Tree dynamic coding method and system based on high-precision positioning and visual analysis

Abstract

The application discloses a tree dynamic coding method and system based on high-precision positioning and visual analysis. The method comprises the following steps: acquiring pose data and images; calculating the geographic coordinates of the tree; preprocessing the images; extracting the tree body features and the environment reference features, and establishing the corresponding relationship between the two; determining whether the tree is an existing coded tree based on the geographic coordinates of the tree, the tree body features and the environment reference features; if not, generating the identity code of the tree, wherein the identity code of the tree comprises a basic code and a dynamic extension code; the basic code comprises the encoding corresponding to the geographic coordinates and a serial number; the dynamic extension code comprises a state code for representing the current operation and maintenance state of the tree and a version code for recording the update number of the robust template. By implementing the method of the application, the stable, unique and traceable digital identity of a single tree in the whole life cycle can be established, and the shortcomings of the prior art in reliability, uniqueness and long-term continuity can be overcome.

Description

Technical Field

[0001] This invention relates to computer vision, and more specifically to a method and system for dynamic tree coding based on high-precision positioning and visual analysis. Background Technology

[0002] With the rapid advancement of urban greening, the number of street trees and landscape plants has increased significantly. Achieving long-term, stable digital identification and full lifecycle management of individual trees has become a key challenge for refined governance. Current technical solutions mainly fall into two categories: physical identification methods and methods based on remote sensing and image recognition, but both have significant limitations.

[0003] First, physical identification methods typically involve hanging or attaching signs (such as nameplates, RFID tags, QR codes, etc.) to trees. For example, some areas use QR code tags affixed to tree trunks to facilitate public access to tree information. While this method is simple and straightforward in its initial implementation and facilitates information association, in the complex outdoor natural environment, physical signs are susceptible to weathering, aging, detachment, or human damage, making it difficult to maintain a stable attachment to trees in the long term. Furthermore, a large number of hanging signs not only affect the aesthetics of the city but may also become embedded in the tree trunk as it thickens, harming the tree's health and failing to adapt to changes in tree morphology. Additionally, this method is labor-intensive, requiring regular inspections and maintenance.

[0004] Secondly, methods based on remote sensing and image recognition primarily utilize aerial imagery, satellite remote sensing, or drone aerial photography for large-scale tree detection and identification. For example, one method for canopy identification is based on pixel-level classifiers and template matching. However, these technologies were originally designed for group surveys and tree species classification, not for individual identification management. Their core limitations lie in the lack of individual uniqueness; they typically only identify the canopy or determine the tree species, failing to distinguish different individuals of the same species. Their positioning accuracy is low, making it difficult to meet the precise requirements of urban environments. Furthermore, their visual feature templates are static and unchanging, making it difficult to cope with morphological changes caused by tree growth, seasonal changes, pruning, etc., and unable to handle the impact of changes in the surrounding environment. Therefore, they cannot establish long-term, stable one-to-one identity associations.

[0005] In summary, existing tree identification management technologies fail to address the core issue of how to assign and maintain a unique, lifelong digital identity for dynamically growing individual trees that requires no physical carrier and can adapt to their changes. Specifically, physical identification methods are prone to damage and detachment, lack long-term reliability, and are costly to maintain manually; methods relying solely on geographic coordinates struggle to guarantee the uniqueness and stability of identifications in complex environments or when trees are transplanted; and methods relying solely on image features are susceptible to the natural growth of trees and changes in the external environment, leading to feature drift and making it difficult to achieve long-term consistency of identity.

[0006] Therefore, it is necessary to design a new method to establish a stable, unique and traceable digital identity for a single tree throughout its entire life cycle, overcoming the shortcomings of existing methods in terms of reliability, uniqueness and long-term continuity. Summary of the Invention

[0007] The purpose of this invention is to overcome the shortcomings of the prior art and provide a dynamic tree coding method and system based on high-precision positioning and visual analysis.

[0008] To achieve the above objectives, the present invention adopts the following technical solution: a dynamic tree coding method based on high-precision positioning and visual analysis, comprising:

[0009] Acquire pose data and images collected by devices installed on the vehicle;

[0010] Calculate the geographic coordinates of the tree based on the pose data;

[0011] The image is preprocessed to obtain a preprocessed image;

[0012] Based on the preprocessed image, tree body features and environmental reference features are extracted, and the correspondence between the tree body features and the environmental reference features is established.

[0013] Based on the tree's geographical coordinates, the tree's physical characteristics, and the characteristics of the environmental reference objects, determine whether the tree is a tree that already has a code;

[0014] If the tree is not an existing coded tree, an identity code for the tree is generated. The identity code for the tree includes a basic code and a dynamic extension code. The basic code includes the code corresponding to the geographic coordinates and a sequence number. The dynamic extension code includes a status code for indicating the current operation and maintenance status of the tree and a version code for recording the number of times the robust template has been updated.

[0015] The further technical solution is as follows: after determining whether the tree is an already coded tree based on the tree's geographical coordinates, the tree's intrinsic features, and the environmental reference features, it further includes:

[0016] If the tree is an already coded tree, the feature weights are dynamically adjusted based on the matching degree between the tree's ontological features and historical tree ontological features, as well as the matching degree between the environmental reference features and historical environmental reference features. These features are then fused with historical templates to generate new feature templates, which are recorded in the tree's identity code.

[0017] The further technical solution is as follows: determining whether a tree is an already coded tree based on the tree's geographical coordinates, the tree's intrinsic features, and the environmental reference features includes:

[0018] Candidate tree sets are determined based on the geographic coordinates of the trees through global similarity comparison;

[0019] When a candidate tree set exists, a refined comparison is performed based on the tree's intrinsic features and the environmental reference features to determine whether the tree is an already coded tree.

[0020] The further technical solution is as follows: the calculation of the geographic coordinates of the tree based on the pose data includes:

[0021] A visual algorithm is used to combine the pose data and the image to calculate the distance and azimuth angle of the tree base relative to the vehicle, so as to obtain the relative distance and angle.

[0022] The relative distance and angle are converted into increments in the geodetic coordinate system using a rotation matrix and updated to the vehicle position to determine the absolute coordinates of the tree, thus obtaining the tree's geographic coordinates.

[0023] The geographic coordinates of the trees are encoded to obtain the corresponding codes.

[0024] The further technical solution is as follows: Extracting tree body features and environmental reference features based on the preprocessed image, and establishing the correspondence between the tree body features and the environmental reference features, includes:

[0025] The preprocessed image is analyzed by a convolutional neural network to extract high-dimensional visual features, including tree trunk, bark texture and key points, and the robustness of the image enhancement algorithm under multiple viewpoints and lighting conditions is utilized.

[0026] Identify and calculate the three-dimensional coordinates of fixed reference objects in the environment involved in the preprocessed image, and determine the relative distance and angle between the trees and the fixed reference objects by combining the heading angle, thus forming environmental reference object features;

[0027] The spatial relationship vector is used to associate the tree's physical features with the environmental reference features to construct the relative positional relationship between the two.

[0028] The further technical solution is as follows: when a candidate tree set exists, a refined comparison is performed based on the tree's intrinsic features and the environmental reference features to determine whether the tree is an already coded tree, including:

[0029] When a candidate tree set exists, the matching degree of the tree ontology features and the environmental reference features with the tree ontology features and environmental reference features corresponding to the candidate tree set is compared to determine whether the tree is an already coded tree.

[0030] The further technical solution is as follows: when a candidate tree set exists, the matching degree between the tree ontology features and the environmental reference features and the tree ontology features and environmental reference features corresponding to the candidate tree set is compared to determine whether the tree is an already coded tree, including:

[0031] When a candidate tree set exists, the matching degree between the tree ontology features and the tree ontology features corresponding to the candidate tree set is calculated to obtain the tree feature matching degree.

[0032] Determine whether the tree feature matching degree meets the requirements;

[0033] If the tree feature matching degree meets the requirements, then the tree is determined to be a tree that has already been coded;

[0034] If the tree feature matching degree does not meet the requirements, the matching degree between the environmental reference feature and the environmental reference feature corresponding to the candidate tree set is calculated to obtain the environmental reference feature matching degree.

[0035] If the environmental reference features match the requirements, then it is determined that the tree's identity has been replaced or significantly altered.

[0036] If the environmental reference feature matching degree does not meet the requirements, then the environmental reference feature matching degree is frozen.

[0037] The further technical solution is as follows: after determining whether the tree's identity has been replaced or significantly altered, it also includes:

[0038] The status code is marked to generate a new tree identification code, which is then associated with the original tree identification code.

[0039] The further technical solution is as follows: The feature weights are dynamically adjusted based on the matching degree between the tree's intrinsic features and historical tree intrinsic features, and the matching degree between the environmental reference features and historical environmental reference features. These adjustments are then fused with historical templates to generate new feature templates, which are recorded in the tree's identity code.

[0040] When the matching degree between the tree body feature and the historical tree body feature, and the matching degree between the environmental reference feature and the historical environmental reference feature both exceed the first set threshold, the weight corresponding to the tree body feature and the weight corresponding to the environmental reference feature are kept close in proportion.

[0041] If the matching degree between the environmental reference feature and the historical environmental reference feature is higher than a first set threshold, and the matching degree between the tree body feature and the historical tree body feature is lower than a second set threshold, then the weight corresponding to the environmental reference feature is increased, and the weight corresponding to the tree body feature is decreased to highlight the environmental feature.

[0042] If the matching degree between the tree feature and the historical tree feature is higher than a first set threshold, and the matching degree between the environmental reference feature and the historical environmental reference feature is lower than a second set threshold, then the weight corresponding to the tree feature is increased, and the weight corresponding to the environmental reference feature is decreased to emphasize the tree feature.

[0043] If the matching degree between the tree body feature and the historical tree body feature, and the matching degree between the environmental reference feature and the historical environmental reference feature are both lower than the second set threshold, the weight corresponding to the environmental reference feature is frozen to avoid erroneous updates.

[0044] This invention also provides a tree dynamic coding system based on high-precision positioning and visual analysis, comprising:

[0045] The acquisition unit is used to acquire pose data and images collected by the device installed on the vehicle;

[0046] A calculation unit is used to calculate the geographic coordinates of the tree based on the pose data;

[0047] A preprocessing unit is used to preprocess the image to obtain a preprocessed image;

[0048] The extraction unit is used to extract tree body features and environmental reference features based on the preprocessed image, and to establish the correspondence between the tree body features and the environmental reference features.

[0049] The comparison unit is used to determine whether the tree is a tree that has been coded based on the tree's geographical coordinates, the tree's physical characteristics, and the characteristics of the environmental reference objects.

[0050] The encoding generation unit is used to generate an identity code for the tree if the tree is not a tree with an existing code. The identity code of the tree includes a basic code and a dynamic extension code. The basic code includes the code corresponding to the geographic coordinates and a sequence number. The dynamic extension code includes a status code for indicating the current operation and maintenance status of the tree and a version code for recording the number of times the robust template has been updated.

[0051] The advantages of this invention compared to existing technologies are as follows: This invention collects pose data and images using a device installed on a vehicle, calculates the geographic coordinates of the tree using this information, preprocesses the image to extract the tree's intrinsic features and environmental reference features, and establishes their correspondence. Then, based on the tree's geographic coordinates, intrinsic features, and environmental features, it determines whether the tree is already in the records. If it is a newly discovered tree, an identity code is generated for it, consisting of a basic code composed of a code corresponding to the geographic coordinates and a sequence number, and a dynamic extended code containing a status code indicating the maintenance status and a version code recording the number of robust template updates. This method achieves a stable, unique, and traceable digital identity for a single tree throughout its entire lifecycle, solving the shortcomings of existing technologies in terms of reliability, uniqueness, and long-term continuity, thereby ensuring the efficiency and accuracy of tree management.

[0052] The present invention will be further described below with reference to the accompanying drawings and specific embodiments. Attached Figure Description

[0053] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0054] Figure 1 A flowchart illustrating the tree dynamic coding method based on high-precision positioning and visual analysis provided in an embodiment of the present invention;

[0055] Figure 2 A schematic diagram illustrating tree identification provided in an embodiment of the present invention;

[0056] Figure 3 A top-view schematic diagram of tree identification provided in an embodiment of the present invention;

[0057] Figure 4 Illustration of tree recognition effect provided in embodiments of the present invention Figure 1 ;

[0058] Figure 5 Illustration of tree recognition effect provided in embodiments of the present invention Figure 2 ;

[0059] Figure 6 A schematic diagram illustrating a detailed comparison provided for embodiments of the present invention;

[0060] Figure 7 A schematic block diagram of a tree dynamic coding system based on high-precision positioning and visual analysis provided in an embodiment of the present invention;

[0061] Figure 8 A schematic block diagram of a computer device provided for an embodiment of the present invention. Detailed Implementation

[0062] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0063] It should be understood that, when used in this specification and the appended claims, the terms "comprising" and "including" indicate the presence of the described features, integrals, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or collections thereof.

[0064] It should also be understood that the terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the invention. As used in this specification and the appended claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise.

[0065] It should also be further understood that the term "and / or" as used in this specification and the appended claims refers to any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0066] Please see Figure 1 , Figure 1 This is a schematic flowchart illustrating a tree dynamic coding method based on high-precision positioning and visual analysis provided in an embodiment of the present invention. This tree dynamic coding method, applied in a server, calculates the tree's geographic coordinates using pose data and images collected from devices on a vehicle, and preprocesses the images to extract tree features and environmental reference features, establishing feature correspondences. The method utilizes global similarity comparison and refined comparison to determine whether a tree is already coded, and generates an identity code for newly identified trees, including a basic code and a dynamic extension code, achieving a stable, unique, and traceable digital identity for a single tree throughout its entire lifecycle. Furthermore, by dynamically adjusting feature weights and fusing historical templates to update the tree feature template, reliability in long-term continuity is ensured, overcoming the shortcomings of existing technologies regarding uniqueness and reliability. This process allows for precise monitoring of tree state changes, maintaining the accuracy and consistency of its digital identity even when trees are replaced or undergo significant changes.

[0067] Figure 1 This is a flowchart illustrating the tree dynamic coding method based on high-precision positioning and visual analysis provided in an embodiment of the present invention. Figure 1 As shown, the method includes the following steps S110 to S150.

[0068] S110. Acquire pose data and images collected by the device installed on the vehicle.

[0069] In this embodiment, pose data refers to information about the vehicle's position and attitude collected by devices installed on the vehicle. Specifically, these devices include, but are not limited to, high-precision acquisition modules and computing units such as Global Navigation Satellite System (GNSS) receivers (supporting RTK differential positioning) and inertial measurement units (IMUs). The GNSS module is used to acquire the vehicle's own centimeter-level absolute geographic coordinates P in real time. vehicle (X v Y v The precise pose data of the vehicle is formed by combining the vehicle's position and the accurate heading angle θ (the angle between the vehicle's heading and true north). This pose data not only provides the vehicle's position information, but also its motion state (such as speed and acceleration) and orientation angle, which is crucial for accurately calculating the geographic coordinates of trees relative to the vehicle.

[0070] Images refer to the results of image acquisition of trees on both sides of the road using high-definition vehicle-mounted cameras. These images undergo preprocessing steps, including noise reduction and illumination normalization, to obtain standardized images, thereby ensuring the accuracy of subsequent feature extraction. To ensure that the trees are located in the center of the field of view, object detection algorithms are usually applied to the acquired images to crop the tree regions, generating cropped images containing only the main parts of the trees. This step helps improve the efficiency and accuracy of subsequent visual analysis, especially when using advanced techniques such as convolutional neural networks (CNNs) to extract tree features (such as trunk texture, bark texture, key points of the main trunk / branch, etc.). In addition, by analyzing tree images taken at different times, from different perspectives, and under different lighting conditions, more robust time-series features can be constructed, further enhancing the reliability of tree identification and identity verification. Images not only contain rich visual information about the trees themselves, but also information about fixed reference objects in the surrounding environment, which is crucial for establishing the association between trees and their geographical location.

[0071] S120. Calculate the geographic coordinates of the tree based on the pose data.

[0072] In this embodiment, the geographic coordinates of a tree refer to its specific location relative to the Earth's surface, determined by combining high-precision vehicle positioning information (including centimeter-level absolute geographic coordinates and precise heading angles) and image data collected by the vehicle's onboard camera, and processed by a visual algorithm. This geographic location is typically expressed in latitude and longitude, and to balance uniqueness and data volume requirements, Geohash encoding is also performed.

[0073] In one embodiment, step S120 described above may include steps S121 to S123.

[0074] S121. Using a visual algorithm, the distance and azimuth angle of the tree base relative to the vehicle are calculated by combining the pose data and the image, so as to obtain the relative distance and angle.

[0075] In this embodiment, relative distance and angle refer to the positional parameters of the tree base relative to the vehicle, determined by analyzing images captured by the vehicle-mounted camera using a visual algorithm. Specifically:

[0076] Relative distance: refers to the straight-line distance d from the center of the vehicle to the base of the tree.

[0077] Azimuth angle: refers to the angle α between the line connecting the center of the tree and the center of the vehicle and the vehicle's heading line.

[0078] These parameters are obtained by detecting and measuring trees in the image and their surrounding environmental references (such as streetlights, manhole covers, etc.), ensuring that the location of trees can be accurately identified even under different lighting conditions or viewing angles.

[0079] S122. Use a rotation matrix to convert the relative distance and angle into an increment in the geodetic coordinate system and update it to the vehicle position to determine the absolute coordinates of the tree, so as to obtain the geographic coordinates of the tree.

[0080] In this embodiment, this step involves transforming the observations in the vehicle coordinate system to the geodetic coordinate system. The specific operation is as follows:

[0081] Construct a rotation matrix R(θ) based on the vehicle's heading angle θ.

[0082] Using formula P tree =P vehicle +R(θ)∗[d⋅cosα, d⋅sinα] T , where P vehicle It is the vehicle's absolute geographic coordinates, converting relative distances and angles into increments in the geodetic coordinate system.

[0083] The final result is the absolute coordinates of the tree (i.e., its geographic coordinates), which have a latitude and longitude precision of 9 decimal places. Generally, 6 decimal places are used for Geohash encoding to ensure a planar precision of about 0.1 meters.

[0084] S123. Encode the geographical coordinates of the tree to obtain the code corresponding to the geographical coordinates.

[0085] In this embodiment, during this step, the geographic coordinates of the trees are converted into a form that is easy to store and retrieve—encoding. Specifically:

[0086] First, the geographic coordinates of the trees are encoded using the Geohash encoding method. Geohash is a geocoding method that encodes two-dimensional latitude and longitude data into a one-dimensional string format, which retains a certain degree of spatial resolution while facilitating database indexing and retrieval.

[0087] The encoded result not only reflects the tree's specific location but also serves as part of the basic coding system to construct a unique identity for the tree. This identity remains unchanged throughout the tree's lifespan, enabling unique and traceable management of the tree.

[0088] Through the above steps, the system can effectively determine the precise location of each tree and generate a unique identification code for it, thereby achieving effective management and long-term tracking of individual trees.

[0089] S130. The image is preprocessed to obtain a preprocessed image.

[0090] In this embodiment, the preprocessed image refers to a standardized image suitable for subsequent feature extraction and analysis, generated by performing a series of processing operations on the original acquired image, such as noise removal, illumination adjustment, and region of interest cropping. These preprocessing steps are crucial for improving image quality and enhancing the accuracy and robustness of feature recognition. Specifically, S130 may include the following key steps:

[0091] Filtering techniques, such as Gaussian filtering and median filtering, are used to smooth images and reduce the impact of noise. This reduces or eliminates random noise in the image, which may originate from electronic noise in the sensor itself or from external environmental factors (such as weather conditions).

[0092] Histogram equalization and adaptive contrast enhancement techniques are used to adjust the image brightness distribution, ensuring good visual quality under various lighting conditions. This addresses the issue of inconsistent image brightness caused by changes in lighting conditions, ensuring the comparability of tree images taken at different times.

[0093] First, the location of trees in the image is located using object detection algorithms (such as YOLO, SSD, etc.);

[0094] Then, based on the detection results, the bounding box is expanded by a certain ratio to ensure that the trees and nearby fixed reference objects (such as streetlights, manhole covers, etc.) are fully included, forming a cropped image.

[0095] The part containing the target tree is precisely cropped from the entire image, eliminating interference from irrelevant background information, thereby focusing on the tree itself and its surrounding environmental references.

[0096] The cropped images are uniformly scaled to a specified size, and their pixel values ​​are normalized to a specific range (such as [0, 1] or [-1, 1]) to better adapt to the requirements of deep learning models such as convolutional neural networks.

[0097] This ensures that all images input into subsequent models have a consistent size and format, facilitating batch processing and accelerating the computation process.

[0098] Through the series of preprocessing operations described above, the original image is converted into a high-quality standard format, which not only improves the efficiency and accuracy of subsequent feature extraction stages but also provides a reliable data foundation for identifying individual trees. Furthermore, effective preprocessing helps the system operate stably in complex and changing real-world environments, effectively addressing issues such as changes in lighting and viewing angles, thus enhancing the robustness and practicality of the entire system.

[0099] S140. Based on the preprocessed image, extract the tree body features and environmental reference features, and establish the correspondence between the tree body features and the environmental reference features.

[0100] In this embodiment, tree ontological features refer to information extracted from tree images to describe their unique physical and visual attributes. These features include, but are not limited to, trunk texture, bark details, and key point locations of the trunk and branches, which together form the basis for identifying and distinguishing different trees.

[0101] Environmental reference features refer to the location information and relative relationships of fixed objects in the environment surrounding trees (such as streetlights, manhole covers, etc.), which helps to improve the accuracy and stability of tree identification.

[0102] In one embodiment, step S140 described above may include steps S141 to S143.

[0103] S141. Analyze the preprocessed image using a convolutional neural network to extract high-dimensional visual features, including tree trunks, bark textures, and key points, and utilize the robustness of image enhancement algorithms under multiple viewpoints and lighting conditions.

[0104] In this embodiment, a pre-trained convolutional neural network (CNN) is used to perform feature extraction on the input standardized image.

[0105] The extracted features include, but are not limited to, the texture and structural information of the trunk and bark, and the key points of the trunk and major branches.

[0106] To enhance the robustness of the model, images taken from various perspectives and under different lighting conditions are used for training or testing to ensure that the system can work stably under different environmental conditions.

[0107] Specifically, a deep learning model suitable for image feature extraction is selected, and these models have performed well in many computer vision tasks. The intermediate layers of the CNN are designed with specific convolutional and pooling layers to capture local details of the trees, such as trunk texture and bark structure. Meanwhile, high-level feature maps are used to identify key points on the trunk and major branches. Transfer learning techniques are employed, using a model pre-trained on a large-scale natural image dataset as a foundation, and then fine-tuning it to meet the needs of tree feature extraction.

[0108] Low-level features, such as edges and color variations, are extracted through convolutional layers. Then, multiple convolutional layers and non-linear activation functions are combined to capture more complex patterns, such as the unique texture of the bark. Heatmap regression or direct regression methods are applied to predict the coordinates of key points on the trunk and major branches. This step facilitates subsequent 3D tree modeling and growth monitoring. Combining feature maps from different levels forms a multi-level, multi-scale feature representation, thus comprehensively describing the morphological characteristics of the tree.

[0109] S142. Identify and calculate the three-dimensional coordinates of fixed reference objects in the environment involved in the preprocessed image, and determine the relative distance and angle between the trees and the fixed reference objects by combining the heading angle, thus forming environmental reference object features.

[0110] In this embodiment, a target detection algorithm is used to identify fixed reference objects (such as streetlights, manhole covers, etc.) in the image, and their three-dimensional coordinates are estimated by stereo vision or other depth perception technologies.

[0111] Calculate the relative distance and angle between the tree and each reference point based on the vehicle's heading angle and current position.

[0112] The above information is integrated into environmental reference features, which serve as supplementary data for subsequent feature comparison and identity verification processes.

[0113] Specifically, object detection frameworks are used to identify fixed reference objects (such as streetlights, manhole covers, etc.) in images. These algorithms can achieve real-time detection while maintaining accuracy. A unique category label is assigned to each type of reference object, and the training set contains a sufficient number of samples to cover all possible scenarios.

[0114] Based on the vehicle's heading angle θ, the relative orientation between the trees and the reference object is adjusted. Specifically, this involves converting the observations from the vehicle coordinate system to the geodetic coordinate system.

[0115] Calculate the straight-line distance between the base of the tree and the center of the camera; determine the angle of the line connecting the tree and the vehicle relative to the vehicle's heading.

[0116] Based on the above calculation results, combined with the vehicle's current position P vehicle The absolute latitude and longitude coordinates P of the tree are obtained through spatial coordinate transformation. tree .

[0117] All the information obtained from the above calculations (such as the 3D coordinates of the reference object, the relative distance and angle between the tree and the reference object, etc.) are encoded into a unified format for easy subsequent processing. The newly generated environmental reference object features, along with the corresponding tree features, are stored in the database for subsequent identification and feature comparison.

[0118] Through the detailed steps described above, the system can not only accurately extract the high-dimensional visual features of trees themselves, but also effectively utilize fixed reference information in the surrounding environment to construct a stable and reliable tree identification system. This method significantly improves the accuracy and reliability of tree identification, and is particularly suitable for applications involving long-term monitoring and management of urban greening resources.

[0119] S143. Use spatial relationship vectors to associate the tree body features and the environmental reference features to construct the relative positional relationship between the two.

[0120] In this embodiment, a spatial relation vector R is defined. te It represents the tree's intrinsic feature F. t Features of its surrounding environment and reference objects F e The relative positional relationship between them.

[0121] Although not every feature point is precisely matched during this process, this relative relationship helps the system to accurately locate and identify specific individual trees when faced with tree growth, damage, or environmental changes.

[0122] The resulting comprehensive feature template not only contains rich ontology information but also integrates reliable environmental reference data, greatly improving the adaptability and reliability of the entire system.

[0123] Through these three steps, the system can effectively extract the unique characteristics of trees and closely integrate them with their surrounding environment, thereby achieving long-term and effective identification and tracking management of individual trees. This method overcomes the limitations of traditional methods that rely on a single identifier, improving the accuracy and stability of tree identification.

[0124] S150. Determine whether the tree is a coded tree based on the tree's geographical coordinates, the tree's physical characteristics, and the environmental reference features.

[0125] In one embodiment, step S150 described above may include steps S151 to S152.

[0126] S151. Based on the geographical coordinates of the trees, a candidate tree set is determined by global similarity comparison.

[0127] In this embodiment, the candidate tree set refers to the set of all existing coded tree records that are close to the current tree location (e.g., within a 5-meter radius) when performing global similarity comparison based on tree geographic coordinates, selected from the database.

[0128] The geographic coordinates of trees (obtained through a high-precision positioning system) are used to quickly filter out a set of historical records that may match the current tree, i.e., a candidate tree set.

[0129] During the data acquisition phase, the vehicle's centimeter-level absolute geographic coordinates P are obtained using a GNSS module (supporting RTK differential positioning). vehicle And the precise heading angle θ.

[0130] The relative distance d and the included angle α between the tree and the vehicle are calculated using a visual algorithm that combines the vehicle's position and heading angle. This is then converted into the tree's absolute latitude and longitude coordinates P. tree .

[0131] The system will use the calculated tree geographic coordinates to query all existing records within a 5-meter radius in the tree identity database to form a preliminary set of candidate trees.

[0132] This step relies on efficient spatial indexing techniques (such as R-trees or Quadtrees) to accelerate the search process and reduce computation.

[0133] If no record is found within the specified range, the current tree is considered a newly discovered tree, triggering the creation process. If multiple potential matches exist, further analysis of the detailed characteristics of these candidates is required for confirmation.

[0134] S152. When a candidate tree set exists, a refined comparison is performed based on the tree's intrinsic features and the environmental reference features to determine whether the tree is a tree that has already been coded.

[0135] Specifically, when a candidate tree set exists, the matching degree between the tree ontology features and the environmental reference features and the tree ontology features and environmental reference features corresponding to the candidate tree set is compared to determine whether the tree is an already coded tree.

[0136] For each tree in the candidate tree set, a detailed comparison is made using its intrinsic characteristics (texture, structure, etc.) and environmental reference characteristics (positional relationship of fixed facilities) to ensure that the final confirmed tree identity is accurate.

[0137] First, the currently collected tree features (including but not limited to trunk texture, bark texture, key point location, etc.) are matched with the historical feature templates of the corresponding trees in the candidate set.

[0138] Calculate the similarity score between the two. If the score is higher than a set threshold (e.g., 90%), the two are considered to be highly consistent and tend to be considered to be the same tree.

[0139] When the ontological feature matching degree is insufficient, the features of environmental reference objects are examined instead. This includes the three-dimensional coordinates and relative positional relationships of permanent facilities around the trees (such as streetlights, manhole covers, etc.).

[0140] Compare the consistency of parameters such as distance and angle between the current tree and the candidate tree and these environmental references to assess whether they are located in the same geographical location.

[0141] If both the tree's intrinsic features and environmental features show a high degree of matching, the tree is finally confirmed as an already coded tree, and a feature fusion update operation is performed.

[0142] If only the environmental features match well but the ontological features differ significantly, it may be due to the trees being replaced or undergoing major changes. In this case, a new identity record needs to be created and associated with the old record, while marking the status code (such as "replaced / transplanted").

[0143] If neither of the two methods achieves a satisfactory match, it is considered a case of suspected transplantation and environmental change, and the comparison results are temporarily frozen for further verification.

[0144] After a successful match, the system will dynamically fuse the old and new features according to a certain weight ratio, generate an updated feature template, and store it in the database as one of the historical versions.

[0145] At the same time, adjust the status code and version code of the corresponding trees to ensure data integrity and traceability.

[0146] Through the above steps, S150 not only achieves effective identification and tracking of tree identities, but also establishes a comprehensive life cycle management system, greatly improving the efficiency and accuracy of urban greening resource management and protection.

[0147] In one embodiment, step S152 described above may include steps S1521 to S1527.

[0148] S1521. When a candidate tree set exists, calculate the matching degree between the tree ontology features and the tree ontology features corresponding to the candidate tree set to obtain the tree feature matching degree.

[0149] S1522. Determine whether the tree feature matching degree meets the requirements;

[0150] S1523. If the tree feature matching degree meets the requirements, then it is determined that the tree is a tree that has been coded.

[0151] S1524. If the tree feature matching degree does not meet the requirements, calculate the matching degree between the environmental reference feature and the environmental reference feature corresponding to the candidate tree set to obtain the environmental reference feature matching degree.

[0152] S1525. If the environmental reference feature matching degree meets the requirements, then it is determined that the tree identity has been replaced or significantly changed.

[0153] S1526. Mark the status code to generate a new tree identification code and associate it with the original tree identification code.

[0154] S1527. If the environmental reference feature matching degree does not meet the requirements, then freeze the environmental reference feature matching degree.

[0155] In this embodiment, when a candidate tree set exists, the matching degree between the ontological features of the current tree (such as trunk texture, bark texture, key points of the trunk / branch and diameter at breast height, etc.) and the corresponding ontological features of each tree in the candidate tree set is first calculated to obtain the tree feature matching degree.

[0156] Based on a pre-set threshold (e.g., 90%), determine whether the calculated tree feature matching degree reaches or exceeds this threshold. If it does, the matching degree is considered to meet the requirements; otherwise, it does not.

[0157] If the tree feature matching degree meets the requirements, it means that the currently identified tree is highly similar to a certain tree in the candidate set, and the system determines that the tree is a tree that already exists in the database.

[0158] If the tree feature matching degree does not meet the requirements, the next step is to calculate the matching degree between the current tree's environmental reference features (such as the relative position and angle with fixed references such as streetlights and guardrails) and the environmental reference features corresponding to the candidate set, in order to obtain the environmental reference feature matching degree.

[0159] If the environmental reference features match the requirements, it indicates that although the tree itself has changed (it may have grown, been pruned, or been partially damaged), the features of its surrounding environment have not changed significantly. The system then identifies the tree as having been replaced or having undergone a major change.

[0160] For trees identified as having been replaced or undergone significant changes, the system will update their status code (e.g., marked as "10 = Replaced / Transplanted"), generate a new tree identification code, and ensure that it is associated with the original tree identification code to guarantee complete lifecycle traceability.

[0161] If the matching degree of environmental reference features does not meet the requirements, that is, if the tree and its surrounding environment have changed significantly and the tree's identity cannot be clearly determined, the system will freeze the matching result and wait for further data collection or manual confirmation in order to avoid erroneous updates or confusion of tree identities.

[0162] This series of steps ensures the accuracy and reliability of tree identification, while also allowing the system to dynamically adapt to the challenges brought about by tree growth and environmental changes.

[0163] S160. If the tree is not an existing coded tree, then generate an identity code for the tree, wherein the identity code for the tree includes a basic code and a dynamic extension code; the basic code includes the code corresponding to the geographic coordinates and the sequence number; the dynamic extension code includes a status code for representing the current operation and maintenance status of the tree and a version code for recording the number of times the robust template is updated.

[0164] In this embodiment, firstly, the absolute geographic coordinates (P) of the new tree are obtained using a high-precision GNSS system. tree To ensure spatial uniqueness, latitude and longitude are converted into string encoding using Geohash encoding or other similar methods. Typically, the last six decimal places of the latitude and longitude are used for encoding to balance uniqueness and data volume. This encoding method takes into account the specific geographical location of the trees, ensuring accurate spatial positioning of each tree.

[0165] To further ensure the uniqueness of the base code, an auto-incrementing sequence number is added. This could be done by sequentially numbering all trees within a specific geographical area, ensuring that each tree has a unique base code.

[0166] Once generated by the system, the basic code is permanently bound to a specific tree and its geographical location. Regardless of subsequent template updates, environmental changes, or operational management operations, the basic code will not change. This code is equivalent to a tree's "ID number," possessing global uniqueness, stability, and immutability; it is the core identifier.

[0167] Status codes are used to indicate the current operational status of trees. Initially, if a tree is healthy and requires no special attention, the status code is set to "00" (normal). Status codes reflect the real-time status of the trees and can be adjusted according to actual needs. Other status codes include, but are not limited to:

[0168] "01": Requires testing (requires access to external information for verification, and manual intervention may be necessary).

[0169] "10": Replaced / Transplanted (Information stored in the database and needs to be re-encoded);

[0170] "11": Damaged / invalid (information stored in the database and needs to be re-encoded);

[0171] Version code: Records the number of times the robust template has been updated, reflecting the evolution history of tree features throughout its lifecycle. For new trees, the initial version code is set to "0", and it increments with subsequent identification, feature fusion, and template updates.

[0172] The status code and version number can be dynamically adjusted based on the identification results and management needs. This is similar to the file information attached to an ID card, which can be continuously updated to adapt to new situations without affecting the uniqueness and stability of the underlying code itself.

[0173] Each basic code is more than just a number; it corresponds to a complete dataset containing information such as the tree's intrinsic features (e.g., trunk texture, structure), environmental features (e.g., relative position to fixed references like streetlights and guardrails), geographic coordinates, and historical versions. When a new tree is identified, the system creates a new dataset and associates all the aforementioned relevant information with it.

[0174] Over time, whenever new identifications or features are updated, this information is added to the corresponding tree's dataset, and the status code and version code are adjusted as needed. This not only maintains the uniqueness of the tree's identity but also provides complete traceability throughout its entire lifecycle.

[0175] In this way, the identity of each tree can be effectively managed and tracked even when the tree is growing, damaged, or the environment changes, ensuring its identifiability and traceability throughout its life cycle. Furthermore, this coding mechanism provides flexibility, allowing for dynamic adjustments to the tree's status and characteristic templates based on actual needs, thus better adapting to various complex scenarios.

[0176] In this embodiment, the tree identification code is uniquely bound to the individual tree and its geographical location upon initial generation, following the principles below:

[0177] Uniqueness: Each tree corresponds to only one basic code throughout its entire life cycle, and this code will not be repeated and will not be changed.

[0178] Stability: Once the basic code is established, it will remain unchanged permanently, just like a tree's "ID number," ensuring a long-term valid identifier.

[0179] Scalability: Although the basic code is fixed, its corresponding "data information" (such as feature templates and operation and maintenance status) can be dynamically updated over time. These updates do not affect the uniqueness and stability of the basic code.

[0180] S170. If the tree is an already coded tree, the feature weights are dynamically adjusted based on the matching degree between the tree's ontological features and historical tree ontological features, as well as the matching degree between the environmental reference features and historical environmental reference features. These features are then fused with historical templates to generate new feature templates, which are recorded in the tree's identity code.

[0181] In one embodiment, step S170 described above may include steps S171 to S174.

[0182] S171. When the matching degree between the tree body feature and the historical tree body feature, and the matching degree between the environmental reference feature and the historical environmental reference feature both exceed the first set threshold, the weight corresponding to the tree body feature and the weight corresponding to the environmental reference feature are kept close to each other.

[0183] In this embodiment, when the matching degree between the tree's physical features (such as trunk texture, structure, etc.) and the tree's physical features in the historical record, as well as the matching degree between the environmental reference features (such as the relative position with fixed references such as streetlights and guardrails) and the environmental reference features in the historical record, both exceed the first set threshold.

[0184] Operation: In this case, it means that the currently collected data is highly consistent with historical data, therefore the weight W corresponding to the tree's intrinsic features is maintained. t The weight W corresponding to the characteristics of the environmental reference object eThe proportions are close to ensure the stability and accuracy of the template.

[0185] S172. If the matching degree between the environmental reference feature and the historical environmental reference feature is higher than a first set threshold, and the matching degree between the tree body feature and the historical tree body feature is lower than a second set threshold, then the weight corresponding to the environmental reference feature is increased, and the weight corresponding to the tree body feature is decreased to highlight the environmental feature.

[0186] In this embodiment, if the matching degree between the environmental reference feature and the historical environmental reference feature is higher than a first preset threshold, and the matching degree between the tree body feature and the historical tree body feature is lower than a second preset threshold, this means that although the tree may have undergone some changes (such as growth or damage), the surrounding environmental references remain unchanged or change only slightly. In this case, the weight W of the environmental reference feature should be increased. e And correspondingly reduce the weight W of the tree's intrinsic features. t This is to highlight the importance of environmental characteristics and ensure accurate tree identification even when environmental changes are minimal.

[0187] S173. If the matching degree between the tree body feature and the historical tree body feature is higher than a first set threshold, and the matching degree between the environmental reference feature and the historical environmental reference feature is lower than a second set threshold, then the weight corresponding to the tree body feature is increased, and the weight corresponding to the environmental reference feature is decreased to emphasize the tree feature.

[0188] In this embodiment, if the matching degree between the tree body feature and the historical tree body feature is higher than a first set threshold, the matching degree between the environmental reference feature and the historical environmental reference feature is lower than a second set threshold.

[0189] This indicates that despite changes in the surrounding environment (such as the addition or removal of reference points), the inherent characteristics of the trees (such as trunk texture and structure) remain highly consistent. In this case, the weight W of the tree's intrinsic features should be increased. t And reduce the weight of environmental reference features W e This emphasizes the importance of tree characteristics in the identification process.

[0190] S174. If the matching degree between the tree body feature and the historical tree body feature, and the matching degree between the environmental reference feature and the historical environmental reference feature are both lower than the second set threshold, the weight corresponding to the environmental reference feature is frozen to avoid erroneous updates.

[0191] When the matching degree between the tree's intrinsic features and historical tree intrinsic features, and the matching degree between environmental reference features and historical environmental reference features are both lower than the second set threshold.

[0192] In this scenario, there may be significant uncertainty and error risk, such as severe damage to trees or major environmental changes. To avoid erroneous template updates, the system will freeze the weights W of environmental reference features. e Only historical information is retained, pending further confirmation or manual intervention.

[0193] Once an appropriate weight allocation strategy has been determined based on the above conditions, the feature fusion formula F will then be used. m =W t ⋅F t +W e ⋅F e The fused feature template is calculated and smoothly integrated with historical templates to generate a new feature template. This new template not only reflects the currently collected information but also takes into account historical data, ensuring long-term tracking and updating of tree features.

[0194] The results of the merged update are written into the dataset corresponding to the tree's identity code, with the version code incrementing with each update. Furthermore, each collected raw data entry is stored as an independent historical record, ensuring data traceability and integrity. In this way, even when trees grow, are damaged, or their environment changes, the identity of each tree can be effectively managed and tracked, achieving uniqueness and traceability throughout its entire lifecycle.

[0195] For example, please see Figures 2 to 3 The in-vehicle system can provide centimeter-level precise positioning for the vehicle, obtaining the absolute geographic coordinates P of the vehicle's center. vehicle (X v Y v The system obtains the vehicle's heading angle θ (the angle between the vehicle's front direction and true north). After obtaining this information, the system preprocesses the image, highlighting the roadside trees in a prominent color, while other fixed reference objects (such as manhole covers, road signs, electrical boxes, streetlights, etc., about 20 types) are marked with different colors. Using a visual algorithm, the system calculates the straight-line distance d between the base of the roadside trees and the camera, as well as the relative azimuth angle α between the line connecting the tree and the vehicle's center and the vehicle's heading line.

[0196] Next, based on the principle of spatial coordinate transformation, the tree's position is converted into absolute latitude and longitude coordinates. This process ensures that the tree's precise and unique absolute position can be obtained regardless of the vehicle's orientation. Typically, the converted latitude and longitude coordinates will have 9 decimal places, but to balance uniqueness and data volume, 6 decimal places are usually used for Geohash encoding, which guarantees a planar accuracy of approximately 0.1 meters.

[0197] In addition, the model extracts features of the roadside trees themselves and environmental features from multiple sources, and dynamically allocates weights for feature fusion accordingly. When a vehicle drives near a tree, the absolute coordinates of the tree are calculated based on the angle between the vehicle and the tree, and the tree's intrinsic features, including trunk texture and key branch points, are extracted. At the same time, environmental feature information such as the relative position and horizontal angle between the tree and surrounding fixed reference objects (such as manhole covers, streetlights, etc.) is recorded.

[0198] Subsequently, as Figures 4 to 6 The system searches the database for all coded trees within a 5-meter radius of the tree's coordinates to form a candidate set. It then compares the collected tree features with those in the candidate set for similarity. If the similarity exceeds a set threshold (generally around 90% to avoid misidentification), the tree is marked as a candidate. Next, the system further compares the collected environmental features with those of the candidate trees, such as the distance from the tree to nearby reference points, the angle between the tree and the reference point, and the type of the reference point. If the environmental features of both trees also match highly, the newly collected information is considered to correspond to a tree already existing in the database.

[0199] Ultimately, the model dynamically integrates the newly collected tree and environmental features with the existing data in the database according to their weights. For example, Figure 3 Example images show vehicles collecting information from the same area at different time periods, illustrating the cases of trees T2 and T3. The model compares T2 and T3 with existing data in the database, such as T1, using the aforementioned comparison process, concluding that T2 and T3 are indeed the tree T1 in the database. Therefore, only the existing data needs to be updated, without creating new records. This method improves data consistency and accuracy while reducing the generation of redundant data.

[0200] This embodiment combines high-precision positioning information with the visual characteristics of trees to provide a novel method for identifying individual trees. This method not only avoids the limitations of relying on a single physical marker, coordinate, or image, but also enhances the stability and reliability of the identification.

[0201] During each identification process, the system automatically merges the currently collected tree features with historical data and dynamically adjusts the weights based on environmental changes and matching accuracy. When feature drift occurs (such as changes caused by tree growth or environmental shifts), the system updates the template using the latest features, thereby ensuring the accuracy and consistency of long-term tracking.

[0202] A multi-dimensional feature template is employed, including features such as the tree's texture and structure, as well as the relative position of surrounding permanent facilities. This not only improves the accuracy of identification in complex environments but also ensures the distinguishability between different trees, increasing the stability of the entire system.

[0203] A unique identification code, generated based on location, tree characteristics, and environmental reference features, ensures that each tree has a unique digital identity throughout its entire lifecycle. This code remains unchanged once generated, providing a solid foundation for tree management and research, and enabling end-to-end traceability from planting to final state.

[0204] In summary, the method in this embodiment establishes a stable and dynamically adaptable tree identification system by integrating location information and visual analysis. This solves the problems of easy loss of identification and failure over time in traditional methods. At the same time, it ensures the uniqueness and traceability of each tree throughout its entire life cycle through unique coding rules.

[0205] The aforementioned dynamic tree coding method based on high-precision positioning and visual analysis collects pose data and images using equipment installed on a vehicle. This information is then used to calculate the tree's geographic coordinates. The images are preprocessed to extract tree features and environmental reference features, establishing their correspondence. Next, based on the tree's geographic coordinates, features, and environmental characteristics, it is determined whether the tree is already in the records. If it is a newly discovered tree, an identity code is generated, consisting of a basic code (corresponding to geographic coordinates and a sequence number) and a dynamic extended code (containing a status code indicating maintenance status and a version code recording the number of robust template updates). This method achieves a stable, unique, and traceable digital identity for each tree throughout its entire lifecycle, addressing the shortcomings of existing technologies in terms of reliability, uniqueness, and long-term continuity, thereby ensuring efficient and accurate tree management.

[0206] Figure 7 This is a schematic block diagram of a tree dynamic coding system 300 based on high-precision positioning and visual analysis provided in an embodiment of the present invention. Figure 7 As shown, corresponding to the above-described tree dynamic coding method based on high-precision positioning and visual analysis, the present invention also provides a tree dynamic coding system 300 based on high-precision positioning and visual analysis. This tree dynamic coding system 300 includes a unit for executing the above-described tree dynamic coding method based on high-precision positioning and visual analysis, and the system can be configured in a server. Specifically, please refer to... Figure 7 The tree dynamic coding system 300 based on high-precision positioning and visual analysis includes an acquisition unit 301, a calculation unit 302, a preprocessing unit 303, an extraction unit 304, a comparison unit 305, and a coding generation unit 306.

[0207] The system comprises: an acquisition unit 301 for acquiring pose data and images collected by a device installed on a vehicle; a calculation unit 302 for calculating the geographic coordinates of the tree based on the pose data; a preprocessing unit 303 for preprocessing the image to obtain a preprocessed image; an extraction unit 304 for extracting tree body features and environmental reference features based on the preprocessed image, and establishing a correspondence between the tree body features and the environmental reference features; a comparison unit 305 for determining whether the tree is an already coded tree based on its geographic coordinates, tree body features, and environmental reference features; and a coding generation unit 306 for generating an identity code for the tree if it is not an already coded tree, wherein the tree identity code includes a basic code and a dynamic extension code; wherein the basic code includes the code corresponding to the geographic coordinates and a sequence number; and the dynamic extension code includes a status code indicating the current operation and maintenance status of the tree and a version code recording the number of times the robust template has been updated.

[0208] In one embodiment, the tree dynamic coding system 300 based on high-precision positioning and visual analysis further includes:

[0209] The fusion update unit 307 is used to dynamically adjust the feature weights based on the matching degree between the tree's ontological features and historical tree ontological features, and the matching degree between the environmental reference features and historical environmental reference features, if the tree is an already coded tree, and to fuse it with the historical template to generate a new feature template, which is then recorded in the tree's identity code.

[0210] In one embodiment, the comparison unit 305 includes:

[0211] The global comparison subunit is used to determine a candidate tree set based on the geographic coordinates of the tree through global similarity comparison; the refined comparison subunit is used to perform a refined comparison based on the tree's ontological features and the environmental reference features when a candidate tree set exists, in order to determine whether the tree is an already coded tree.

[0212] In one embodiment, the computing unit 302 includes:

[0213] The distance and angle calculation subunit is used to calculate the distance and azimuth angle of the tree base relative to the vehicle using a visual algorithm combined with the pose data and the image, so as to obtain the relative distance and angle; the coordinate calculation subunit is used to convert the relative distance and angle into an increment in the geodetic coordinate system using a rotation matrix, and update it to the vehicle position to determine the absolute coordinates of the tree, so as to obtain the geographic coordinates of the tree; the encoding subunit is used to encode the geographic coordinates of the tree, so as to obtain the code corresponding to the geographic coordinates.

[0214] In one embodiment, the extraction unit 304 includes:

[0215] The ontology feature extraction subunit is used to analyze the preprocessed image through a convolutional neural network, extract high-dimensional visual features including tree trunk, bark texture, and key points, and utilize the robustness of image enhancement algorithms under multiple viewpoints and lighting conditions; the environment feature extraction subunit is used to identify and calculate the three-dimensional coordinates of fixed reference objects in the environment involved in the preprocessed image, and determine the relative distance and angle between the tree and the fixed reference object by combining the heading angle, forming environmental reference object features; the relationship construction subunit is used to associate the tree ontology features and the environmental reference object features using spatial relationship vectors to construct the relative positional relationship between the two.

[0216] In one embodiment, the refined comparison subunit is used to compare the matching degree of the tree ontological features and the environmental reference features with the tree ontological features and environmental reference features corresponding to the candidate tree set when a candidate tree set exists, so as to determine whether the tree is an already coded tree.

[0217] In one embodiment, the refined comparison subunit includes:

[0218] The system includes: an ontology feature matching degree calculation module, used to calculate the matching degree between the ontology features of the tree and the ontology features of the tree corresponding to the candidate tree set when a candidate tree set exists, to obtain the tree feature matching degree; a judgment module, used to determine whether the tree feature matching degree meets the requirements; a first determination module, used to determine that the tree is an already coded tree if the tree feature matching degree meets the requirements; an environmental feature matching degree calculation module, used to calculate the matching degree between the environmental reference object features and the environmental reference object features corresponding to the candidate tree set if the tree feature matching degree does not meet the requirements, to obtain the environmental reference object feature matching degree; a second determination module, used to determine that the tree identity has been replaced or significantly changed if the environmental reference object feature matching degree meets the requirements; and a freezing module, used to freeze the environmental reference object feature matching degree if the environmental reference object feature matching degree does not meet the requirements.

[0219] In one embodiment, the refined comparison subunit further includes:

[0220] The association module is used to mark the status code to generate a new tree identification code and associate it with the original tree identification code.

[0221] In one embodiment, the fusion update unit 307 includes:

[0222] The first adjustment subunit is configured to maintain a close weight ratio between the tree's intrinsic feature and the environmental reference feature when both the matching degree between the tree's intrinsic feature and historical tree intrinsic features and the matching degree between the environmental reference feature and historical environmental reference feature exceed a first preset threshold. The second adjustment subunit is configured to increase the weight of the environmental reference feature and decrease the weight of the tree's intrinsic feature if the matching degree between the environmental reference feature and historical environmental reference feature is higher than the first preset threshold and the matching degree between the tree's intrinsic feature and historical tree intrinsic features is lower than the second preset threshold. The system emphasizes environmental features. A third adjustment subunit is used to increase the weight of the tree's physical features and decrease the weight of the environmental reference features to emphasize tree features when the matching degree between the tree's physical features and historical tree physical features is higher than a first set threshold, and the matching degree between the environmental reference features and historical environmental reference features is lower than a second set threshold. A fourth adjustment subunit is used to freeze the weight of the environmental reference features to avoid erroneous updates when both the matching degree between the tree's physical features and historical tree physical features and the matching degree between the environmental reference features and historical environmental reference features are lower than the second set threshold.

[0223] It should be noted that those skilled in the art can clearly understand that the specific implementation process of the tree dynamic coding system 300 based on high-precision positioning and visual analysis and each unit can be referred to the corresponding description in the foregoing method embodiments. For the sake of convenience and brevity, it will not be repeated here.

[0224] The aforementioned tree dynamic coding system 300 based on high-precision positioning and visual analysis can be implemented as a computer program, which can be used in various ways, such as... Figure 8 It runs on the computer device shown.

[0225] Please see Figure 8 , Figure 8 This is a schematic block diagram of a computer device provided in an embodiment of this application. The computer device 500 can be a server, wherein the server can be a standalone server or a server cluster composed of multiple servers.

[0226] See Figure 8 The computer device 500 includes a processor 502, a memory, and a network interface 505 connected via a system bus 501. The memory may include a non-volatile storage medium 503 and internal memory 504.

[0227] The non-volatile storage medium 503 may store an operating system 5031 and a computer program 5032. The computer program 5032 includes program instructions that, when executed, cause the processor 502 to execute a dynamic tree encoding method based on high-precision positioning and visual analysis.

[0228] The processor 502 provides computing and control capabilities to support the operation of the entire computer device 500.

[0229] The internal memory 504 provides an environment for the operation of the computer program 5032 in the non-volatile storage medium 503. When the computer program 5032 is executed by the processor 502, the processor 502 can execute a tree dynamic coding method based on high-precision positioning and visual analysis.

[0230] This network interface 505 is used for network communication with other devices. Those skilled in the art will understand that... Figure 8 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device 500 to which the present application is applied. The specific computer device 500 may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0231] The processor 502 is used to run the computer program 5032 stored in the memory to implement all the steps of the tree dynamic coding method based on high-precision positioning and visual analysis.

[0232] It should be understood that in the embodiments of this application, the processor 502 may be a central processing unit (CPU), or it may be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor.

[0233] It will be understood by those skilled in the art that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program includes program instructions and can be stored in a storage medium, which is a computer-readable storage medium. The program instructions are executed by at least one processor in the computer system to implement the process steps of the embodiments of the above methods.

[0234] Therefore, the present invention also provides a storage medium. This storage medium can be a computer-readable storage medium. The storage medium stores a computer program, wherein when executed by a processor, the computer program causes the processor to perform all the steps of the tree dynamic encoding method based on high-precision positioning and visual analysis.

[0235] The storage medium can be any computer-readable storage medium capable of storing program code, such as a USB flash drive, portable hard drive, read-only memory (ROM), magnetic disk, or optical disk.

[0236] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0237] In the embodiments provided by this invention, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the system embodiments described above are merely illustrative. For example, the division of each unit is only a logical functional division, and there may be other division methods in actual implementation. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed.

[0238] The steps in the method of this invention can be adjusted, merged, or reduced in order according to actual needs. The units in the system of this invention can be merged, divided, or reduced according to actual needs. Furthermore, the functional units in the various embodiments of this invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0239] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, a terminal, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention.

[0240] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in the present invention, and these modifications or substitutions should all be covered within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A tree dynamic coding method based on high-precision positioning and visual analysis, characterized in that, include: Acquire pose data and images collected by devices installed on the vehicle; Calculate the geographic coordinates of the tree based on the pose data; The image is preprocessed to obtain a preprocessed image; Based on the preprocessed image, tree body features and environmental reference features are extracted, and the correspondence between the tree body features and the environmental reference features is established. The determination of whether a tree is an already coded tree is based on its geographic coordinates, tree characteristics, and environmental reference features. Specifically, a candidate tree set is determined by global similarity comparison based on the tree's geographic coordinates. When a candidate tree set exists, a refined comparison is performed based on the tree characteristics and environmental reference features to determine whether the tree is an already coded tree. If the tree is not an existing coded tree, an identity code for the tree is generated. The identity code for the tree includes a basic code and a dynamic extension code. The basic code includes the code corresponding to the geographic coordinates and a sequence number. The dynamic extension code includes a status code for indicating the current operation and maintenance status of the tree and a version code for recording the number of times the robust template has been updated.

2. The tree dynamic coding method based on high-precision positioning and visual analysis according to claim 1, characterized in that, After determining whether a tree is an already coded tree based on its geographical coordinates, its intrinsic features, and the features of the environmental reference points, the process further includes: If the tree is an already coded tree, the feature weights are dynamically adjusted based on the matching degree between the tree's ontological features and historical tree ontological features, as well as the matching degree between the environmental reference features and historical environmental reference features. These features are then fused with historical templates to generate new feature templates, which are recorded in the tree's identity code.

3. The tree dynamic coding method based on high-precision positioning and visual analysis according to claim 1, characterized in that, The calculation of the tree's geographic coordinates based on the pose data includes: A visual algorithm is used to combine the pose data and the image to calculate the distance and azimuth angle of the tree base relative to the vehicle, so as to obtain the relative distance and angle. The relative distance and angle are converted into increments in the geodetic coordinate system using a rotation matrix and updated to the vehicle position to determine the absolute coordinates of the tree, thus obtaining the tree's geographic coordinates. The geographic coordinates of the trees are encoded to obtain the corresponding codes.

4. The tree dynamic coding method based on high-precision positioning and visual analysis according to claim 1, characterized in that, The step of extracting tree features and environmental reference features from the preprocessed image and establishing the correspondence between the tree features and the environmental reference features includes: The preprocessed image is analyzed by a convolutional neural network to extract high-dimensional visual features, including tree trunk, bark texture and key points, and the robustness of the image enhancement algorithm under multiple viewpoints and lighting conditions is utilized. Identify and calculate the three-dimensional coordinates of fixed reference objects in the environment involved in the preprocessed image, and determine the relative distance and angle between the trees and the fixed reference objects by combining the heading angle, thus forming environmental reference object features; The spatial relationship vector is used to associate the tree's physical features with the environmental reference features to construct the relative positional relationship between the two.

5. The tree dynamic coding method based on high-precision positioning and visual analysis according to claim 4, characterized in that, When a candidate tree set exists, a refined comparison is performed based on the tree's intrinsic features and the environmental reference features to determine whether the tree is an already coded tree, including: When a candidate tree set exists, the matching degree of the tree ontology features and the environmental reference features with the tree ontology features and environmental reference features corresponding to the candidate tree set is compared to determine whether the tree is an already coded tree.

6. The tree dynamic coding method based on high-precision positioning and visual analysis according to claim 5, characterized in that, When a candidate tree set exists, comparing the matching degree between the tree ontological features and the environmental reference features and the tree ontological features and environmental reference features corresponding to the candidate tree set to determine whether the tree is an already coded tree includes: When a candidate tree set exists, the matching degree between the tree ontology features and the tree ontology features corresponding to the candidate tree set is calculated to obtain the tree feature matching degree. Determine whether the tree feature matching degree meets the requirements; If the tree feature matching degree meets the requirements, then the tree is determined to be a tree that has already been coded; If the tree feature matching degree does not meet the requirements, the matching degree between the environmental reference feature and the environmental reference feature corresponding to the candidate tree set is calculated to obtain the environmental reference feature matching degree. If the environmental reference features match the requirements, then it is determined that the tree's identity has been replaced or significantly altered. If the environmental reference feature matching degree does not meet the requirements, then the environmental reference feature matching degree is frozen.

7. The tree dynamic coding method based on high-precision positioning and visual analysis according to claim 6, characterized in that, After determining whether the tree's identity has been replaced or significantly altered, the process further includes: The status code is marked to generate a new tree identification code, which is then associated with the original tree identification code.

8. The tree dynamic coding method based on high-precision positioning and visual analysis according to claim 1, characterized in that, The process of dynamically adjusting feature weights based on the matching degree between the tree's intrinsic features and historical tree intrinsic features, and the matching degree between the environmental reference features and historical environmental reference features, and fusing them with historical templates to generate new feature templates, which are then recorded in the tree's identity encoding, includes: When the matching degree between the tree body feature and the historical tree body feature, and the matching degree between the environmental reference feature and the historical environmental reference feature both exceed the first set threshold, the weight corresponding to the tree body feature and the weight corresponding to the environmental reference feature are kept close in proportion. If the matching degree between the environmental reference feature and the historical environmental reference feature is higher than a first set threshold, and the matching degree between the tree body feature and the historical tree body feature is lower than a second set threshold, then the weight corresponding to the environmental reference feature is increased, and the weight corresponding to the tree body feature is decreased to highlight the environmental feature. If the matching degree between the tree feature and the historical tree feature is higher than a first set threshold, and the matching degree between the environmental reference feature and the historical environmental reference feature is lower than a second set threshold, then the weight corresponding to the tree feature is increased, and the weight corresponding to the environmental reference feature is decreased to emphasize the tree feature. If the matching degree between the tree body feature and the historical tree body feature, and the matching degree between the environmental reference feature and the historical environmental reference feature are both lower than the second set threshold, the weight corresponding to the environmental reference feature is frozen to avoid erroneous updates.

9. A tree dynamic coding system based on high-precision positioning and visual analysis, characterized in that, include: The acquisition unit is used to acquire pose data and images collected by the device installed on the vehicle; A calculation unit is used to calculate the geographic coordinates of the tree based on the pose data; A preprocessing unit is used to preprocess the image to obtain a preprocessed image; The extraction unit is used to extract tree body features and environmental reference features based on the preprocessed image, and to establish the correspondence between the tree body features and the environmental reference features. The comparison unit is used to determine whether a tree is an already coded tree based on the tree's geographical coordinates, the tree's intrinsic features, and the environmental reference features. Specifically, it determines a candidate tree set based on the tree's geographical coordinates through global similarity comparison. When a candidate tree set exists, it performs a refined comparison based on the tree's intrinsic features and the environmental reference features to determine whether the tree is an already coded tree. The encoding generation unit is used to generate an identity code for the tree if the tree is not a tree with an existing code. The identity code of the tree includes a basic code and a dynamic extension code. The basic code includes the code corresponding to the geographic coordinates and a sequence number. The dynamic extension code includes a status code for indicating the current operation and maintenance status of the tree and a version code for recording the number of times the robust template has been updated.

Images