Tree feature monitoring method, device and equipment and storage medium

Abstract

The invention discloses a tree feature monitoring method, device and equipment and a storage medium, and the scheme can comprise the steps: determining a to-be-monitored region range for monitoring tree features, and enabling the to-be-monitored region range to comprise a plurality of to-be-monitored trees which are distributed in a scattered manner; the method comprises the following steps: acquiring point cloud data of an environment including a tree to be monitored in a geographic range by using a horizontally mounted multi-line laser radar; extracting point cloud features from the point cloud data; matching front and back frames of the obtained point cloud features, calculating the position and attitude change of the subsequent frame relative to the first frame, and taking the first frame data in the point cloud data as an original point to obtain the pose track point of the motion path of the multi-line laser radar in the to-be-monitored area range; and after the single-frame scanning data matched with the target position is obtained, obtaining tree features of a specific to-be-monitored tree in a specific area based on the single-frame scanning data matched with the target position.

Description

Technical Field

[0001] The present invention relates to the technical field of tree characteristic monitoring, and in particular to a tree characteristic monitoring method, device, equipment and storage medium. Background Art

[0002] In the field of forestry monitoring and perception recognition, there are strict requirements for the monitoring and analysis of important indicators such as tree monomer segmentation and breast diameter, especially the accuracy of the dynamic observation of breast diameter changes is required to be within 3cm. Existing Solution 1: Mobile measurement solution, using vehicle-mounted or airborne, backpack laser radar for scanning, generating point cloud map data, on this basis, tree segmentation and breast diameter measurement are performed. This solution uses a typical mobile measurement method. Although it is based on different carriers, the accuracy has different degrees of loss. Different sensor integration solutions will introduce integration errors and calibration errors. The accuracy is difficult to guarantee, especially in the area where the signal is missing. The accuracy loss is serious and the measurement requirements cannot be guaranteed. Existing Solution 2: Static station scanning solution, using the traditional station laser scanning method, setting up targets for later splicing, obtaining point cloud map data, and measuring on this basis. This solution uses a station scanning method, and requires additional deployment of special target objects, and later special software for point cloud splicing. The work is cumbersome and inefficient, and the professional difficulty is high. Summary of the invention

[0003] The present invention provides a method, device, equipment and storage medium for monitoring tree characteristics, so as to overcome at least one technical problem existing in the prior art.

[0004] According to a first aspect of an embodiment of the present invention, a tree feature monitoring method is provided, comprising:

[0005] Determine the scope of a to-be-monitored area for monitoring tree characteristics, wherein the to-be-monitored area includes a plurality of dispersed trees to be monitored, and use a horizontally installed multi-line laser radar to collect point cloud data of an environment including the trees to be monitored within the geographic scope, wherein the point cloud data includes single-frame scan data of the trees to be monitored collected by the multi-line laser radar, and the single-frame scan data carries timestamp information when the single-frame scan data is obtained;

[0006] Extracting point cloud features from the point cloud data, wherein the extracted point cloud features include geometric line features or surface features;

[0007] Match the previous and next frames of the point cloud features obtained, calculate the position and posture changes of the subsequent frames relative to the first frame, and take the first frame data in the point cloud data as the origin to obtain the posture trajectory points of the motion path of the multi-line laser radar in the area to be monitored; at the same time, construct a global point cloud map of the environment within the area to be monitored based on the point cloud data;

[0008] When it is necessary to view the tree features of a specific area or a specific tree to be monitored in the global point cloud map, firstly, according to the location information of the monitored target, the time point information or spatial position information corresponding to the target is located on the trajectory in the constructed global point cloud map; the original point cloud data collected by the time point information or spatial position information is indexed by the posture information on the trajectory, so as to obtain a single frame scanning data matching the target position;

[0009] The tree characteristics of the specific area or the specific trees to be monitored are obtained based on the single frame scanning data matching the target position.

[0010] Preferably, the calculating of the position and posture changes of the subsequent frames relative to the first frame specifically includes:

[0011] Extract geometric features from the scanned point cloud data, including geometric line and surface features;

[0012] Match the features of the current frame with the features of the first frame to find the corresponding relationship between the two frames of data;

[0013] By processing the matched feature points, applying the RANSAC algorithm to remove outliers, and then using the matched feature points to calculate the rotation and translation parameters, the posture transformation of the subsequent frames relative to the first frame is obtained. The transformation is represented by a rigid body transformation matrix or a quaternion representing the posture and position offset.

[0014] Preferably, the locating the time point information or spatial position information corresponding to the target on the trajectory in the constructed global point cloud map specifically includes:

[0015] Obtaining acquisition time information or position information of specific point cloud data corresponding to the specific area or the specific tree to be monitored from the global point cloud map;

[0016] The time point information is obtained based on the time information, or the spatial position information is obtained based on the posture information.

[0017] According to a second aspect of an embodiment of the present invention, there is provided a tree feature monitoring device, comprising:

[0018] A point cloud data acquisition module is used to determine the scope of a to-be-monitored area for monitoring tree characteristics, wherein the to-be-monitored area includes a plurality of dispersed trees to be monitored, and to use a horizontally installed multi-line laser radar to collect point cloud data of an environment including the trees to be monitored within the geographic scope, wherein the point cloud data includes single-frame scan data of the trees to be monitored collected by the multi-line laser radar, and the single-frame scan data carries timestamp information when the single-frame scan data is obtained;

[0019] A feature extraction module, used to extract point cloud features from the point cloud data, where the extracted point cloud features include geometric line features or surface features;

[0020] A point cloud map construction module is used to match the previous and next frames of the obtained point cloud features, calculate the position and posture changes of the subsequent frames relative to the first frame, and use the first frame data in the point cloud data as the origin to obtain the position and posture trajectory points of the motion path of the multi-line laser radar in the range of the area to be monitored; at the same time, a global point cloud map of the environment within the range of the area to be monitored is constructed based on the point cloud data;

[0021] The single-frame scanning data acquisition module is used for, when it is necessary to view the tree features of a specific area or a specific tree to be monitored in the global point cloud map, first locating the time point information or spatial position information corresponding to the target on the trajectory in the constructed global point cloud map according to the position information of the monitored target; using the posture information on the trajectory to index the original point cloud data collected by the time point information or spatial position information, so as to obtain the single-frame scanning data matching the target position;

[0022] The tree feature acquisition module is used to obtain the tree features of the specific area or the specific trees to be monitored based on the single frame scanning data matching the target position.

[0023] According to a third aspect of an embodiment of the present invention, a tree feature monitoring device is provided, which is used to execute the tree feature monitoring method described above.

[0024] According to a fourth aspect of an embodiment of the present invention, there is provided a computer-readable medium having computer-readable instructions stored thereon, wherein the computer-readable instructions can be executed by a processor to implement the tree feature monitoring method as described in any of the preceding items.

[0025] One embodiment of the present specification can achieve at least the following beneficial effects: high-precision monitoring of tree features can be achieved. Specifically, point cloud data within the area to be monitored is collected using a horizontally installed multi-line laser radar, and geometric line and surface features are extracted from these data. Then, front and back frame matching and posture transformation calculations are performed to construct a global point cloud map. When it is necessary to view the features of a specific area or a specific tree, the corresponding single-frame scan data can be called based on the constructed map and trajectory information, thereby achieving local high-precision measurement. This not only improves the overall efficiency of tree monitoring, but also ensures that the measurement accuracy of key parameters such as breast diameter is within 3 cm, meeting the needs for high-precision monitoring of tree features. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0027] Figure 1 A flowchart of a tree feature monitoring method provided in an embodiment of this specification;

[0028] Figure 2 The embodiments of this specification provide corresponding to Figure 1 A structural schematic diagram of a tree characteristic monitoring device;

[0029] Figure 3 The embodiments of this specification provide corresponding to Figure 1 A structural schematic diagram of a tree characteristic monitoring device. DETAILED DESCRIPTION

[0030] In order to make the purpose, technical solutions and advantages of one or more embodiments of this specification clearer, the technical solutions of one or more embodiments of this specification will be clearly and completely described below in combination with the specific embodiments of this specification and the corresponding drawings. Obviously, the described embodiments are only part of the embodiments of this specification, not all of the embodiments. Based on the embodiments in this specification, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of one or more embodiments of this specification.

[0031] It should be understood that although the terms first, second, third, etc. may be used in this application document to describe various information, this information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other.

[0032] In order to solve the defects in the prior art, this solution provides the following embodiments:

[0033] Figure 1 This is a flow chart of a tree feature monitoring method applied to a service platform provided in an embodiment of this specification. From a program perspective, the execution subject of the flow can be a program installed on a server of the service platform.

[0034] like Figure 1 As shown, the process may include the following steps.

[0035] Step 102: Determine the scope of a monitored area for monitoring tree characteristics, wherein the monitored area includes a plurality of dispersed trees to be monitored, and use a horizontally installed multi-line laser radar to collect point cloud data of the environment within the geographical range including the trees to be monitored, wherein the point cloud data includes single-frame scan data of the trees to be monitored collected by the multi-line laser radar, and the single-frame scan data carries timestamp information when the single-frame scan data is obtained.

[0036] In this step, the steps of using a horizontal multi-line laser radar for data collection can be as follows: 1. Configure and deploy the multi-line laser radar, ensure that it is installed horizontally to cover the required monitoring area, and the multi-line laser radar can emit laser beams from different angles to capture the three-dimensional information of the surrounding environment. 2. Turn on the radar for scanning, it will emit lasers and receive the reflected signals, calculate the spatial coordinates of each point according to the round-trip time and incident angle of the beam, and form the original point cloud data. 3. In the process of data collection, it is not limited to a specific carrier or application scenario, and can be fixedly installed or carried on a mobile platform such as a drone or vehicle for dynamic scanning. 4. Ensure that the motion trajectory of the radar is synchronized with the timestamp of the point cloud data during the collection process, which is crucial for subsequent point cloud processing and precision enhancement. Through these steps, the basic point cloud data of trees and their environment can be obtained, laying the foundation for further feature extraction, matching and local precision enhancement. In the above steps, ensuring that the multi-line laser radar is installed horizontally is to fully cover the required monitoring area, thereby effectively capturing the three-dimensional information of trees and their environment in the area. This ensures that the laser beams emitted from different angles can scan the target area evenly, improving the quality and integrity of data collection.

[0037] The reason why the technical solution of this application uses multi-line laser radar is that it can provide richer scanning data, realize the enhancement of local features of trees and fast data collection. This helps to generate local tree features associated with the global point cloud map, so as to perform accurate measurements and meet the needs of high-precision monitoring of trees, especially in the dynamic observation of important indicators such as breast diameter, which requires an accuracy of less than 3 cm. Therefore, the use of multi-line laser radar is a key technical means to achieve the combination of efficient collection and high-precision measurement.

[0038] Step 104: extracting point cloud features from the point cloud data, where the extracted point cloud features include geometric line features or surface features.

[0039] In this step, geometric line features in point cloud data processing generally refer to geometric features with linear structures identified and extracted from point clouds. These features are manifested as a series of point sets in point cloud data, which are roughly arranged in the form of straight lines or approximate straight lines in space, such as the edges of tree branches, corners of buildings, etc. In the technical solution of the present application, the extracted line features are used to assist subsequent point cloud matching and pose estimation, thereby achieving high-precision local point cloud imaging. Geometric surface features refer to surface features identified and extracted in point cloud data. These features generally represent the plane or curved surface parts of the scanned object. In the technical solution of the present application, by identifying these geometric surface features, it can assist in the processing and analysis of point cloud data. For example, in tree feature monitoring, surface features may involve circular cross-sections of tree trunks, plane interfaces of branch bifurcations, etc. This information is crucial for accurately measuring parameters such as the diameter at breast height of trees.

[0040] In this step, feature extraction of the scanned data is mainly performed by identifying and separating the geometric line and surface features in the point cloud. This involves using algorithms to analyze the point cloud data to find the edges, planes, and other meaningful structural information, which are essential for subsequent matching and positioning. The specific extraction process may include techniques such as curvature-based segmentation methods or the use of RANSAC (random sampling consensus) algorithms to fit planes and line segments, so as to accurately extract parts with significant shape features from the original point cloud.

[0041] Step 106: Match the previous and next frames of the point cloud features obtained, and calculate the position and posture changes of the subsequent frames relative to the first frame, and use the first frame data in the point cloud data as the origin to obtain the posture trajectory points of the motion path of the multi-line laser radar in the area to be monitored; at the same time, construct a global point cloud map of the environment within the area to be monitored based on the point cloud data.

[0042] In this step, the front and back frame matching operation is first performed, that is, the point cloud features of the current frame (i.e., the newly acquired frame) are matched with the point cloud features of the first frame (or the previous frame) to find the corresponding relationship between the two frames of data. This is done by comparing the position and shape of the feature points. Then the pose transformation calculation is performed, that is, the matched feature points are processed, such as using the RANSAC algorithm to remove outliers, and then the position change (translation) and attitude change (rotation) of the subsequent frames relative to the first frame are calculated based on these reliable matching points. Finally, a rigid body transformation matrix or quaternion representation of the attitude and position offset is obtained, which describes the pose trajectory points of the multi-line laser radar on the motion path within the area to be monitored. As the above process is repeated, the first frame data is used as the origin, and the data of all frames are gradually accumulated, and they are spliced ​​together to form a complete global point cloud map of the environment in the area to be monitored. This map not only shows the overall structure of the entire area, but also provides a basis for more accurate positioning of specific targets in the future.

[0043] In this step, the trajectory points describe the trajectory of the attitude change of the LiDAR relative to the initial origin during the data acquisition process. These points record the position and direction information of the scanner at different time points. They are obtained by matching the previous and next frames of the point cloud data and are used to reconstruct the motion trajectory of the scanner.

[0044] The global point cloud map generated in this step is considered "non-high-precision" because it is mainly used for scene reconstruction and browsing, focusing on providing an overall view of the environment. This kind of map may sacrifice a certain degree of detail accuracy when integrating large-scale data in order to process and display large-area point cloud information more quickly. In short, in order to improve processing speed and efficiency, when building a global point cloud map, the system may not retain the highest accuracy of all original collected data, but speed up the processing process by appropriately simplifying it. That is, in order to improve processing speed and efficiency, it is sometimes necessary to sacrifice a certain degree of detail accuracy to process and display large-area point cloud information more quickly. This practice makes the generated map more suitable for the overall reconstruction and browsing of the scene, but not necessarily for high-precision measurement of specific objects. However, for parts that require high-precision measurement, local accuracy enhancement can be achieved by calling single-frame point cloud data at specific locations, thereby meeting the strict requirements for the accuracy of tree features such as tree diameter monitoring.

[0045] Step 108: When it is necessary to view the tree features of a specific area or a specific tree to be monitored within the global point cloud map, first locate the time point information or spatial position information corresponding to the target on the trajectory in the constructed global point cloud map according to the position information of the monitored target; use the posture information on the trajectory to index the original point cloud data collected by the time point information or spatial position information, so as to obtain a single frame scanning data matching the target position.

[0046] In this step, the location information of the monitoring target is first determined, that is, the specific location of the specific area or the specific tree to be monitored that needs to be checked is first clarified. Then, the location is located in the global point cloud map, that is, according to the known location of the monitoring target, the time point information or spatial position information corresponding to the target is found on the trajectory in the global point cloud map that has been constructed. This step uses the previously calculated posture trajectory points to achieve precise positioning. Then, the original point cloud data is indexed, that is, the posture information on the trajectory (that is, the posture trajectory points mentioned in step 106) is used to directly index the original point cloud data collected at the corresponding time point or spatial position from the timestamp synchronized data, so as to obtain a single-frame scan data that accurately matches the target position. Finally, the tree features are obtained. Specifically, based on the above-obtained single-frame scan data that matches the target position, further analysis and processing are performed to extract the detailed features of the specific area or the specific tree to be monitored.

[0047] Step 110: Obtain tree features of the specific area or the specific trees to be monitored based on the single-frame scan data matching the target position.

[0048] In this step, the tree characteristics of a specific area or a specific tree to be monitored are obtained based on a single frame of scan data that matches the target position, because this frame of data is directly derived from the high-resolution scan result of the tree at a precise moment. By using the posture information on the trajectory to index the original point cloud data collected at that time point or spatial position, it can be ensured that the acquired data has local high-precision characteristics, so that it can accurately reflect the specific characteristics of the tree, such as parameters such as breast diameter.

[0049] In the above technical solution, the process of calling out the corresponding single-frame point cloud using the trajectory throughout the whole process involves the following aspects: 1. The trajectory is associated with the timestamp: because in the data acquisition process, each frame of point cloud data has a corresponding timestamp, which enables each point cloud data to be associated with the precise moment of scanning. 2. Locating a specific frame: when it is necessary to call a high-precision point cloud of a specific area or tree, first locate the time point or spatial position corresponding to the target on the trajectory in the constructed non-high-precision point cloud map according to the location information of the monitoring target or the preset monitoring point. 3. Extracting a single-frame point cloud: using the posture information on the trajectory (i.e., the posture trajectory point mentioned in step 106), the original point cloud data collected at the time point or spatial position can be directly indexed. In this way, a single-frame high-precision point cloud that accurately matches the target position is "called out". 4. Local high-precision processing: Since the frame of point cloud that is "called out" is directly derived from the original high-resolution scan, it retains the local high-precision information, which is suitable for fine measurement of tree diameter at breast height, etc., and meets the needs of local precision enhancement. In this way, although the overall map may not be high-precision, by calling the single-frame point cloud corresponding to a specific trajectory point, high-quality data can be obtained in local areas that require high-precision measurement, thereby achieving the purpose of efficient and accurate tree feature monitoring.

[0050] Figure 1 The method in can achieve high-precision monitoring of tree characteristics. Specifically, by using a horizontally installed multi-line lidar to collect point cloud data in the area to be monitored, and extracting geometric line and surface features from these data, and then performing front and back frame matching and pose transformation calculations to construct a global point cloud map. When it is necessary to view the characteristics of a specific area or a specific tree, the corresponding single-frame scan data can be called based on the constructed map and trajectory information to achieve local high-precision measurement. This not only improves the overall efficiency of tree monitoring, but also ensures that the measurement accuracy of key parameters such as breast diameter is within 3cm, meeting the needs of high-precision monitoring of tree characteristics.

[0051] based on Figure 1 The method, the examples of this specification also provide some specific implementation plans of the method, which are described below.

[0052] In an optional embodiment, the calculating of the position and posture changes of the subsequent frames relative to the first frame may specifically include:

[0053] Extract geometric features from the scanned point cloud data, including geometric line and surface features;

[0054] Match the features of the current frame with the features of the first frame to find the corresponding relationship between the two frames of data;

[0055] By processing the matched feature points, applying the RANSAC algorithm to remove outliers, and then using the matched feature points to calculate the rotation and translation parameters, the posture transformation of the subsequent frames relative to the first frame is obtained. The transformation is represented by a rigid body transformation matrix or a quaternion representing the posture and position offset.

[0056] In an optional embodiment, locating the time point information or spatial position information corresponding to the target on the trajectory in the constructed global point cloud map may specifically include:

[0057] Obtaining acquisition time information or position information of specific point cloud data corresponding to the specific area or the specific tree to be monitored from the global point cloud map;

[0058] The time point information is obtained based on the time information, or the spatial position information is obtained based on the posture information.

[0059] Based on the same idea, the embodiments of this specification also provide a device corresponding to the above method. Figure 2 The embodiments of this specification provide corresponding to Figure 1 A schematic diagram of the structure of a tree characteristic monitoring device. Figure 2 As shown, the device may include:

[0060] The point cloud data acquisition module 202 is used to determine the scope of the area to be monitored for monitoring tree characteristics, wherein the scope of the area to be monitored includes a plurality of dispersed trees to be monitored, and a horizontally installed multi-line laser radar is used to collect point cloud data of the environment including the trees to be monitored within the geographical scope, wherein the point cloud data includes single-frame scan data of the trees to be monitored collected by the multi-line laser radar, and the single-frame scan data carries timestamp information when the single-frame scan data is obtained.

[0061] The feature extraction module 204 is used to extract point cloud features from the point cloud data. The extracted point cloud features include geometric line features or surface features.

[0062] The point cloud map construction module 206 is used to match the previous and next frames of the point cloud features obtained, calculate the position and posture changes of the subsequent frames relative to the first frame, and use the first frame data in the point cloud data as the origin to obtain the posture trajectory points of the motion path of the multi-line laser radar in the area to be monitored; at the same time, a global point cloud map of the environment within the area to be monitored is constructed based on the point cloud data.

[0063] The single-frame scanning data acquisition module 208 is used to locate the time point information or spatial position information corresponding to the target on the trajectory in the constructed global point cloud map according to the position information of the monitoring target when it is necessary to view the tree characteristics of a specific area or a specific tree to be monitored in the global point cloud map; and use the posture information on the trajectory to index the original point cloud data collected by the time point information or spatial position information, so as to obtain the single-frame scanning data matching the target position.

[0064] The tree feature acquisition module 210 is used to obtain the tree features of the specific area or the specific trees to be monitored based on the single frame scanning data matching the target position.

[0065] It is understood that the above modules refer to computer programs or program segments for executing one or more specific functions. In addition, the distinction between the above modules does not mean that the actual program codes must also be separated.

[0066] Based on the same idea, the embodiments of this specification also provide a device corresponding to the above method.

[0067] Figure 3 Schematic diagram of a tree feature monitoring device provided in the embodiment of this specification. Figure 3 As shown, this hardware device may include:

[0068] at least one processor 310; and,

[0069] A memory 330 is communicatively connected to the at least one processor; wherein,

[0070] The memory 330 stores instructions 320 executable by the at least one processor 310. The instructions are executed by the at least one processor 310 to enable the hardware device to:

[0071] Determine the scope of a to-be-monitored area for monitoring tree characteristics, wherein the to-be-monitored area includes a plurality of dispersed trees to be monitored, and use a horizontally installed multi-line laser radar to collect point cloud data of an environment including the trees to be monitored within the geographic scope, wherein the point cloud data includes single-frame scan data of the trees to be monitored collected by the multi-line laser radar, and the single-frame scan data carries timestamp information when the single-frame scan data is obtained;

[0072] Extracting point cloud features from the point cloud data, wherein the extracted point cloud features include geometric line features or surface features;

[0073] Match the previous and next frames of the point cloud features obtained, calculate the position and posture changes of the subsequent frames relative to the first frame, and take the first frame data in the point cloud data as the origin to obtain the posture trajectory points of the motion path of the multi-line laser radar in the area to be monitored; at the same time, construct a global point cloud map of the environment within the area to be monitored based on the point cloud data;

[0074] When it is necessary to view the tree features of a specific area or a specific tree to be monitored in the global point cloud map, firstly, according to the location information of the monitored target, the time point information or spatial position information corresponding to the target is located on the trajectory in the constructed global point cloud map; the original point cloud data collected by the time point information or spatial position information is indexed by the posture information on the trajectory, so as to obtain a single frame scanning data matching the target position;

[0075] The tree features of the specific area or the specific trees to be monitored are obtained based on the single frame scanning data matching the target position.

[0076] An embodiment of the present invention further provides a computer-readable medium having computer-readable instructions stored thereon. The computer-readable instructions can be executed by a processor to implement a tree feature monitoring method.

[0077] Those skilled in the art can understand that the accompanying drawings are only schematic diagrams of an embodiment, and the modules or processes in the accompanying drawings are not necessarily required to implement the present invention.

[0078] Those skilled in the art can understand that the modules in the device in the embodiment can be distributed in the device in the embodiment according to the description of the embodiment, or can be changed accordingly and located in one or more devices different from the embodiment. The modules in the above embodiment can be combined into one module, or can be further divided into multiple sub-modules.

[0079] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A tree feature monitoring method, characterized in that: include: Determine the scope of a to-be-monitored area for monitoring tree characteristics, wherein the to-be-monitored area includes a plurality of dispersed trees to be monitored, and use a horizontally installed multi-line laser radar to collect point cloud data of an environment including the trees to be monitored within the geographic scope, wherein the point cloud data includes single-frame scan data of the trees to be monitored collected by the multi-line laser radar, and the single-frame scan data carries timestamp information when the single-frame scan data is obtained; Extracting point cloud features from the point cloud data, wherein the extracted point cloud features include geometric line features or surface features; Match the previous and next frames of the point cloud features obtained, calculate the position and posture changes of the subsequent frames relative to the first frame, and take the first frame data in the point cloud data as the origin to obtain the posture trajectory points of the motion path of the multi-line laser radar in the area to be monitored; at the same time, construct a global point cloud map of the environment within the area to be monitored based on the point cloud data; When it is necessary to view the tree features of a specific area or a specific tree to be monitored in the global point cloud map, firstly, according to the location information of the monitored target, the time point information or spatial position information corresponding to the target is located on the trajectory in the constructed global point cloud map; the original point cloud data collected by the time point information or spatial position information is indexed by the posture information on the trajectory, so as to obtain a single frame scanning data matching the target position; The tree features of the specific area or the specific trees to be monitored are obtained based on the single frame scanning data matching the target position.

2. The tree feature monitoring method according to claim 1, characterized in that: The calculating of the position and posture changes of the subsequent frames relative to the first frame specifically includes: Extract geometric features from the scanned point cloud data, including geometric line and surface features; Match the features of the current frame with the features of the first frame to find the corresponding relationship between the two frames of data; By processing the matched feature points, applying the RANSAC algorithm to remove outliers, and then using the matched feature points to calculate the rotation and translation parameters, the posture transformation of the subsequent frames relative to the first frame is obtained. The transformation is represented by a rigid body transformation matrix or a quaternion representing the posture and position offset.

3. The tree feature monitoring method according to claim 1, characterized in that: The locating the time point information or spatial position information corresponding to the target on the trajectory in the constructed global point cloud map specifically includes: Obtaining acquisition time information or position information of specific point cloud data corresponding to the specific area or the specific tree to be monitored from the global point cloud map; The time point information is obtained based on the time information, or the spatial position information is obtained based on the posture information.

4. A tree characteristic monitoring device, characterized in that: include: A point cloud data acquisition module is used to determine the scope of a to-be-monitored area for monitoring tree characteristics, wherein the to-be-monitored area includes a plurality of dispersed trees to be monitored, and to use a horizontally installed multi-line laser radar to collect point cloud data of an environment including the trees to be monitored within the geographic scope, wherein the point cloud data includes single-frame scan data of the trees to be monitored collected by the multi-line laser radar, and the single-frame scan data carries timestamp information when the single-frame scan data is obtained; A feature extraction module, used to extract point cloud features from the point cloud data, where the extracted point cloud features include geometric line features or surface features; A point cloud map construction module is used to match the previous and next frames of the obtained point cloud features, calculate the position and posture changes of the subsequent frames relative to the first frame, and use the first frame data in the point cloud data as the origin to obtain the position and posture trajectory points of the motion path of the multi-line laser radar in the range of the area to be monitored; at the same time, a global point cloud map of the environment within the range of the area to be monitored is constructed based on the point cloud data; The single-frame scanning data acquisition module is used for, when it is necessary to view the tree features of a specific area or a specific tree to be monitored in the global point cloud map, first locating the time point information or spatial position information corresponding to the target on the trajectory in the constructed global point cloud map according to the position information of the monitored target; using the posture information on the trajectory to index the original point cloud data collected by the time point information or spatial position information, so as to obtain the single-frame scanning data matching the target position; The tree feature acquisition module is used to obtain the tree features of the specific area or the specific trees to be monitored based on the single frame scanning data matching the target position.

5. A tree characteristic monitoring device comprising: at least one processor; as well as, a memory communicatively connected to the at least one processor; wherein, The processor stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to: Determine the scope of a to-be-monitored area for monitoring tree characteristics, wherein the to-be-monitored area includes a plurality of dispersed trees to be monitored, and use a horizontally installed multi-line laser radar to collect point cloud data of an environment including the trees to be monitored within the geographic scope, wherein the point cloud data includes single-frame scan data of the trees to be monitored collected by the multi-line laser radar, and the single-frame scan data carries timestamp information when the single-frame scan data is obtained; Extracting point cloud features from the point cloud data, wherein the extracted point cloud features include geometric line features or surface features; Match the previous and next frames of the point cloud features obtained, calculate the position and posture changes of the subsequent frames relative to the first frame, and take the first frame data in the point cloud data as the origin to obtain the posture trajectory points of the motion path of the multi-line laser radar in the area to be monitored; at the same time, construct a global point cloud map of the environment within the area to be monitored based on the point cloud data; When it is necessary to view the tree features of a specific area or a specific tree to be monitored in the global point cloud map, firstly, according to the location information of the monitored target, the time point information or spatial position information corresponding to the target is located on the trajectory in the constructed global point cloud map; the original point cloud data collected by the time point information or spatial position information is indexed by the posture information on the trajectory, so as to obtain a single frame scanning data matching the target position; The tree features of the specific area or the specific trees to be monitored are obtained based on the single frame scanning data matching the target position.

6. A computer-readable medium having computer-readable instructions stored thereon, wherein the computer-readable instructions can be executed by a processor to implement the tree characteristic monitoring method according to any one of claims 1 to 3.

Images