command signal set by technicians to conduct forestry inspections.
[0030] The testing scope is the distribution area of forests that need to be tested, as set by the technicians.
[0031] Image detection information refers to images captured by cameras on drones within the detection range.
[0032] Surrounding environment information refers to information such as wind, temperature, and terrain distribution around the detection range. This information can be generated by parameters detected by sensors such as temperature and wind speed sensors on the drone, combined with terrain distribution information pre-input by the operator.
[0033] S11: Obtain tree image information by combining image detection information with preset tree features.
[0034] Tree features are the characteristics of trees, such as color, texture, and shape, defined by technicians. Tree image information refers to images of trees within the detection range, identified by extracting images containing tree features from the image detection information.
[0035] S12: Identify tree type and tree shape from tree image information.
[0036] Tree type refers to the species of trees within the detection range, and tree shape refers to the shape of trees within the detection range. Tree type and tree shape are identified from tree image information.
[0037] S13: Combine tree type, tree shape and surrounding environment information to build a tree detection model.
[0038] A tree detection model is a digital 3D model used to simulate and analyze the current state and future changes of trees. It is created by using tree type and shape as the basic geometric model, and then employing 3D reconstruction algorithms (such as multi-view stereo vision or LiDAR point cloud-based modeling methods) to convert the tree shape into a digital mesh or voxel model. The tree type is then attached as an attribute label, and mechanical parameters such as wood density, elastic modulus, and ultimate strength are retrieved from a pre-defined tree attribute lookup table. Finally, surrounding environmental information (such as real-time wind speed sensor data and pre-input terrain slope data) is bound to the model as boundary conditions or force parameters, forming a comprehensive data model that serves as the tree detection model.
[0039] The tree attribute reference table stores mechanical parameters such as wood density, elastic modulus, and ultimate strength corresponding to different tree types. The parameters in the tree attribute database are set in advance by those skilled in the art based on actual conditions through experiments.
[0040] S14: Obtain tree fall information based on the detected tree model.
[0041] Tree falling information refers to information such as the time point when a tree within the detection range is about to fall and the direction of the fall. The tree falling information is obtained by performing force analysis on the detected tree model through mechanical simulation algorithms (such as finite element analysis or simplified mechanical model).
[0042] S15: Generate a tree fall report based on tree fall information and upload it.
[0043] A tree fall report is a report that presents tree fall information in the form of a structured document or data package. It is completed by inputting the various data in the tree fall information into a preset report template, filling in the corresponding data, and then uploading the completed report to the operator's system for the operator to review and make decisions.
[0044] Methods for obtaining information about fallen trees include: S20: Simulate the center of gravity of a tree from a detected tree model.
[0045] The tree centroid refers to the center point of the overall mass distribution of a tree. The coordinates of the center position of the overall mass distribution of a tree are calculated from the tree detection model using discretization integration, and this coordinate is used as the tree centroid.
[0046] S21: Obtain the tree's stress parameters based on the tree's center of gravity and surrounding environment information.
[0047] Tree stress parameters refer to the magnitude, direction, and specific location of various external forces acting on the tree model. These mainly include wind load, gravity (which is implicitly included in the center of gravity calculation), and lateral force caused by slope.
[0048] Wind speed and direction data are retrieved from the surrounding environment, and combined with tree shape (canopy projection area and height can be calculated and retrieved using a tree detection model) to calculate wind force using fluid dynamics formulas. Finally, the total force parameters (such as resultant force and resultant moment) obtained by vector synthesis of wind force and the lateral component of gravity in the slope direction are used as the force parameters of the tree.
[0049] S22: Identify abnormal information based on tree image information and preset abnormal features.
[0050] Anomaly information refers to parameter information about abnormal damage on trees that could cause them to fall. This includes: damage type (such as cavities, cracks, decay, insect infestation), damage location (three-dimensional coordinates on the trunk or branches), damage size (such as the volume of cavities, the length and depth of cracks, and the area of decayed areas), and damage degree (such as the residual strength coefficient of the wood). By identifying the type and location of abnormal features from tree image information, and then comparing the tree's baseline shape (a reference table or a simulation of the undamaged complete shape based on the current healthy shape of the tree) with the tree shape, the size of the abnormal features is obtained. Based on this size, the damage degree is matched from the tree attribute reference table. The damage degree, anomaly type, anomaly location, and anomaly size are then integrated to form the anomaly information.
[0051] The tree attribute lookup table stores the degree of damage corresponding to the size of different abnormal features; the larger the size, the greater the degree of damage.
[0052] S23: Combine anomaly information with tree stress parameters to simulate tree toppling information from the detected tree model.
[0053] Anomaly information and tree stress parameters are input into the tree detection model. Finite element analysis is used to calculate the stress distribution and deformation inside the tree under a given load. If the maximum principal stress exceeds the ultimate strength of the wood (retrieved from a lookup table based on tree type), or if the model becomes unstable (e.g., the overturning moment exceeds the stabilizing moment), the tree is deemed to be at risk of falling. The falling direction is determined by a combination of the force direction, center of gravity shift, and the weakest area of the structure. The final output includes tree falling information containing probability, direction, and time window.
[0054] In this embodiment, the simulation process can be performed iteratively, and the calculation is repeated after each update of abnormal information or environmental data.
[0055] The verification methods for tree fall information also include: S30: Determine whether the tree image information contains preset hole features.
[0056] The cavity feature refers to the characteristic of tree cavities within a detection range set by technicians. By determining whether tree image information contains cavity features, it can be determined whether an organism has drilled a hole inside the tree.
[0057] S31: No adjustment is made when the tree image information does not contain hole features.
[0058] If the tree image information does not contain cavity features, it means that no organism has drilled holes inside the tree, and no adjustments will be made.
[0059] S32: When the tree image information contains cavity features, the scanning information of the tree image information is collected.
[0060] Scanning information refers to information used to scan the shape of the tree's exterior and interior, as well as the objects inside. When the tree image information contains hollow features, it indicates that an organism has drilled a hole inside the tree. In this case, a laser radar pre-installed on a drone emits laser pulses at the tree and receives the reflected signals from the tree trunk surface and the boundaries of the internal cavities to generate three-dimensional point cloud data as scanning information.
[0061] S33: Form the internal shape of the tree based on the scanned information.
[0062] The internal shape of a tree refers to the digital three-dimensional form of the cavities, channels, nests, and other structures inside the trunk. By filtering and denoising the scanned information, and then identifying the interface between the cavity and the wood based on the difference in signal reflection intensity or dielectric constant, the extracted boundary point cloud is transformed into a continuous triangular mesh or voxel model to form the internal shape of the tree.
[0063] S34: Update tree fall information based on the tree's internal shape, tree type, and surrounding environment information.
[0064] New information on tree fall can be obtained by analyzing the internal shape of the tree, the tree type, and the surrounding environment.
[0065] Methods for updating tree fall information include: S40: Obtain the detected biota by using information about the surrounding environment and the detection range.
[0066] The detection biome refers to the collection of biological species that are active within the geographical environment and ecological characteristics of the tree and have the habit of nesting on tree trunks. The detection biome is matched from a preset biological activity checklist by the surrounding environmental information and the detection range.
[0067] The biological activity control table stores different surrounding environmental information and the corresponding detection biological groups for different detection ranges. The parameters in the biological activity control table are set in advance by those skilled in the art based on actual conditions.
[0068] S41: Obtain cavity parameters based on tree image information and cavity features.
[0069] Cavity parameters refer to the quantitative description of visible cavities on the surface of a tree trunk. They are obtained by identifying parameters such as size, shape, orientation, location, and edge features (smoothness, fresh sawdust, resin secretion) corresponding to the cavity features from tree image information.
[0070] S42: Identify labeled biomes from the detected biomes based on tree type and cavity parameters.
[0071] The marker biota refers to the biological species that are active on the hollow tree and leave hollow parameters from the test biota. The corresponding organisms are matched with the tree type and hollow parameters from the biological activity table, and the set of organisms that match the corresponding organisms is retrieved from the test biota as the marker biota.
[0072] The biological activity reference table stores the active organisms corresponding to different tree types and cavity parameters.
[0073] S43: Combine the marked biota with tree type to obtain the baseline internal shape.
[0074] The baseline internal shape refers to the internal three-dimensional structural template shape of the nest of the marked organism in the corresponding tree species. The baseline internal shape is obtained by matching the marked organism group with the tree type from the biological activity reference table.
[0075] The biological activity reference table stores the baseline internal shape corresponding to different marked biological groups and tree types.
[0076] S44: Update tree fall information based on the baseline internal shape and the tree's internal shape.
[0077] New information on tree fall is obtained by analyzing the internal shape of the baseline and the internal shape of the tree.
[0078] Other methods for updating tree fall information include: S50: The detection location ratio is obtained based on the internal shape of the tree and the detected tree model.
[0079] The detection position ratio refers to the proportion of the spatial position of the tree's internal shape on the trunk to its corresponding position in the detected tree model. It is obtained by taking the proportion of the tree's internal shape's position on the detected tree model as the detection position ratio. For example, the standard tree template is [0, 1]. The actual numerical ratio of the detected tree model is obtained according to the template, and then the set of proportions of the tree's internal shape's position on the detected tree model is taken as the detection position ratio (e.g., [0.3, 0.7]).
[0080] S51: Obtain the mark reference shape based on the detection position ratio and the reference internal shape.
[0081] The marking reference shape refers to the internal shape contained in the detection position ratio within the reference internal shape. Referring to S50, the internal shape contained in the detection position ratio is retrieved from the reference internal shape (the standard tree template in S50 is [0, 1]) as the marking reference shape.
[0082] S52: Compare the consistency between the marker reference shape and the internal shape of the tree to determine the type of event to be detected.
[0083] The detection event type refers to the activity stage of the marked organism community on the tree, including: Active digging stage: the organisms are still digging holes, and the cavity is expanding; Maintenance stage: the organisms have basically completed nest building and only make minor repairs; Abandonment stage: the organisms have left, and the cavity no longer changes or begins to decay naturally; Abnormal expansion stage: other organisms have intervened or structural damage has occurred, and the cavity exceeds the baseline range.
[0084] By analyzing the consistency between the marker reference shape and the internal shape of the tree for each event type, the event type corresponding to the consistency between the internal shape of the tree and the marker reference shape is taken as the detection event type.
[0085] S53: Combine the detection event type, tree shape, and labeled biomes to simulate future hollowing trends from the detected tree model.
[0086] The future hollowing trend refers to the prediction of the evolution direction and rate of change of the internal cavity of the tree trunk in the future based on the current detection event type and the behavioral habits of the marked organisms. The future hollowing trend is obtained by combining the detection event type, tree shape and marked organism group (biological behavior model) into the detection tree model to simulate the volume, direction and time series of the internal shape of the tree.
[0087] Biological behavior models are obtained by matching the corresponding biological behavior models from a biological activity control table by labeling biological groups.
[0088] S54: Update tree fall information based on future hollowing trends and detected event types.
[0089] The future trend is to gradually increase the volume of cavities or change their shape in the tree detection model to generate predictive tree models for multiple future time points. Then, the stress analysis of the tree model at each predictive time point will be carried out again to obtain new information on tree falling.
[0090] The methods for verifying event types also include: S60: Obtain the characteristics of labeled organisms based on labeled organism populations.
[0091] Tagged biomarkers are visual or behavioral features extracted from a tagged biomass that can be used to identify and locate specific biological individuals in an image. These features include physical outlines, movement patterns, and vocal characteristics. Tagged biomarkers are matched from a biological activity checklist using a tagged biomass.
[0092] The biological activity reference table stores the marker biological characteristics corresponding to different marker biological groups.
[0093] S61: Update tree image information and combine it with labeled biological features to obtain the target organism.
[0094] The target organism refers to the organism that is actually active on the tree. Tree image information is continuously collected, and the organism corresponding to the marked biological characteristics in the tree image information is identified as the target organism.
[0095] S62: Obtain the biological change frequency based on the update status of the target organism and tree image information.
[0096] Biological change frequency refers to the frequency with which a target organism appears, leaves, or is active on a tree. This is calculated by recording the time, duration, and location of each appearance of the target organism in various tree images, and then calculating the appearance frequency as (number of appearances / total number of data collections) × 100%. Average dwell time is calculated as total dwell time / number of appearances. Entry / exit frequency refers to the number of times an organism enters or exits its nest per unit time. Day / night distribution refers to the ratio of appearances during the day to those at night. Finally, the calculated data are combined to determine the biological change frequency.
[0097] S63: Update scan information to obtain change information.
[0098] Change information refers to parameters that actually change the internal shape of a tree. Change information is obtained by re-collecting scan information and calculating the difference between corresponding data before and after the update.
[0099] S64: Combine change information with biological change frequency to update the detected event type.
[0100] By inputting change information and biological change frequency into a biological activity control table to match labeled event types, the event types that match the labeled event types in the detected event types are taken as new detected event types.
[0101] The biological activity reference table stores different event types that correspond to changes in biological information and the frequency of biological changes.
[0102] Methods for updating tree fall information based on future hollowing trends and detected event types include: S70: Obtain the duration of future changes based on the type of detected event.
[0103] The duration of future change refers to the length of time required to predict that an organism will exhibit a unit trend change under the condition of a detected event type. The duration of future change is determined by matching the biological activity control table with the detected event type.
[0104] S71: Combine future hollowing trends with a tree detection model to obtain the trend of center of gravity change.
[0105] The center of gravity change trend refers to the hollowing trend in the future hollowing trend when the center of gravity of a tree changes. By inputting the future hollowing trend into the detection tree model to simulate the center of gravity, the hollowing trend of each segment when the center of gravity changes in the future hollowing trend is taken as the center of gravity change trend.
[0106] S72: Obtain the total duration of change based on the trend of the center of gravity change and the future duration of the change.
[0107] The total duration of change refers to the total time required for an organism of the detection event type to show a trend of change in the center of gravity in a tree. Referring to S70, the total coefficient is obtained by calculating the quotient of the trend of change in the center of gravity and the unit trend change of the detection event type. The product of the total coefficient and the future duration of change is then calculated as the total duration of change.
[0108] S73: Obtain historical environmental information based on the detection range.
[0109] Historical environmental information refers to environmental data records within the detection range over a past period (such as the past 1 year, 5 years, or 10 years). The surrounding environmental information for a historical period is retrieved from the system based on the detection range as historical environmental information.
[0110] S74: Combine the total duration of change, surrounding environmental information, and historical environmental information to obtain future environmental information.
[0111] Future environmental information refers to the environmental information that appears within the detection range during the total duration of future changes. It is obtained by retrieving the current environmental information corresponding to the future time period based on the total duration of changes from historical environmental information.
[0112] S75: Update tree fall information using future environmental information and total duration of change.
[0113] By inputting future environmental information and the total duration of change into the tree detection model, a new stress analysis is performed to obtain new information on tree toppling.
[0114] Verification methods for tree fall reports include: S80: Retrieve the marked image information around the tree image information from the image detection information.
[0115] The labeled image information refers to the images around the current tree, which are retrieved from the image detection information to serve as the labeled image information.
[0116] S81: Identify marked items based on marked image information.
[0117] Marked items refer to objects (such as trees, buildings, rocks, and piles) around the current tree. These objects are identified from the marked image information and designated as marked items.
[0118] S82: Obtain the force information applied to the marker based on the information of the marked object and the surrounding environment.
[0119] The force information of the marker refers to the information on the force exerted by the marker on the current tree. By retrieving the terrain slope from the surrounding environment information, and then combining the terrain slope with the marker, the direction and component of the force output by the marker towards the tree are obtained as the force information of the marker.
[0120] S83: Update the tree detection model and tree fall report by combining the marked force information.
[0121] A new tree detection model is obtained by inputting the force information of the marker into the tree detection model, and a new tree fall report is obtained by using the new tree detection model.
[0122] Verification methods for tree fall reports include: S90: Retrieve the direction of tree fall from the tree fall report.
[0123] The direction in which a tree falls refers to the direction in which it falls when it falls. This can be obtained by retrieving the direction in which a tree falls from a tree fall report.
[0124] S91: The extent of the fall is determined based on the direction and shape of the fallen tree.
[0125] The fall impact range refers to the spatial range covered or potentially impacted by a fallen tree. It is determined by analyzing the trajectory of three-dimensional data of the falling direction and tree shape, and using this trajectory to define the spatial range of the fall impact range.
[0126] S92: Obtain target image information by marking image information and the extent of the impact of the dumping.
[0127] Target image information refers to the marked image information contained within the dumping impact area, which is used as the target image information.
[0128] S93: Combine target image information to retrieve target tree reports.
[0129] A target tree report refers to a report of fallen trees in a target image. It is generated by identifying the corresponding tree from the target image information and retrieving the corresponding tree fall report from the system as the target tree report.
[0130] In this embodiment, each tree within the detection range has a report of falling trees.
[0131] S94: Update the target tree report based on the detected tree model and the extent of the fall impact, and add the target tree report to the tree fall report.
[0132] By inputting the fall impact range, the detected tree model, and the target tree report into the target tree model, the impact of the tree fall can be re-identified, a new target tree report can be generated, and the target tree report can be added to the tree fall report. This allows the impact of the tree fall to be sent to the operator for review.
[0133] Based on the same inventive concept, embodiments of the present invention provide an intelligent data processing system based on forestry ecological monitoring, comprising: The acquisition module is used to acquire image detection information and surrounding environment information; A memory for storing a program for an intelligent data processing method based on forestry ecological monitoring; The processor is used to load and execute programs stored in memory.
[0134] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional modules is used as an example. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. The specific working process of the system, device, and unit described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.
[0135] The above description is merely a preferred embodiment of the present invention. The scope of protection of the present invention is not limited to the above embodiments. All technical solutions falling within the scope of the present invention's concept are within the scope of protection of the present invention. It should be noted that for those skilled in the art, any improvements and modifications made without departing from the principles of the present invention should also be considered within the scope of protection of the present invention.
Claims
1. A smart data processing method based on forestry ecological monitoring, characterized in that, include: S10: Responds to monitoring signals and preset detection range to control the UAV to collect image detection information and surrounding environment information; S11: Obtain tree image information by combining image detection information with preset tree features; S12: Identify tree type and tree shape from tree image information; S13: Combine tree type, tree shape and surrounding environment information to establish a tree detection model; S14: Obtain tree fall information based on the detected tree model; S15: Generate a tree fall report based on tree fall information and upload it.
2. The intelligent data processing method based on forestry ecological monitoring according to claim 1, characterized in that, Methods for obtaining information about fallen trees include: S20: Simulate the center of gravity of the tree from the detected tree model; S21: Obtain the tree's stress parameters based on the tree's center of gravity and surrounding environment information; S22: Identify abnormal information based on tree image information and preset abnormal features; S23: Combine anomaly information with tree stress parameters to simulate tree toppling information from the detected tree model.
3. The intelligent data processing method based on forestry ecological monitoring according to claim 2, characterized in that, The verification methods for tree fall information also include: S30: Determine whether the tree image information contains preset hole features; S31: No adjustment is made when the tree image information does not contain hole features; S32: When the tree image information contains cavity features, the scanning information of the tree image information is collected; S33: Forming the internal shape of a tree based on scanned information; S34: Update tree fall information based on the tree's internal shape, tree type, and surrounding environment information.
4. The intelligent data processing method based on forestry ecological monitoring according to claim 3, characterized in that, Methods for updating tree fall information include: S40: Obtain the detected biota by combining information about the surrounding environment and the detection range; S41: Obtain cavity parameters based on tree image information and cavity features; S42: Identify labeled biomes from detected biomes based on tree type and cavity parameters; S43: Combine the marked biota with tree type to obtain the baseline internal shape; S44: Update tree fall information based on the baseline internal shape and the tree's internal shape.
5. The intelligent data processing method based on forestry ecological monitoring according to claim 4, characterized in that, Other methods for updating tree fall information include: S50: Detection location ratio is obtained based on the internal shape of the tree and the detected tree model; S51: Obtain the mark reference shape based on the detection position ratio and the reference internal shape; S52: Compare the consistency between the marked reference shape and the internal shape of the tree to determine the type of event to be detected; S53: Simulate future hollowing trends from the detected tree model by combining the detected event type, tree shape, and labeled biomes; S54: Update tree fall information based on future hollowing trends and detected event types.
6. The intelligent data processing method based on forestry ecological monitoring according to claim 5, characterized in that, The methods for verifying event types also include: S60: Obtaining labeled biological characteristics based on labeled biological populations; S61: Update tree image information and combine it with labeled biological features to obtain the target organism; S62: Obtain the biological change frequency based on the update status of the target organism and tree image information; S63: Update scan information to obtain change information; S64: Combine change information with biological change frequency to update the detected event type.
7. The intelligent data processing method based on forestry ecological monitoring according to claim 6, characterized in that, Methods for updating tree fall information based on future hollowing trends and detected event types include: S70: Obtain the duration of future changes based on the type of detected event; S71: Combine future hollowing trends with tree detection models to obtain the trend of center of gravity change; S72: Obtain the total duration of change based on the trend of the center of gravity change and the future duration of the change; S73: Obtain historical environmental information based on the detection range; S74: Combine the total duration of change, surrounding environmental information, and historical environmental information to obtain future environmental information; S75: Update tree fall information using future environmental information and total duration of change.
8. The intelligent data processing method based on forestry ecological monitoring according to claim 7, characterized in that, Verification methods for tree fall reports include: S80: Retrieve the marked image information surrounding the tree image information from the image detection information; S81: Identifying marked items based on marked image information; S82: Obtain the force information applied to the marker based on the information of the marked object and the surrounding environment; S83: Update the tree detection model and tree fall report by combining the marked force information.
9. The intelligent data processing method based on forestry ecological monitoring according to claim 8, characterized in that, Verification methods for tree fall reports include: S90: Retrieve the direction of tree fall from the tree fall report; S91: The extent of the fall impact is determined based on the direction and shape of the fallen trees; S92: Obtain target image information by marking image information and the extent of the impact of the dumping; S93: Combine target image information to retrieve target tree reports; S94: Update the target tree report based on the detected tree model and the extent of the fall, and add the target tree report to the tree fall report.
10. An intelligent data processing system based on forestry ecological monitoring, characterized in that, include: The acquisition module is used to acquire image detection information and surrounding environment information; A memory for storing a program that implements the intelligent data processing method based on forestry ecological monitoring as described in any one of claims 1 to 9; The processor is used to load and execute programs stored in memory.