Forest landscape area accurate measurement system based on laser radar

Abstract

The invention relates to the technical field of forest resource investigation, and discloses a forest landscape area accurate measurement system based on a laser radar, which comprises an equipment end, a data end and an interface end, the data end comprises a data acquisition module, a data preprocessing module, a data fusion module and an area measurement module; by arranging the data fusion module, feature points extracted by the radar data preprocessing unit are matched with feature points extracted by the remote sensing data preprocessing unit, a set of optimal matching feature point pairs is obtained, and then an improved optimization algorithm is used for optimizing a weighted iteration nearest point algorithm, so that the optimal matching feature point pairs are obtained. A weight concept is introduced into a weighted iteration nearest point algorithm to improve registration precision and robustness, an improved optimization algorithm is adopted to obtain an optimal weight value, meanwhile, a neural network is introduced into the optimization algorithm, the nonlinear and multi-modal data modeling capacity of the neural network is brought into full play, and a complex data relation is effectively captured.

Description

Technical Field

[0001] The present invention relates to the technical field of forest resource surveys, and more particularly to a precise measurement system for forest landscape area based on lidar. Background Art

[0002] A forest landscape refers to a complete forest and non-forest ecosystem with an area exceeding 500 square kilometers; traditional methods for measuring forest landscape area usually rely on manual measurement and remote sensing technology. Manual measurement is not only time-consuming and laborious but also has low measurement accuracy; remote sensing images can cover large areas and are suitable for macroscopic measurement of forest landscape unit area. However, remote sensing technologies such as optical remote sensing and microwave remote sensing are easily restricted by weather conditions, such as cloud cover, terrain undulation, vegetation cover, etc., resulting in the resolution and accuracy of remote sensing images being affected by various factors, and thus the error of the measurement results.

[0003] The published document with the publication number CN118379340A discloses a method for determining the area of a regular triangular network forest plot. This method divides the forest plot into multiple triangular regions; calculates the floor areas corresponding to the multiple triangular regions respectively; and calculates the sum of the floor areas corresponding to the multiple triangular regions as the area of the forest plot, which can improve the safety and stability of the power system and the accuracy and efficiency of forest plot area measurement.

[0004] However, for various measurement methods, the acquisition of measurement data is the basis and key to improving the accuracy of area measurement. If errors occur in the collected measurement data, even if the most precise area measurement method is used, the final result will still be deviated; with the development of lidar technology, the three-dimensional spatial data obtained by lidar can accurately measure the area of forest landscapes. Lidar technology can provide extremely high angular, distance, and velocity resolutions and can provide high-precision positioning. Therefore, when collecting measurement data, it can effectively improve the accuracy of measurement data and thus ensure the accuracy of area measurement.

[0005] In view of this, the present invention proposes a precise measurement system for forest landscape area based on lidar, which uses lidar technology and combines remote sensing technology at the same time to improve measurement accuracy while ensuring measurement efficiency. Summary of the Invention

[0006] In order to overcome the above-mentioned defects of the prior art, the present invention provides a precise measurement system for forest landscape area based on lidar to solve the problems existing in the above background art.

[0007] The present invention provides the following technical solution: A precise measurement system for forest landscape area based on lidar, including a device end, a data end, and an interface end; The device side includes a lidar device and a remote sensing device, and the device side collects data respectively according to the data acquisition instruction; The data side is used to receive the data collected by the device side and perform analysis, including a data acquisition module, a data preprocessing module, a data fusion module, and an area measurement module; The interface side is used for man-machine interaction display of data; The data acquisition module is used to send a data acquisition instruction to the device side and acquire the collected lidar data and remote sensing data; The data preprocessing module is used to perform preprocessing operations on the data of the data acquisition module; the data preprocessing module includes a lidar data preprocessing unit and a remote sensing data preprocessing unit; The data fusion module is used to match the feature points extracted by the lidar data preprocessing unit and the feature points extracted by the remote sensing data preprocessing unit, obtain a set of optimal matching feature point pairs, then optimize the weighted iterative closest point algorithm using an improved optimization algorithm, and then perform data registration; The area measurement module extracts the boundary of the target object based on the edge detection algorithm, combines the data registered by the data fusion module, clusters the feature points using the clustering algorithm, obtains the categories corresponding to the feature points and then performs area block division, and measures and calculates the area based on the divided area blocks.

[0008] Preferably, the lidar device is used to obtain lidar data of the target object through lidar technology and transmit it to the data side, and the remote sensing device is used to obtain remote sensing data of the target object through remote sensing technology and transmit it to the data side; the target object is a forest landscape that needs to be measured for area.

[0009] Preferably, the data acquisition module includes a lidar data acquisition unit and a remote sensing data acquisition unit. The lidar data acquisition unit is used to receive the lidar data transmitted by the lidar device and then transmit it to the lidar data preprocessing unit; the remote sensing data acquisition unit is used to receive the remote sensing data transmitted by the remote sensing device and then transmit it to the remote sensing data preprocessing unit; The lidar data preprocessing unit is used to perform point cloud preprocessing on the lidar data, extract feature points and then transmit them to the data fusion module. The remote sensing data preprocessing unit is used to perform image preprocessing on the remote sensing data, extract feature points and then transmit them to the data fusion module; the feature points extracted by the lidar data preprocessing unit are labeled as lidar data feature points, and the feature points extracted by the remote sensing data preprocessing unit are labeled as remote sensing data feature points.

[0010] Preferably, the specific method for the data fusion module to match the feature points is: The feature points of the point cloud data are successively represented as D1, D2, D3, …, D n ; the feature points of the remote sensing data are successively represented as Y1, Y2, Y3, …, Y m ; calculate the matching degree between the feature points of the point cloud data and the feature points of the remote sensing data for matching, and take the two points with the highest matching degree as the best matching feature point pair, and all the best matching feature point pairs form a set U; The calculation formula of the matching degree is expressed as: , where PP ij is the matching degree between the i-th feature point of the point cloud data and the j-th feature point of the remote sensing data, ρ1 is the first matching coefficient, and ρ2 is the second matching coefficient; among them, the first matching coefficient is expressed as: , and the second matching coefficient is expressed as: ; among them, D ir is the feature point of the i-th point cloud data in the r-th dimension, Y jr is the feature point of the j-th point cloud data in the r-th dimension, R is the total dimension; i = 1, 2, 3, …, n, j = 1, 2, 3, …, m; The point cloud data feature point and the remote sensing data feature point corresponding to the highest matching degree, that is, the largest value of the matching degree, form the best matching feature point pair, and all the best matching feature point pairs form a set U: , where u g is the g-th best matching feature point pair, g = 1, 2, 3, …, G.

[0011] Preferably, the specific process of the data registration by the data fusion module is as follows: Based on the set U, use the best matching feature point pair to obtain the transformation matrix from the remote sensing data to the point cloud data. The transformation matrix is obtained by using the optimized weighted iterative closest point algorithm, and the transformation matrix includes a rotation matrix R and a translation vector T.

[0012] Preferably, the optimized iterative closest point algorithm includes the following steps: Step S11: Obtain each best matching feature point pair in the set U, and obtain the initial transformation matrix through the center point of the best matching feature point pair; Step S12: Use the improved optimization algorithm to obtain the weight of each best matching feature point pair; Step S13: Error minimization: Based on the rotation matrix R and the translation vector T of the initial transformation matrix, calculate the error function E(R, T); Step S14: Perform a rigid transformation on the transformation matrix to update the transformation matrix, and use the updated transformation matrix as the new initial transformation matrix; Step S15: Repeat steps S11 to S14 for iteration until the value of the error function is less than the preset threshold, then stop the iteration.

[0013] Preferably, the center point of the best matching feature point pair is the mean feature point of the point cloud data feature point and the remote sensing data feature point; the initial transformation matrix is the identity matrix; The error function E(R, T) is expressed by the calculation formula: , where k g is the weight of the gth best matching feature point pair, and RY j is the rotation matrix about Y j ; this error function represents the error between the remote sensing data Y j after being transformed by the transformation matrix and the point cloud data D i .

[0014] Preferably, obtaining the weight of each best matching feature point pair by using the improved optimization algorithm in step S12 includes: Step S21: Assign an arbitrary weight value to each best matching feature point pair. The range of the weight value is from 0 to 1. There are G weight values for G best matching feature point pairs, forming a weight value set K. ; Suppose there are N weight value sets, and the N weight value sets are represented as K1, K2, K3,..., KN. Step S22: Regard the N weight value sets as a monkey group. There are N monkeys in each monkey group, corresponding to the N weight value sets respectively; Initialize the monkey group, and define the scale of the monkey group, that is, the number of monkeys, the survival rate LIFE, and the discovery rate FIND. Step S23: Define the survival function. Step S24: Define the survival threshold. Step S25: Stop traversing when a calibrated surviving monkey group appears, and replace the remaining monkeys outside the calibrated surviving monkey group. Step S26: Randomly change the monkeys that have not been replaced among the remaining monkeys to get new monkeys. Step S27: Combine the new monkeys, the replaced monkeys, and the calibrated surviving monkey group to obtain a new generation of complete monkey group; Repeat the iteration until the change in the survival value of the monkey groups before and after iteration is less than the preset survival threshold, then stop the iteration to obtain the final monkey group. Step S28: Take the monkey with the highest survival value in the final monkey group as the weight value set corresponding to each best matching feature point pair as the weight.

[0015] Preferably, the formula of the survival function is expressed as: ; where f(I) is the survival function value of the weight value set corresponding to the Ith monkey, and SC IThe predicted survival value for the set of weight values corresponding to the \(i\)-th monkey; \(i = 1, 2, 3, \ldots, N\); The way to obtain the predicted survival value is as follows: Take the data corresponding to each monkey as the research data, input the research data into the survival value prediction model, and obtain the predicted survival value; For each monkey, calculate the value of the survival function of each monkey, denoted as the survival value; sort the survival values of all monkeys in the monkey group in descending order to obtain a monkey sequence arranged from high to low according to the survival value; calculate the cumulative survival value of each monkey, that is, the sum of the survival values of the first \(h\) monkeys; The calculation formula of the survival threshold is expressed as: , where \(YS\) is the survival threshold; \(LIEF\) is the preset survival rate, and \(LIFE\in[0, 1]\).

[0016] Preferably, the determination method of the calibrated survival monkey group is: Traverse each monkey in the monkey group. When the cumulative survival value, that is, the sum of the survival values of each monkey, is greater than or equal to the survival threshold, the monkey group composed of each monkey corresponding to the cumulative survival value is the calibrated survival monkey group; For the remaining monkeys outside the calibrated survival monkey group, that is, the monkeys in the monkey group that are not in the calibrated survival monkey group, the way to replace the remaining monkeys is: Determine the number \(Q\) of monkeys that need to be replaced: ; where \(round\) is the operation of rounding a numerical value, \(syh\) is the number of remaining monkeys outside the calibrated survival monkey group, and \(FIND\) is the preset discovery rate, and \(FIND\in[0, 1]\); Randomly select \(Q\) new monkeys from the monkey group to replace the \(Q\) monkeys at the end of the monkey sequence among the remaining monkeys, that is, the \(Q\) monkeys with smaller survival values; For the monkeys among the remaining monkeys that are not replaced, randomly change any one of their weight value elements to obtain new monkeys.

[0017] The technical effects and advantages of the present invention: The present invention is provided with a data fusion module, which facilitates the matching of the feature points extracted by the radar data preprocessing unit and the feature points extracted by the remote sensing data preprocessing unit to obtain a set of the best-matched feature point pairs. Then, the improved optimization algorithm is used to optimize the weighted iterative closest point algorithm, and the registered radar point cloud data and remote sensing data are fused to generate a comprehensive data set with higher information content and accuracy. The registration parameters are optimized through the optimization algorithm to reduce the error between the matched feature point pairs and improve the accuracy of feature point registration, thereby providing more accurate data for subsequent area measurement. Registering the feature points of the radar data and the feature points of the remote sensing data can ensure the spatial consistency of the two types of data and significantly improve the accuracy of the data. The lidar point cloud data provides accurate three-dimensional spatial coordinates, while the remote sensing data provides rich surface information. The lidar point cloud data may have occlusions or missing data in some areas, while the remote sensing data can provide additional information for these areas, and vice versa. Therefore, registering the lidar point cloud data and the remote sensing data can fill these information gaps, enhance the integrity of the data, and improve the spatial accuracy of the measurement. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 It is a structural diagram of a precise forest landscape area measurement system based on lidar according to the present invention.

[0019] Figure 2 It is a structural diagram of the data terminal according to the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0020] Next, the technical solutions in the present invention will be clearly and completely described in conjunction with the drawings in the present invention. In addition, the forms of the various structures described in the following embodiments are merely examples, and a precise forest landscape area measurement system based on lidar involved in the present invention is not limited to the various structures described in the following embodiments. All other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the scope of protection of the present invention.

[0021] Such as Figure 1As shown in the figure, the present invention provides a precise measurement system for forest landscape area based on lidar, including a device end, a data end, and an interface end; the device end includes a lidar device and a remote sensing device, the lidar device is used to obtain lidar data of a target object through lidar technology and transmit it to the data end, and the remote sensing device is used to obtain remote sensing data of the target object through remote sensing technology and transmit it to the data end; the lidar device includes, but is not limited to, airborne lidar systems and ground lidar systems, etc., which are devices that use lidar technology for data acquisition, and the remote sensing device is a device that uses remote sensing technology for remote sensing data acquisition; the lidar device continuously emits laser beams and receives reflected signals during data acquisition to generate high-precision three-dimensional point cloud data; the target object is the forest landscape for which area measurement is required. The data end is used to receive the data collected by the device end and perform analysis, such as Figure 2 shown, including a data acquisition module, a data preprocessing module, a data fusion module, and an area measurement module; The data acquisition module is used to send data acquisition instructions to the device end, obtain the collected lidar data and remote sensing data, and the device end collects the data respectively according to the data acquisition instructions; the data acquisition module includes a lidar data acquisition unit and a remote sensing data acquisition unit, the lidar data acquisition unit is used to receive the lidar data transmitted by the lidar device and then transmit it to the radar data preprocessing unit; the remote sensing data acquisition unit is used to receive the remote sensing data transmitted by the remote sensing device and then transmit it to the remote sensing data preprocessing unit. The data preprocessing module is used to perform preprocessing operations on the data of the data acquisition module to obtain directly usable data; the data preprocessing module includes a radar data preprocessing unit and a remote sensing data preprocessing unit, the radar data preprocessing unit is used to perform point cloud preprocessing on the lidar data, extract feature points and then transmit them to the data fusion module, and the remote sensing data preprocessing unit is used to perform image preprocessing on the remote sensing data, extract feature points and then transmit them to the data fusion module; the point cloud preprocessing includes, but is not limited to, preprocessing steps such as noise filtering, outlier removal, and downsampling to improve the quality of subsequent data fusion; the image preprocessing includes, but is not limited to, geometric correction and radiometric correction of remote sensing images to improve image quality and eliminate geometric distortion. The data fusion module is used to match the feature points extracted by the radar data preprocessing unit and the feature points extracted by the remote sensing data preprocessing unit, obtain a set of the best-matched feature point pairs, then optimize the weighted iterative closest point algorithm using an improved optimization algorithm, and then perform data registration. The registered radar point cloud data and remote sensing data are fused to generate a comprehensive data set with higher information content and accuracy. Its purpose is to optimize the registration parameters through the optimization algorithm to reduce the error between the matched feature point pairs, improve the accuracy of feature point registration, and thus provide more accurate data for subsequent area measurement. Registering the feature points of radar data and the feature points of remote sensing data can ensure the spatial consistency of the two types of data and significantly improve the accuracy of the data. The lidar point cloud data provides accurate three-dimensional spatial coordinates, while the remote sensing data provides rich surface information. There may be occlusions or missing data in some areas of the lidar point cloud data, and the remote sensing data can provide additional information for these areas, and vice versa. Therefore, registering the radar point cloud data and the remote sensing data can fill in these information gaps, enhance the integrity of the data, and improve the spatial accuracy of the measurement. The area measurement module extracts the boundary of the target object based on the edge detection algorithm, combines the data registered by the data fusion module, uses the clustering algorithm to cluster the feature points, obtains the corresponding categories of the feature points, and then performs area block division, and calculates the area based on the divided area blocks. The interface terminal is used for man-machine interaction display of the data.

[0022] In this embodiment, it should be specifically noted that after the radar data preprocessing unit extracts the feature points, they are labeled as point cloud data feature points, and any one of algorithms such as 3D-SIFT and SHOT algorithms can be used for feature point extraction. The SHOT algorithm is a local feature descriptor for three-dimensional point cloud data, which can capture the local geometry and texture features of the point cloud surface. After the remote sensing data preprocessing unit extracts the feature points, they are labeled as remote sensing data feature points, and any one of algorithms such as SIFT, SURF, and Harris algorithms can be used for feature point extraction. In practical applications, suitable feature extraction algorithms can be reasonably selected according to the characteristics of different data and the requirements of subsequent registration. The data features include spectral features, texture features, structural features, and terrain features, etc.

[0023] In this embodiment, it should be specifically noted that the specific method for the data fusion module to match the feature points is as follows: The point cloud data feature points are successively represented as D1, D2, D3, …, D n ; The remote sensing data feature points are successively represented as Y1, Y2, Y3, …, Y m; Calculate the matching degree between the feature points of the point cloud data and the feature points of the remote sensing data for matching, and take the two points with the highest matching degree as the best matching feature point pair. All the best matching feature point pairs form a set U; The calculation formula of the matching degree is expressed as: , where PP ij is the matching degree between the i-th feature point of the point cloud data and the j-th feature point of the remote sensing data, ρ1 is the first matching coefficient, and ρ2 is the second matching coefficient; among them, the first matching coefficient is expressed as: , and the second matching coefficient is expressed as: ; where D ir is the feature point of the i-th point cloud data in the r-th dimension, and Y jr is the feature point of the j-th point cloud data in the r-th dimension, and R is the total dimension; i = 1, 2, 3,..., n, j = 1, 2, 3,..., m; The point cloud data feature point and the remote sensing data feature point corresponding to the highest matching degree, that is, the maximum value of the matching degree, form the best matching feature point pair. All the best matching feature point pairs form a set U: , where u g is the g-th best matching feature point pair, and g = 1, 2, 3,..., G.

[0024] In this embodiment, it should be specifically noted that the specific process of the data registration by the data fusion module is as follows: Based on the set U, use the best matching feature point pair to obtain the transformation matrix from the remote sensing data to the point cloud data. The transformation matrix is obtained by using the optimized weighted iterative closest point (ICP) algorithm. The optimized weighted iterative closest point algorithm introduces the concept of weight. Introducing the concept of weight can improve the registration accuracy and robustness. The improved optimization algorithm is used to obtain the best weight value. At the same time, a neural network is introduced into the optimization algorithm to give full play to the non-linear and multi-modal data modeling capabilities of the neural network and effectively capture complex data relationships; according to the parallel computing processing mechanism, the computing efficiency is improved; the transformation matrix includes a rotation matrix R and a translation vector T; The optimized iterative closest point algorithm includes the following steps: Step S11: Obtain each best matching feature point pair in the set U, and obtain the initial transformation matrix through the center point of the best matching feature point pair; Step S12: Use the improved optimization algorithm to obtain the weight of each best matching feature point pair; Step S13: Error minimization: Based on the rotation matrix R and the translation vector T of the initial transformation matrix, calculate the error function E(R, T); Step S14: Perform a rigid transformation on the transformation matrix to update the transformation matrix, and use the updated transformation matrix as the new initial transformation matrix; Step S15: Repeat steps S11 to S14 for iteration until the value of the error function is less than the preset threshold, at which point the iteration stops.

[0025] In this embodiment, it should be specifically noted that the center point of the best matching feature point pair is the mean feature point of the point cloud data feature point and the remote sensing data feature point; the initial transformation matrix can be set as the identity matrix; The error function E(R, T) is expressed by the following calculation formula: , where k g is the weight of the g-th best matching feature point pair, RY j is the rotation matrix with respect to Y j ; this error function represents the error between the remote sensing data Y j after being transformed by the transformation matrix and the point cloud data D i .

[0026] In this embodiment, it should be specifically noted that obtaining the weight of each best matching feature point pair by using the improved optimization algorithm in step S12 includes: Step S21: Assign an arbitrary weight value to each best matching feature point pair. The range of the weight value is from 0 to 1. For G best matching feature point pairs, there are G weight values, forming a weight value set K. ; Suppose there are N weight value sets, then the N weight value sets are represented as K1, K2, K3,..., KN; Step S22: Regard the N weight value sets as a monkey group. There are N monkeys in each monkey group, corresponding to the N weight value sets respectively; Initialize the monkey group, and define the scale of the monkey group, that is, the number of monkeys, the survival rate LIFE, and the discovery rate FIND. Step S23: Define the survival function; Step S24: Define the survival threshold; Step S25: Stop traversing when a calibrated survival monkey group appears, and replace the remaining monkeys outside the calibrated survival monkey group; Step S26: Make random changes to the monkeys that have not been replaced among the remaining monkeys to obtain new monkeys; Step S27: Combine the new monkeys, the replaced monkeys, and the calibrated survival monkey group to obtain a new generation of complete monkey group; Repeat the iteration until the change in the survival value of the monkey groups before and after iteration is less than the preset survival threshold, at which point the iteration stops, and the final monkey group is obtained; Step S28: Use the monkey with the highest survival value in the final monkey group as the weight value set corresponding to each best matching feature point pair as the weight.

[0027] In this embodiment, it should be specifically noted that the formula of the survival function is expressed as: ; Among them, f(I) is the survival function value of the weight value set corresponding to the I-th monkey, SC I is the predicted survival value of the weight value set corresponding to the I-th monkey; I = 1, 2, 3, ..., N; For each monkey, calculate the value of the survival function of each monkey, which is recorded as the survival value; sort the survival values of all monkeys in the monkey group in descending order to obtain a monkey sequence arranged from high to low according to the survival value; calculate the cumulative survival value of each monkey, that is, the sum of the survival values of the previous h monkeys; The calculation formula of the survival threshold is expressed as: , where YS is the survival threshold; LIEF is the preset survival rate, satisfying LIFE∈[0,1]; the specific value can be set by technical personnel in the field; The method for determining the surviving monkey group is as follows: Traverse each monkey in the monkey group. When the cumulative survival value, that is, the sum of the survival values of each monkey, is greater than or equal to the survival threshold, the monkey group consisting of each monkey corresponding to the cumulative survival value is the calibrated survival monkey group. For example, if the sum of the survival values of the five monkeys is exactly equal to the survival threshold when traversing to the fifth monkey, the monkey group consisting of these five monkeys is the calibrated survival monkey group. The remaining monkeys outside the calibrated surviving monkey group, i.e., the monkeys in the monkey group that are not calibrated surviving monkey groups, are replaced in the following manner: Determine the number of monkeys Q that need to be replaced: ; Among them, round is the rounding operation for a value, syh is the number of monkeys remaining outside the calibrated surviving monkey group, and FIND is the preset discovery rate, which satisfies FIND∈[0,1]. The specific value can be set by technical personnel in the field. Randomly select Q new monkeys from the monkey group to replace the Q monkeys at the end of the monkey sequence in the remaining monkeys, that is, the Q monkeys with smaller survival values; For the remaining monkeys that have not been replaced, randomly change any of their weight value elements to obtain new monkeys.

[0028] In this embodiment, it should be specifically explained that the method for obtaining the predicted survival value is: The data corresponding to each monkey is used as research data, wherein the research data is the weight value of each best matching feature point pair corresponding to the weight value set, and the research data is input into the survival value prediction model to obtain the predicted survival value; The training process of the survival value prediction model is: The weight of each matching feature point pair is used as analysis data, the matching degrees corresponding to multiple sets of analysis data are collected in advance, and the analysis data and the corresponding matching degrees are converted into a corresponding set of feature vectors; Use each set of feature vectors as the input of the survival value prediction model. The survival value prediction model outputs a set of prediction matching degrees corresponding to each set of analysis data, and uses the value of the prediction matching degree as the predicted survival value. Take the actual matching degree corresponding to each set of research data as the prediction target, where the actual matching degree is the matching degree corresponding to the analysis data collected in advance as described above. Take minimizing the sum of prediction errors of all analysis data as the training target. The calculation formula of the prediction error is as follows: , where ε p is the prediction error, p is the group number of the feature vectors corresponding to the analysis data, θ p is the prediction matching degree corresponding to the p-th set of analysis data, and μ p is the actual matching degree corresponding to the p-th set of analysis data. Train the survival value prediction model until the sum of prediction errors reaches convergence and then stop training; The matching degree corresponding to the analysis data is obtained by those skilled in the art through multiple uses of the iterative closest point algorithm for multiple sets of different analysis data to obtain the corresponding registration accuracy. The greater the registration accuracy, the greater the corresponding matching degree. The registration accuracy can be evaluated by calculating the mean square error or the average distance error; The survival value prediction model is specifically a deep neural network model, which includes an input layer, a hidden layer, and an output layer. Each hidden layer includes multiple neurons, and there are connections between each neuron and the neurons in the next layer. The connections contain weights, which determine the importance and influence of data transmission in the neural network. An activation function is applied to each neuron between the hidden layer and the output layer. The activation function introduces non-linearity, allowing the network to learn more complex patterns and features.

[0029] In this embodiment, it should be specifically noted that the area measurement module combines the registered data with the band information in the remote sensing data and uses a clustering algorithm for clustering. The clustering algorithm can be any one of clustering algorithms such as K-means clustering, spectral clustering, hierarchical clustering, and DBSCAN clustering. After clustering, L categories are formed, which respectively correspond to L forest landscape units. The forest landscape unit is a forest landscape structure, including various patches, corridors, etc. In the forest landscape, patches can be different types of plant communities, such as lakes, grasslands, coniferous forests, etc.; corridors can be rivers, roads, etc.; Use an edge detection algorithm to obtain the boundary of the target object. The enclosed area formed by each category within the boundary range is used as an area block. Measure and calculate the areas of all area blocks, and sum up the areas of all area blocks to obtain the area of the target object. The extraction of boundary points by the edge detection algorithm is a prior art, and this embodiment will not elaborate on it here. The area of the area block can be calculated by numerical integration or geometric methods for the area enclosed by a closed curve; The interface terminal outputs the total area of the target object, and at the same time outputs the names and areas of the forest landscape units corresponding to each category; while obtaining the area of the target object, it can understand the area composition of each landscape unit of the target object.

[0030] Finally: The above are only the preferred embodiments of the present invention and are not used to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

[0031] The above is only the specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present application can easily think of changes or replacements, which should all be covered by the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claimed rights.

Claims

1. A precise measurement system for forest landscape area based on lidar, characterized in that: It includes a device side, a data side, and an interface side; The device side includes a lidar device and a remote sensing device, and the device side collects data respectively according to data acquisition instructions; The data side is used to receive the data collected by the device side and perform analysis, including a data acquisition module, a data preprocessing module, a data fusion module, and an area measurement module; The interface side is used for human-computer interaction display of data; The data acquisition module is used to send data acquisition instructions to the device side and obtain the collected lidar data and remote sensing data; The data preprocessing module is used to perform preprocessing operations on the data of the data acquisition module; the data preprocessing module includes a lidar data preprocessing unit and a remote sensing data preprocessing unit; The data fusion module is used to match the feature points extracted by the lidar data preprocessing unit with the feature points extracted by the remote sensing data preprocessing unit, obtain a set of best-matched feature point pairs, then optimize the weighted iterative closest point algorithm using an improved optimization algorithm, and then perform data registration; The area measurement module extracts the boundary of the target object based on the edge detection algorithm, combines the data after registration by the data fusion module, clusters the feature points using the clustering algorithm, obtains the categories corresponding to the feature points and then performs area block division, and measures and calculates the area based on the divided area blocks.

2. The precise measurement system for forest landscape area based on lidar according to claim 1, characterized in that: The lidar device is used to obtain lidar data of the target object through lidar technology and transmit it to the data side, and the remote sensing device is used to obtain remote sensing data of the target object through remote sensing technology and transmit it to the data side; the target object is a forest landscape that needs to be measured for area.

3. The precise measurement system for forest landscape area based on lidar according to claim 2, wherein: The data acquisition module includes a lidar data acquisition unit and a remote sensing data acquisition unit. The lidar data acquisition unit is used to receive the lidar data transmitted by the lidar device and then transmit it to the lidar data preprocessing unit; The remote sensing data acquisition unit is used to receive the remote sensing data transmitted by the remote sensing device and then transmit it to the remote sensing data preprocessing unit; The lidar data preprocessing unit is used to perform point cloud preprocessing on the lidar data, extract feature points and then transmit them to the data fusion module. The remote sensing data preprocessing unit is used to perform image preprocessing on the remote sensing data, extract feature points and then transmit them to the data fusion module; After the lidar data preprocessing unit extracts feature points, they are labeled as lidar data feature points. After the remote sensing data preprocessing unit extracts feature points, they are labeled as remote sensing data feature points.

4. The precise measurement system for forest landscape area based on lidar according to claim 3, wherein: The specific method for the data fusion module to match feature points is as follows: The feature points of the point cloud data are successively represented as D1, D2, D3, …, D n ; the feature points of the remote sensing data are successively represented as Y1, Y2, Y3, …, Y m ; calculate the matching degree between the feature points of the point cloud data and the feature points of the remote sensing data for matching, and take the two points with the highest matching degree as the best matching feature point pair. All the best matching feature point pairs form a set U; The calculation formula of the matching degree is expressed as: , where PP ij is the matching degree between the i-th feature point of the point cloud data and the j-th feature point of the remote sensing data, ρ1 is the first matching coefficient, and ρ2 is the second matching coefficient; among them, the first matching coefficient is expressed as: , and the second matching coefficient is expressed as: ; where D ir is the feature point of the i-th point cloud data in the r-th dimension, and Y jr is the feature point of the j-th point cloud data in the r-th dimension, and R is the total dimension; i = 1, 2, 3, …, n, j = 1, 2, 3, …, m; The point cloud data feature points corresponding to the highest matching degree, i.e., the largest value of the matching degree, and the remote sensing data feature points form the best matching feature point pairs, and all the best matching feature point pairs form a set U: , where u g is the g-th best matching feature point pair, and g = 1, 2, 3, …, G.

5. The precise measurement system for forest landscape area based on lidar according to claim 4, characterized in that: The specific process for the data fusion module to perform data registration is as follows: Based on the set U, use the best-matched feature point pairs to obtain a transformation matrix from remote sensing data to point cloud data. The transformation matrix is obtained using the optimized weighted iterative closest point algorithm, and the transformation matrix includes a rotation matrix R and a translation vector T.

6. The precise measurement system for forest landscape area based on lidar according to claim 5, wherein: The optimized iterative closest point algorithm includes the following steps: Step S11: Obtain each best-matched feature point pair in the set U, and obtain an initial transformation matrix through the center point of the best-matched feature point pair; Step S12: Use the improved optimization algorithm to obtain the weights of each pair of best-matching feature points; Step S13: Error minimization: Based on the rotation matrix R and translation vector T of the initial transformation matrix, calculate the error function E(R, T); Step S14: Perform a rigid transformation on the transformation matrix to update the transformation matrix, and use the updated transformation matrix as the new initial transformation matrix; Step S15: Repeat Steps S11 to S14 iteratively until the value of the error function is less than the preset threshold, then stop the iteration.

7. The precise measurement system for forest landscape area based on lidar according to claim 6, characterized in that: The center point of the pair of best-matching feature points is the mean feature point of the point cloud data feature point and the remote sensing data feature point; the initial transformation matrix is the identity matrix; The error function E(R, T) is expressed by the calculation formula as follows: , where k g is the weight of the g-th best matching feature point pair, and RY j is the rotation matrix with respect to Y j ; this error function represents the error between the remote sensing data Y j after being transformed by the transformation matrix and the point cloud data D i .

8. The precise measurement system for forest landscape area based on lidar according to claim 7, characterized in that: The process of obtaining the weights of each pair of best-matching feature points by using the improved optimization algorithm in Step S12 includes: Step S21: Assign an arbitrary weight value to each pair of best-matching feature points. The range of the weight value is from 0 to 1. For G pairs of best-matching feature points, there are G weight values, which form a weight value set K. ; Assume there are N weight value sets, and the N weight value sets are represented as K1, K2, K3, …, KN. Step S22: Regard N sets of weight values as a monkey population. There are N monkeys in each monkey population, corresponding to N sets of weight values respectively; Initialize the monkey population, and define the size of the monkey population, i.e., the number of monkeys, the survival rate LIFE, and the discovery rate FIND; Step S23: Define the survival function; Step S24: Define the survival threshold; Step S25: Stop traversing when a calibrated survival monkey population appears, and replace the remaining monkeys outside the calibrated survival monkey population; Step S26: Randomly change the un-replaced monkeys among the remaining monkeys to get new monkeys; Step S27: Combine the new monkeys, the replaced monkeys, and the calibrated survival monkey population to obtain a new generation of complete monkey population; Repeat the iteration until the change in the survival values of the monkey populations before and after iteration is less than the preset survival threshold, then stop the iteration to obtain the final monkey population; Step S28: Take the monkey with the highest survival value in the final monkey population as the corresponding set of weight values as the weights of each pair of best-matching feature points.

9. The precise measurement system for forest landscape area based on lidar according to claim 8, characterized in that: The formula of the survival function is expressed as: ; where f(I) is the survival function value of the weight value set corresponding to the I-th monkey, and SC I is the predicted survival value of the weight value set corresponding to the I-th monkey; I = 1, 2, 3, …, N; The way to obtain the predicted survival value is: Take the data corresponding to each monkey as the research data, input the research data into the survival value prediction model to obtain the predicted survival value; For each monkey, calculate the value of the survival function of each monkey, denoted as the survival value; Sort the survival values of all monkeys in the monkey population in descending order to obtain a monkey sequence arranged from high to low according to the survival value; Calculate the cumulative survival value of each monkey, that is, the sum of the survival values of the first h monkeys; The calculation formula of the survival threshold is expressed as: , where YS is the survival threshold; LIEF is the preset survival rate, and LIFE ∈ [0, 1].

10. The precise measurement system for forest landscape area based on lidar according to claim 9, characterized in that: The determination method of the calibrated survival monkey population is: Traverse each monkey in the monkey population. When the cumulative survival value, that is, the sum of the survival values of each monkey, is greater than or equal to the survival threshold, the monkey population composed of each monkey corresponding to the cumulative survival value is the calibrated survival monkey population; The remaining monkeys outside the calibrated survival monkey population, that is, the monkeys in the monkey population that are not in the calibrated survival monkey population. The way to replace the remaining monkeys is: Determine the number of monkeys Q to be replaced: ; where, round is the operation of rounding a numerical value, syh is the remaining number of monkeys outside the calibrated surviving monkey group, FIND is the preset discovery rate, and FIND ∈ [0, 1]; Randomly select Q new monkeys from the monkey population to replace the Q monkeys at the end of the monkey sequence among the remaining monkeys, that is, the Q monkeys with smaller survival values; For the un-replaced monkeys among the remaining monkeys, randomly change any one of their weight value elements to get new monkeys.

Images