Automatic monitoring device and method for radial growth of tree trunk

Abstract

The application discloses a kind of automatic monitoring device and method of tree trunk radial growth, it is related to measurement technical field, device includes: electric rotating platform, laser ranging sensor, data processing equipment and support platform;Electric rotating platform drives laser ranging sensor to rotate a week, and laser ranging sensor determines the distance of all sheltering objects to sensor in real time;Data processing equipment receives the real-time ranging data sent by laser ranging sensor;After measuring a circle, the ranging data of current ranging data set is clustered, and a plurality of ranging data clusters are obtained;Right-angle coordinate modeling and circle fitting are carried out on each ranging data cluster, to determine the position and diameter of each tree trunk;Through the diameter of tree trunk obtained by continuous measurement at different times, the radial growth rate of tree trunk is calculated. By implementing the present application, the growth rate of multiple tree trunks can be measured automatically in a non-contact form, solving the problem of contact measurement affecting tree growth and only single measurement in the prior art.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of measurement, in particular to a tree trunk radial growth automatic monitoring device and method. BACKGROUND

[0002] Under the background of global change, studying the productivity of forest trees and their response to environmental factors is of great significance to reveal the adaptability of tree growth to climate change and more accurately quantify the dynamic changes of future forest ecosystems. Among them, tree trunk radial growth is one of the key indicators reflecting productivity.

[0003] Traditional tree trunk radial growth measurement methods include diameter tape measurement, tree ring analysis method, micro-core method, and new pressure type and pull rope type tree radial growth measuring instruments. The diameter tape method has problems such as poor observation continuity and risk of damage to trees. New instruments may also cause mechanical damage to trees, and most products can only measure single trees within a specific diameter range, which is low in efficiency. SUMMARY

[0004] The present application provides a tree trunk radial growth automatic monitoring device and method, which can realize non-contact automatic measurement of the growth rate of multiple tree trunks, solving the problem of contact measurement affecting tree growth and only single tree measurement in the prior art.

[0005] An embodiment of the present application provides a tree trunk radial growth automatic monitoring device, comprising: an electric rotating platform, a laser ranging sensor, a data processing device, and a support platform; wherein the laser ranging sensor is carried on the electric rotating platform, and the electric rotating platform is carried on the support platform;

[0006] The electric rotating platform is used to drive the laser ranging sensor to rotate according to a preset pulse frequency; wherein the rotation axis of the laser ranging sensor and the electric rotating platform is the same;

[0007] The laser ranging sensor is used to periodically emit laser and receive echo, and then determine the measurement distance by calculating the round-trip time of laser; according to the measurement distance and the corresponding time stamp, real-time ranging data is generated; and the real-time ranging data is sent to the data processing device;

[0008] The data processing device is configured to receive all real-time ranging data of the laser ranging sensor during one rotation, cluster all the real-time ranging data to obtain a plurality of ranging data clusters, model each ranging data cluster in rectangular coordinates to determine a corresponding Cartesian coordinate graph of each ranging data cluster, perform circle fitting on each Cartesian coordinate graph to determine a fitting circle center and a diameter corresponding to each ranging data cluster, and determine the position and diameter of the tree trunk; calculate the radial growth rate of each tree trunk diameter obtained through continuous measurement at different times; and wherein each ranging data cluster includes at least three real-time ranging data.

[0009] Further, the automatic tree trunk radial growth monitoring device further comprises a power supply device.

[0010] The power supply device is connected with the electric rotating platform, the laser ranging sensor and the data processing device respectively.

[0011] Further, the electric rotating platform comprises a stepper motor controller, a stepper motor driver, a stepper motor and a hollow rotating platform.

[0012] The stepper motor controller is configured to set a preset pulse frequency and send the preset pulse frequency to the stepper motor driver.

[0013] The stepper motor driver is configured to generate a pulse signal according to the preset pulse frequency and send the pulse signal to the stepper motor.

[0014] The stepper motor is configured to receive the pulse signal sent by the stepper motor driver and drive the hollow rotating platform to rotate according to the pulse signal.

[0015] The hollow rotating platform is configured to drive the laser ranging sensor to rotate; wherein the rotating shafts of the laser ranging sensor and the hollow rotating platform are the same.

[0016] Further, the data processing device clusters all the real-time ranging data to obtain a plurality of ranging data clusters, including:

[0017] According to the preset neighborhood radius and the preset minimum point number, the data point type of all the real-time ranging data is determined; the data point type includes a core point, a boundary point or a noise point.

[0018] The classification operation is repeatedly performed to obtain a plurality of ranging data clusters.

[0019] The classification operation includes:

[0020] select a core point from the current core point set to be classified as a target core point, and connect all ranging data of which the target core point has the maximum density as a target ranging data cluster; wherein, the initial core point set to be classified is a set composed of core points selected from all real-time ranging data;

[0021] mark all core points in the target ranging data cluster as visited core points;

[0022] delete the visited core points from the current core point set to be classified to obtain an updated core point set to be classified;

[0023] determine whether the updated core point set to be classified is empty, if yes, all target ranging data clusters are regarded as a plurality of ranging data clusters, and if no, the updated core point set to be classified is regarded as the current core point set to be classified in the next classification operation.

[0024] Further, the data processing device determines the data point type of all real-time ranging data according to a preset neighborhood radius and a preset minimum point number, including:

[0025] model each real-time ranging data in rectangular coordinates to determine the Cartesian coordinate position of each real-time ranging data, and then obtain all real-time ranging data points;

[0026] for each real-time ranging data point, query the number of other real-time ranging data points within the preset neighborhood radius of the real-time ranging data point to obtain the neighborhood point number of the real-time ranging data point;

[0027] compare the neighborhood point number with the preset minimum point number;

[0028] if the neighborhood point number is greater than or equal to the preset minimum point number, the real-time ranging data point is determined as a core point;

[0029] if the neighborhood point number is less than the preset minimum point number and the neighborhood point number is not equal to 0, the real-time ranging data point is determined as a boundary point;

[0030] if the neighborhood point number is equal to 0, the real-time ranging data point is determined as a noise point.

[0031] Further, the data processing device models each ranging data cluster in rectangular coordinates to determine the Cartesian coordinate diagram corresponding to each ranging data cluster, including:

[0032] for each ranging data cluster, determine the rotation angle of each real-time ranging data according to the time stamp of each real-time ranging data in the ranging data cluster and the preset rotation speed;

[0033] According to the rotation angle and the measured distance of each real-time ranging data, a right-angle coordinate model of each real-time ranging data in the ranging data cluster is established, and a Cartesian coordinate position of each real-time ranging data in the ranging data cluster is determined, and then a Cartesian coordinate graph corresponding to the ranging data cluster is obtained.

[0034] Further, the data processing device performs circle fitting on each Cartesian coordinate graph to determine a fitting circle center and a diameter corresponding to each ranging data cluster, including:

[0035] The circle fitting on each Cartesian coordinate graph is performed by using the least square method to determine the fitting circle center and the diameter corresponding to each ranging data cluster.

[0036] On the basis of the above-mentioned device embodiment, the present application correspondingly provides a method embodiment, which is suitable for the data processing device in the trunk radial growth automatic monitoring device provided by any embodiment of the present application.

[0037] The trunk radial growth automatic monitoring method comprises the following steps:

[0038] All real-time ranging data in the process of rotating one circle of the laser ranging sensor are received.

[0039] All real-time ranging data are clustered to obtain a plurality of ranging data clusters; wherein each ranging data cluster comprises at least three real-time ranging data.

[0040] A right-angle coordinate model of each ranging data cluster is established to determine a Cartesian coordinate graph corresponding to each ranging data cluster.

[0041] Circle fitting is performed on each Cartesian coordinate graph to determine a fitting circle center and a diameter corresponding to each ranging data cluster, which is used to determine the trunk position and the trunk diameter.

[0042] The diameter of each trunk obtained by continuous measurement at different times is used to calculate the radial growth rate of the trunk.

[0043] Compared with the prior art, the beneficial effects of the present application embodiment are as follows:

[0044] The trunk radial growth automatic monitoring device of the present application comprises an electric rotating platform, a laser ranging sensor, a data processing device and a support platform. The electric rotating platform drives the laser ranging sensor to rotate according to a preset pulse frequency. During the rotation, the laser ranging sensor automatically measures the distance data of all surrounding shielding objects to the sensor at a high frequency. Then, the data processing device receives all real-time ranging data collected by the laser ranging sensor during one rotation. The clustering algorithm is used to cluster all real-time ranging data, so as to separate the ranging data of multiple tree trunks, and each tree trunk generates a ranging data cluster. Considering the data effectiveness, each ranging data cluster includes at least three real-time ranging data. The right-angle coordinate modeling is performed on each ranging data cluster to determine the corresponding Cartesian coordinate graph of each ranging data cluster, and the circle fitting is performed on each Cartesian coordinate graph to determine the fitting circle center and diameter of each ranging data cluster, which are used to determine the position and diameter of each tree trunk, and realize the automatic measurement of the diameters of multiple tree trunks. Finally, the radial growth rate of each tree trunk is calculated based on the diameters of the tree trunks obtained by continuous measurement at different times.

[0045] In summary, the present application can realize the full-automatic measurement of the growth rates of multiple individual tree trunks in a forest community in a non-contact form by rotating the laser ranging sensor to sample multiple tree trunk scenes and combining clustering and clustering, and improve the measurement efficiency, thereby solving the problems in the prior art that the contact type measurement affects the growth of the tree trunk and can only achieve single tree measurement. BRIEF DESCRIPTION OF DRAWINGS

[0046] Figure 1 is a structural schematic diagram of a trunk radial growth automatic monitoring device provided by an embodiment of the present application;

[0047] Figure 2 is a structural schematic diagram of an electric rotating platform provided by an embodiment of the present application;

[0048] Figure 3 is a density relationship schematic diagram of a DBSCAN algorithm provided by an embodiment of the present application;

[0049] Figure 4 is a fitting effect diagram of a ranging data cluster provided by an embodiment of the present application;

[0050] Figure 5 is a flowchart of a trunk radial growth automatic monitoring method provided by an embodiment of the present application;

[0051] In the drawings, the reference signs of the drawings of the specification are as follows:

[0052] 1, electric rotating platform; 101, stepper motor controller; 102, stepper motor driver; 103, stepper motor; 104, hollow rotating platform; 2, laser ranging sensor; 3, data processing device; 4, support platform; 5, power supply device. DETAILED DESCRIPTION

[0053] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0054] In the description of the present application, it should be understood that the terms "first", "second" are only for the purpose of description, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features.

[0055] As shown in the drawings, Figure 1 In order to solve the problem of contact measurement affecting the growth of the trunk and only single trunk measurement in the prior art, an embodiment of the present application provides a trunk radial growth automatic monitoring device, which comprises: an electric rotating platform 1, a laser ranging sensor 2, a data processing device 3 and a support platform 4; wherein the laser ranging sensor 2 is carried on the electric rotating platform 1, and the electric rotating platform 1 is carried on the support platform 4.

[0056] The electric rotating platform 1 is used to drive the laser ranging sensor 2 to rotate according to a preset pulse frequency; wherein the rotation axis of the laser ranging sensor 2 and the electric rotating platform 1 is the same.

[0057] Specifically, the electric rotating platform 1 drives the laser ranging sensor 2 to rotate synchronously according to the preset pulse frequency, so that the laser beam of the range finder forms a 360° scan in the horizontal plane. Wherein the pulse frequency determines the speed and angle accuracy of rotation.

[0058] It should be noted that the rotation axis of the laser ranging sensor 2 and the electric rotating platform 1 is the same, which is used to ensure that the laser ranging sensor always performs circumferential scanning with the same axis as the center during rotation, avoiding ranging error caused by axis deviation.

[0059] As shown in the drawings, Figure 2 The structure of the electric rotating platform is shown in the drawings, and in a preferred embodiment, the electric rotating platform 1 comprises: a stepper motor controller 101, a stepper motor driver 102, a stepper motor 103 and a hollow rotating platform 104.

[0060] The step motor controller 101 is configured to set a preset pulse frequency and send the preset pulse frequency to the step motor driver 102.

[0061] The step motor driver 102 is configured to generate a pulse signal according to the preset pulse frequency and send the pulse signal to the step motor 103.

[0062] The step motor 103 is configured to receive the pulse signal sent by the step motor controller 101 and drive the hollow rotating platform 104 to rotate according to the pulse signal.

[0063] The hollow rotating platform 104 is configured to drive the laser ranging sensor 2 to rotate, wherein the laser ranging sensor 2 and the hollow rotating platform 104 have the same rotating axis.

[0064] In an embodiment of the present application, the electric rotating platform 1 is composed of the step motor controller 101, the step motor driver 102, the step motor 103 and the hollow rotating platform 104, which cooperatively provide controllable rotating power for the laser ranging sensor 2 to realize distance monitoring of the tree in the radial direction.

[0065] Specifically, the step motor controller 101 sets a preset pulse frequency and sends the preset pulse frequency to the step motor driver 102; the step motor driver 102 generates a pulse signal according to the preset pulse frequency and sends the pulse signal to the step motor 103, the pulse signal is a direct instruction for driving the step motor to rotate, and each pulse corresponds to a fixed rotation angle of the step motor 103; the step motor 103 accurately controls the rotation angle and speed after receiving the pulse signal sent by the step motor driver 102, and then drives the hollow rotating platform 104 to rotate, thereby converting the electrical signal into a mechanical rotating action; the hollow rotating platform 104 directly drives the laser ranging sensor 2 to rotate, and requires the laser ranging sensor 2 to coincide with the rotating axis of the platform, so as to ensure that the laser ranging sensor 2 moves in a circular motion with the same center when rotating, avoid distance measurement error caused by axis deviation, and enable the laser beam to stably scan the tree in the radial direction.

[0066] It should be noted that in the electric rotating platform 1, the step motor controller 101 is connected with the step motor driver 102, the step motor driver 102 is connected with the step motor 103, and the step motor 103 is fixed below the hollow rotating platform 104.

[0067] The laser ranging sensor 2 is configured to periodically emit laser and receive echo, and then determine the measurement distance by calculating the round-trip time of the laser; generate real-time ranging data according to the measurement distance and the corresponding time stamp; and send the real-time ranging data to the data processing device 3.

[0068] Specifically, the laser ranging sensor 2 periodically and continuously emits laser, and receives the echo reflected by the laser after colliding with the trees and other objects. By using the physical principle "distance = (light speed x round-trip time) ÷ 2", the distance between the laser ranging sensor 2 and a point on the surface of the tree is calculated by measuring the round-trip time of the laser from emission to reception. Then, the laser ranging sensor 2 sends the measured distance value and the time stamp to the data processing device 3.

[0069] The data processing device 3 is configured to receive all real-time ranging data of the laser ranging sensor 2 during one rotation.

[0070] Specifically, in the automatic monitoring device for the radial growth of the tree trunk, the data processing device 3 first receives the real-time ranging data sent by the laser ranging sensor 2 through the communication interface (such as USB or wireless transmission). When the laser ranging sensor 2 completes the 360° circular scanning with the electric rotating platform 1, the data processing device 3 receives and processes all the real-time ranging data collected by the laser ranging sensor 2, and associates each data point with the corresponding time stamp, that is, each real-time ranging data includes the corresponding distance value and the time stamp.

[0071] In an embodiment, the data processing device 3 is a computer, the laser ranging sensor 2 completes the 360° circular scanning with the electric rotating platform 1 at a position of 1.3 m from the ground at a speed of 1 hour / revolution, and the laser ranging sensor 2 sends the collected real-time ranging data to the data processing device 3 through the communication interface during the scanning.

[0072] It should be noted that the data processing device 3 can be an embedded device or a special data processing terminal in addition to the computer. In order to effectively avoid the ranging interference of low shrubs and canopy leaves and improve the data effectiveness, the scanning height of the laser ranging sensor 2 is usually limited within a preset range, and in this embodiment, the preset range is 1.3 m to 1.5 m.

[0073] The data processing device 3 is further configured to cluster all the real-time ranging data to obtain a plurality of ranging data clusters; and each ranging data cluster includes at least three real-time ranging data.

[0074] Specifically, since the ranging data received by the data processing device 3 is scattered, the spatial clustering algorithm is used to cluster all the real-time ranging data. When there are multiple trees in the scanning range, the ranging points of different tree trunks form independent clusters due to the spatial position difference, and each cluster corresponds to a cross-sectional point set of a tree trunk, which provides a basis for subsequent diameter calculation and shape analysis of a single tree trunk.

[0075] In a preferred embodiment, the data processing device 3 clusters all the real-time ranging data to obtain a plurality of ranging data clusters, including:

[0076] The data point types for all real-time ranging data are determined based on the preset neighborhood radius and the preset minimum number of points; the data point types include: core points, boundary points, or noise points.

[0077] Repeat the classification operation to obtain several range measurement data clusters;

[0078] Classification operations include:

[0079] Select a core point from the current set of core points to be classified as the target core point, and take all the ranging data connected by the target core point with the highest density as a target ranging data cluster; wherein, the initial set of core points to be classified is a set of core points selected from all real-time ranging data.

[0080] Mark all core points in the target ranging data cluster as visited core points;

[0081] Remove visited core points from the current set of core points to be classified, and obtain an updated set of core points to be classified.

[0082] Determine whether the updated set of core points to be classified is empty. If it is, treat all target ranging data clusters as several ranging data clusters. If not, use the updated set of core points to be classified as the current set of core points to be classified when performing the classification operation next time.

[0083] In one embodiment of the present invention, the DBSCAN (Density-Based Spatial Clustering of Applications with Noise) algorithm is used to cluster all ranging data in the current ranging dataset.

[0084] It should be noted that the specific principle of the DBSCAN algorithm is as follows:

[0085] Assuming a known sample set ,in, These are individual sample points. This is the total number of sample points, and the neighborhood radius is . The minimum number of points is (The distance to a certain sample is) (The threshold for the number of samples in the neighborhood).

[0086] like Figure 3 The diagram shows the density relationship of the DBSCAN algorithm. For any sample... If with Centered on, with radius as Number of sample points in the neighborhood Not less than the minimum number of points , then it is determined that is a core object. For another sample , and , if is located in the neighborhood of , and is a core object, then it is said that is density direct.

[0087] For and , if there exists a sample sequence , satisfying , , and is density direct, then it is said that is density reachable. It should be noted that the sample sequence is a segment of sample points selected from the data set . For and

[0088] , if there exists a core object sample , , and , such that and are both density reachable, then it is said that and are density connected. In the present application, DBSCAN algorithm is used to cluster all ranging data in the current ranging data set. First, the density characteristics of data points are defined, and the neighborhood radius is set to and the minimum number of points is set to

[0089] , wherein the neighborhood radius is the area within a radius of centered on a single data point, which is called the neighborhood of the point, and the minimum number of points is the number of points required in the neighborhood to determine whether the density of the point is sufficient to form a cluster. In a preferred embodiment, the data processing device 3 determines the data point types of all real-time ranging data according to the preset neighborhood radius and the preset minimum number of points, including:

[0090] Each real-time ranging data is modeled in rectangular coordinates to determine the Cartesian coordinate position of each real-time ranging data, and then all real-time ranging data points are obtained.

[0091]

[0092] ​​​​​For each real-time ranging data point, query the number of other real-time ranging data points within the preset neighborhood radius to obtain the number of neighborhood points of the real-time ranging data point.

[0093] Compare the number of neighboring points with the preset minimum number of points;

[0094] If the number of neighboring points is greater than or equal to the preset minimum number of points, then the real-time ranging data point is determined as the core point;

[0095] If the number of neighboring points is less than the preset minimum number of points, and the number of neighboring points is not equal to 0, then the real-time ranging data point is determined as a boundary point;

[0096] If the number of neighboring points is 0, then the real-time ranging data point is determined to be a noise point.

[0097] In one embodiment of the present invention, the angle value corresponding to the real-time ranging data is calculated based on the timestamp of the real-time ranging data and the preset pulse frequency of the driving electric rotating platform. Based on the angle and distance values ​​of each real-time ranging data point, polar coordinates are constructed and converted to Cartesian coordinates to determine the Cartesian coordinate position of each real-time ranging data point. After completing this coordinate transformation for all real-time ranging data points, all real-time ranging data points represented by coordinates are obtained. For each real-time ranging data point... Calculate real-time ranging data points Point set in the neighborhood The mathematical expression is:

[0098] ;

[0099] in, Indicates real-time ranging data points The set of other data points contained within the region centered at r and with radius r. Represents the set of all real-time ranging data points. This represents the set D excluding real-time ranging data points. Any real-time ranging data point other than [other data points] Represents real-time ranging data points and real-time ranging data points The Euclidean distance between them Represents the neighborhood radius.

[0100] Determine real-time ranging data points Point set in the neighborhood Is it greater than or equal to the minimum number of points? ,like This indicates the real-time ranging data points. The surrounding data points are dense enough to determine the real-time ranging data points. For the core point, if , the data point is determined as a boundary point, in the case that the data point does not belong to the neighborhood of any core point, the data point is determined as a noise point.

[0101] The classification operation is repeatedly executed to generate a plurality of ranging data clusters. Specifically, before the classification operation is repeatedly executed, a set of core points to be classified is initialized, and the core points selected from all real-time ranging data constitute the current set of core points to be classified. Each time the classification operation is executed, a core point is selected from the current set of core points to be classified as a target core point, and all ranging data connected to the target core point through maximum density are found, which form a target ranging data cluster. The core points in the cluster are marked as visited, and the visited core points are deleted from the current set of core points to be classified to obtain an updated set. It is checked whether the updated set is empty. If it is empty, it means that all core points have completed clustering, and all target ranging data clusters currently generated are taken as output to obtain a plurality of ranging data clusters. If it is not empty, the updated set is taken as the set of core points to be classified in the next round of classification operation, and the classification operation is repeatedly executed until the set of core points to be classified is empty.

[0102] It should be noted that in the clustering logic of the DBSCAN algorithm, when a sample point can be directly connected by two core objects, and the two core objects belong to different clusters, the cluster attribution of the sample point has order dependence, that is, it belongs to the cluster to which it is divided first.

[0103] Through the classification operation of such cyclic iteration, the density clustering logic of DBSCAN is used to automatically identify the continuous dense region of trunk ranging data, so that each cluster corresponds to the cross-section point set of a single tree trunk, while filtering noise points, and finally realizing accurate density clustering and division of ranging data. In addition, in order to ensure the effectiveness of the clustering result, each ranging data cluster includes at least 3 real-time ranging data, so as to avoid misjudgment caused by a few data points forming a cluster, and ensure that each cluster can truly reflect the outline of a tree trunk.

[0104] The data processing device 3 is also configured to perform right-angle coordinate modeling on each ranging data cluster to determine a Cartesian coordinate graph corresponding to each ranging data cluster.

[0105] In a preferred embodiment, the data processing device 3 performs right-angle coordinate modeling on each ranging data cluster to determine a Cartesian coordinate graph corresponding to each ranging data cluster, including:

[0106] ​For each ranging data cluster, the rotation angle of each real-time ranging data is determined according to the time stamp of each real-time ranging data in the ranging data cluster and the preset rotation speed;

[0107] According to the rotation angle of each real-time ranging data and the measured distance, each real-time ranging data in the ranging data cluster is modeled in rectangular coordinates to determine the Cartesian coordinate position of each real-time ranging data in the ranging data cluster, and then the Cartesian coordinate graph corresponding to the ranging data cluster is obtained.

[0108] In an embodiment of the present application, a rectangular coordinate system is established with the space position of the laser ranging sensor 2 as the origin. Since the laser ranging sensor 2 rotates at a preset rotation speed with the motorized rotating platform 1, the rotation angle of each real-time ranging data is determined according to the time stamp of each real-time ranging data in the ranging data cluster and the preset rotation speed:

[0109] ;

[0110] wherein, represents the rotation angle of the real-time ranging data, represents the initial angle of the laser ranging sensor 2, represents the preset rotation speed, represents the measurement time of the real-time ranging data, i.e. the time stamp of the real-time ranging data, represents the measurement time when the laser ranging sensor 2 is at the initial angle.

[0111] Then, by using the distance value (corresponding to the radial length from the origin to the trunk) and the angle value (corresponding to the included angle between the radial and the coordinate axis) of the real-time ranging data in each ranging data cluster, the mathematical modeling is performed on each ranging data cluster by converting the polar coordinates to rectangular coordinates, so as to convert the polar coordinates into a point set in the rectangular coordinate system, and finally generate the Cartesian coordinate graph corresponding to each ranging data cluster.

[0112] The data processing device 3 is also configured to perform circle fitting on each Cartesian coordinate graph to determine the fitting circle center and diameter corresponding to each ranging data cluster, so as to determine the trunk position and trunk diameter.

[0113] Specifically, the circle fitting is performed on the point set in each Cartesian coordinate graph. In the case of uniform trunk growth, the circle is the most ideal and simple mathematical model. By mathematical method, an optimal circle is found to fit all coordinate points in the cluster as much as possible. The center of the circle is the position of the corresponding trunk in the coordinate system, and the radius of the circle multiplied by 2 is the trunk diameter.

[0114] In a preferred embodiment, the data processing device 3 performs circle fitting on each Cartesian coordinate graph to determine the fitting circle center and diameter corresponding to each ranging data cluster, comprising:

[0115] By performing circle fitting on each Cartesian coordinate graph using the least squares method, the center and diameter of the fitted circle corresponding to each range measurement data cluster are determined.

[0116] In one embodiment of the present invention, in order to accurately describe the shape of the tree trunk, a least squares method is used to fit a circle to each Cartesian coordinate graph. The goal of the least squares method is to minimize the sum of the squares of the distances from all points to the fitted circle, thereby obtaining the most suitable circle parameters. Let the equation of the fitted circle be:

[0117] ;

[0118] in, Represents the coordinates of the center of the circle. Indicates the radius of the circle.

[0119] Construct an objective function to quantify the fitting error. The objective function is defined as follows:

[0120] ;

[0121] in, This represents the Cartesian coordinates of the c-th point in the ranging data cluster. This represents the total number of data points in the ranging data cluster. This reflects the sum of squared distances from all points in the ranging data cluster to the fitted circle, representing the error between the fitted circle and the measurement points. The calculation method utilizes Python's NumPy library for matrix operations, iteratively calculating the error between the fitted circle and the measurement points. Minimum.

[0122] Finally, when the error is minimized Then, the center of the fitted circle is taken as the position of the tree trunk, and the diameter of the tree trunk is obtained by multiplying the radius of the circle by 2.

[0123] In this embodiment, nine ranging data clusters were obtained after clustering, and circle fitting was performed to obtain the following... Figure 4 The nine subplots showing the fitting results are shown below. Figure 4 (a) to Figure 4 (i) Each sub-graph corresponds to the fitting result of a range measurement data cluster, clearly showing the distribution relationship between the fitting circle (representing the trunk section) and the original range measurement points.

[0124] The data processing device 3 is also used to calculate the radial growth rate of each tree trunk by measuring the diameter of each trunk continuously over different times.

[0125] Specifically, considering the growth rate difference of different tree species, like slow-growing tree species (such as pine, ginkgo, etc.), the growth cycle is long, the diameter change range is small and slow, and a longer time is needed to accumulate measurement data to accurately capture the growth trend and effectively determine whether the growth is abnormal; while fast-growing tree species (such as poplar, eucalyptus, etc.) grow fast, and the diameter change is relatively significant, and a short time of measurement data can reflect the growth state. Therefore, the trunk radial growth rate can be analyzed only after the trunk diameter of the preset rotation period is collected, and the preset rotation period needs to be determined according to the tree species to be measured in the monitoring area and combined with the pulse frequency of the electric rotating platform 1.

[0126] It should be noted that if slow-growing tree species and fast-growing tree species exist in the monitoring area at the same time, the number of preset measurement periods needs to be selected according to the growth characteristics of slow-growing tree species to ensure that the subtle growth changes of slow-growing tree species can be effectively captured, and to avoid missing abnormal judgment due to insufficient data accumulation.

[0127] Before calculating the trunk radial growth rate, the trunk diameter of each trunk in each measurement period is tracked according to the trunk position of each trunk in each measurement period. Specifically, based on the spatial position coordinates of the trunk, the newly measured trunk point set is matched with the historical data by using minimum distance matching or spatial topological relationship, the data of the same trunk at different time periods is confirmed, and the continuous tracking of the trunk diameter change over time is realized.

[0128] For each trunk, the trunk diameter change amount in two specific measurement periods in each measurement period is calculated:

[0129] ;

[0130] wherein, represents the diameter change amount of the i-th trunk, represents the trunk diameter of the i-th trunk in the s-th measurement period, represents the trunk diameter of the i-th trunk in the w-th measurement period, s>w.

[0131] After calculating the trunk diameter change amount of the two specific measurement periods, the growth rate of the trunk is calculated combined with the time interval of the two specific measurement periods:

[0132] ;

[0133] wherein, represents the growth rate of the i-th trunk, represents the measurement time of the i-th trunk in the s-th measurement period, represents the measurement time of the i-th trunk in the w-th measurement period.

[0134] Next, based on the growth rate of each tree trunk at each measurement time, the radial growth trend of the trunk is determined, forming a growth rate change graph. Specifically, if the growth rate value gradually increases over time, such as from 1 mm / d to 3 mm / d, it indicates that the radial growth of the trunk is accelerating, possibly during the peak growth season when the environment is conducive to cell division and thickening. If the growth rate fluctuates within a certain range, such as repeatedly between 2-3 mm / d without a continuous increase or decrease, it represents that the radial growth of the trunk is relatively stable, and environmental factors have not significantly changed the growth, or the tree is in a stable growth period. If the growth rate value continuously decreases, such as from 4 mm / d to 1 mm / d, it means that the radial growth of the trunk is slowing down, possibly due to environmental stress, such as drought or low temperature, which inhibits radial thickening.

[0135] In a preferred embodiment, the automatic monitoring device for radial growth of tree trunks further includes: a power supply device 5;

[0136] The power supply equipment 5 is connected to the electric rotating platform 1, the laser rangefinder 2, and the data processing equipment 3, respectively.

[0137] In one embodiment of the present invention, in order to ensure that the automatic monitoring device for radial growth of tree trunk can operate continuously, the automatic monitoring device for radial growth of tree trunk also includes a power supply device 5, which is connected to the electric rotating platform 1, the laser rangefinder 2 and the data processing device 3 respectively, so as to provide power to the connecting components.

[0138] Preferably, the power supply device 5 can be a solar cell, a storage battery, or a combination of both.

[0139] The automatic monitoring device for radial growth of tree trunks of the present invention uses a laser ranging sensor 2 to rotate and sample multiple tree trunks. The data processing device 3 uses density clustering to divide the ranging data into independent ranging data clusters, solving the problem of multi-target aliasing. The least squares method is used to fit a circle equation to each ranging data cluster. Compared with the contact measurement of traditional methods such as tree ring analysis and micro-tree core method, it will not cause damage to the trees. At the same time, compared with the limitations of existing technologies where contact measurement affects tree trunk growth and can only measure a single tree, the present invention adopts a non-contact measurement method, which will not cause physical interference to tree trunk growth, ensuring that the natural growth state of the trees is not affected. Furthermore, it can measure the diameter of multiple tree trunks simultaneously in a single measurement, effectively improving the measurement efficiency.

[0140] like Figure 5 As shown, based on the above-described apparatus embodiments, corresponding method embodiments are provided;

[0141] One embodiment of the present invention provides an automatic monitoring method for radial growth of tree trunks, applicable to the data processing equipment in the automatic monitoring device for radial growth of tree trunks provided in any embodiment of the present invention;

[0142] The method for automatically monitoring the radial growth of a tree trunk comprises at least the following steps:

[0143] Step S1: receiving all real-time ranging data of the laser ranging sensor during one rotation;

[0144] Step S2: clustering all real-time ranging data to obtain a plurality of ranging data clusters; wherein each ranging data cluster comprises at least three real-time ranging data;

[0145] Step S3: modeling each ranging data cluster in rectangular coordinates to determine a Cartesian coordinate graph corresponding to each ranging data cluster;

[0146] Step S4: fitting a circle to each Cartesian coordinate graph to determine the center and diameter of the fitted circle corresponding to each ranging data cluster, which is used to determine the position and diameter of the tree trunk;

[0147] Step S5: calculating the radial growth rate of each tree trunk by using the diameters of the tree trunk obtained by continuous measurement at different times.

[0148] In the description of the present specification, the description of the terms "one embodiment", "some embodiments", "an example", "a specific example" or "some examples" means that the specific features, structures, materials or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present application. Moreover, the specific features, structures, materials or characteristics described can be combined in any appropriate manner in any one or more embodiments or examples. In addition, different embodiments or examples described in the present specification and the features of different embodiments or examples can be combined and combined by those skilled in the art without contradiction.

[0149] The above is the preferred embodiment of the present application. It should be noted that for those skilled in the art, without departing from the principles of the present application, a number of improvements and refinements can be made, which are also considered within the scope of protection of the present application.

Claims

1. A device for automatic monitoring of the radial growth of a tree trunk, characterized in that it comprises: The application relates to a laser ranging sensor, a data processing device and a support platform; wherein the laser ranging sensor is mounted on the electric rotating platform, and the electric rotating platform is mounted on the support platform. The electric rotating platform is used for driving the laser ranging sensor to rotate according to a preset pulse frequency; wherein the rotating shafts of the laser ranging sensor and the electric rotating platform are the same. The laser ranging sensor is used for periodically emitting laser and receiving echo, and then determining a measuring distance by calculating the laser round-trip time; real-time ranging data is generated according to the measuring distance and a corresponding time stamp; and the real-time ranging data is sent to the data processing device. The data processing device is used for receiving all real-time ranging data of the laser ranging sensor during one rotation; clustering all real-time ranging data to obtain a plurality of ranging data clusters; modeling each ranging data cluster in rectangular coordinates to determine a corresponding Cartesian coordinate graph of each ranging data cluster; fitting each Cartesian coordinate graph to determine a fitting circle center and a diameter of each ranging data cluster, which are used for determining the trunk position and the trunk diameter; and calculating the trunk radial growth rate of each trunk by the trunk diameters obtained through continuous measurement at different times; wherein each ranging data cluster comprises at least three real-time ranging data. The clustering of all real-time ranging data to obtain a plurality of ranging data clusters comprises: determining the data point types of all real-time ranging data according to a preset neighborhood radius and a preset minimum point number; the data point types include core points, boundary points or noise points; repeatedly performing a classification operation to obtain a plurality of ranging data clusters; The classification operation comprises: selecting a core point from a current core point set to be classified as a target core point, and connecting all ranging data of the target core point with the maximum density to a target ranging data cluster; wherein the initial core point set to be classified is a set composed of core points selected from all real-time ranging data; marking all core points in the target ranging data cluster as visited core points; deleting the visited core points from the current core point set to be classified to obtain an updated core point set to be classified; determining whether the updated core point set to be classified is empty; if yes, all target ranging data clusters are regarded as the plurality of ranging data clusters; if not, the updated core point set to be classified is regarded as the current core point set to be classified in the next classification operation; determining the data point types of all real-time ranging data according to a preset neighborhood radius and a preset minimum point number, comprises: modeling each real-time ranging data in rectangular coordinates to determine the Cartesian coordinate position of each real-time ranging data, and then obtaining all real-time ranging data points; for each real-time ranging data point, the number of other real-time ranging data points within the preset neighborhood radius of the real-time ranging data point is inquired to obtain the neighborhood point number of the real-time ranging data point; comparing the neighborhood point number with the preset minimum point number; if the neighborhood point number is greater than or equal to the preset minimum point number, the real-time ranging data point is determined as a core point. ​ If the number of neighborhood points is less than the preset minimum number of points and the number of neighborhood points is not equal to 0, the real-time ranging data point is determined as a boundary point; If the number of neighborhood points is equal to 0, the real-time ranging data point is determined as a noise point.

2. The automatic tree stem radial growth monitoring device according to claim 1, characterized in that, Further comprising: A power supply device; The power supply device is connected with the electric rotating platform, the laser ranging sensor and the data processing device respectively.

3. The automatic tree stem radial growth monitoring device according to claim 1, characterized in that, The electric rotating platform comprises a step motor controller, a step motor driver, a step motor and a hollow rotating platform; The step motor controller is configured to set a preset pulse frequency and send the preset pulse frequency to the step motor driver; The step motor driver is configured to generate a pulse signal according to the preset pulse frequency and send the pulse signal to the step motor; The step motor is configured to receive the pulse signal sent by the step motor driver and drive the hollow rotating platform to rotate according to the pulse signal; The hollow rotating platform is configured to drive the laser ranging sensor to rotate; wherein the rotating shafts of the laser ranging sensor and the hollow rotating platform are the same.

4. The automatic tree stem radial growth monitoring device according to claim 1, characterized in that, The data processing device models each ranging data cluster in rectangular coordinates, determines a Cartesian coordinate graph corresponding to each ranging data cluster, comprising: For each ranging data cluster, the rotation angle of each real-time ranging data is determined according to the time stamp of each real-time ranging data in the ranging data cluster and the preset rotation speed; The real-time ranging data in the ranging data cluster is modeled in rectangular coordinates according to the rotation angle and the measured distance of each real-time ranging data, the Cartesian coordinate position of each real-time ranging data in the ranging data cluster is determined, and then the Cartesian coordinate graph corresponding to the ranging data cluster is obtained.

5. The automatic tree stem radial growth monitoring device according to claim 1, characterized in that, The data processing device performs circle fitting on each Cartesian coordinate graph to determine the center and diameter of the fitting circle corresponding to each ranging data cluster, comprising: The circle fitting on each Cartesian coordinate graph is performed by the least square method to determine the center and diameter of the fitting circle corresponding to each ranging data cluster.

6. A method of automatic monitoring of the radial growth of a tree trunk, characterized in that, The data processing device is suitable for the automatic tree trunk radial growth monitoring device of any one of claims 1-5; The automatic tree trunk radial growth monitoring method comprises: Receiving all real-time ranging data of the laser ranging sensor during one rotation; Clustering all real-time ranging data to obtain a plurality of ranging data clusters; wherein each ranging data cluster comprises at least 3 real-time ranging data; Modeling each ranging data cluster in rectangular coordinates to determine a Cartesian coordinate graph corresponding to each ranging data cluster; Performing circle fitting on each Cartesian coordinate graph to determine the center and diameter of the fitting circle corresponding to each ranging data cluster, which is used to determine the position and diameter of the tree trunk; Calculating the radial growth rate of the diameter of each tree trunk by continuously measuring the diameter of each tree trunk at different times; The clustering of all real-time ranging data to obtain a plurality of ranging data clusters comprises: Determining the data point type of all real-time ranging data according to the preset neighborhood radius and the preset minimum number of points; the data point type comprises a core point, a boundary point or a noise point; Repeating the classification operation to obtain a plurality of ranging data clusters; The classification operation comprises: selecting a core point from the current core point set to be classified as a target core point, and connecting all ranging data of which the target core point has the maximum density as a target ranging data cluster; wherein the initial core point set to be classified is a set of core points selected from all real-time ranging data; marking all core points in the target ranging data cluster as visited core points; deleting the visited core points from the current core point set to be classified to obtain an updated core point set to be classified; determining whether the updated core point set to be classified is empty, if yes, regarding all target ranging data clusters as the ranging data clusters, and if no, regarding the updated core point set to be classified as the current core point set to be classified for the next classification operation; determining the data point type of all real-time ranging data according to the preset neighborhood radius and the preset minimum point number, comprising: modeling each real-time ranging data in rectangular coordinates to determine the Cartesian coordinate position of each real-time ranging data, and then obtaining all real-time ranging data points; for each real-time ranging data point, querying the number of other real-time ranging data points within the preset neighborhood radius of the real-time ranging data point to obtain the neighborhood point number of the real-time ranging data point; comparing the neighborhood point number with the preset minimum point number; if the neighborhood point number is greater than or equal to the preset minimum point number, the real-time ranging data point is determined as a core point; if the neighborhood point number is less than the preset minimum point number and the neighborhood point number is not equal to 0, the real-time ranging data point is determined as a boundary point; if the neighborhood point number is equal to 0, the real-time ranging data point is determined as a noise point.

Images