A method and system for evaluating camphor tree biomass

Abstract

The present invention discloses a camphor tree biomass assessment method and system, which relates to the technical field of camphor tree image processing, including segmenting image data according to information loss degree, extracting structural parameters of camphor tree from the segmented image data; setting the interval length of each environmental factor data according to the degree of change of each environmental factor data, combining different interval segments of multiple environmental factor data into a growth environment scenario, and determining the environmental impact coefficient corresponding to the growth environment scenario; determining the basic growth rate of camphor tree; and determining the correspondence between the environmental factor data of camphor tree and the point cloud data of camphor tree according to the biological factor data of camphor tree, and assessing and predicting the biomass of camphor tree through the growth model of camphor tree. The growth model of camphor tree is trained, and the influence of environmental factors and biological factors are combined, so as to improve the assessment accuracy of camphor tree biomass, be able to predict the future biomass of camphor tree, and ensure the reliability of camphor tree biomass assessment.

Description

Technical Field

[0001] The present application relates to the technical field of camphor tree image processing, and more specifically, to a camphor tree biomass assessment method and system. Background Art

[0002] Cinnamomum longepaniculatum is an important spice and medicinal plant, and the accurate assessment of its biomass is of great significance for resource management and sustainable utilization. Traditional biomass assessment methods rely on field measurements and harvesting methods, which are time-consuming, labor-intensive and costly. In recent years, with the development of remote sensing technology, artificial intelligence algorithms and cloud computing platforms, it has become possible to establish an automated, high-precision biomass assessment system using these technologies. These technologies can use scanning equipment to collect point cloud data, three-dimensional modeling and other methods to obtain the morphological parameters of Cinnamomum longepaniculatum, and combine machine learning algorithms to establish a biomass estimation model to achieve accurate assessment and real-time monitoring of Cinnamomum longepaniculatum biomass.

[0003] In the prior art, when evaluating the biomass of Cinnamomum camphora, the influence of environmental factors and biological factors of Cinnamomum camphora itself is not taken into account, resulting in poor accuracy and low reliability of Cinnamomum camphora biomass evaluation.

[0004] Therefore, how to improve the accuracy and reliability of Cinnamomum oleifera biomass assessment is a technical problem that needs to be solved. Summary of the invention

[0005] The present invention provides a method for evaluating the biomass of camphor trees, which is used to solve the technical problems of poor accuracy and low reliability of the biomass evaluation of camphor trees in the prior art. The method comprises:

[0006] Acquire point cloud data of camphor tree, pre-process the point cloud data, extract feature points in the point cloud data, convert the point cloud data into image data, extract feature points in the image data, and determine the information loss degree by comparing the difference between the feature points in the point cloud data and the feature points in the image data, segment the image data according to the information loss degree, and extract the structural parameters of camphor tree from the segmented image data;

[0007] Acquire the environmental factor data of camphor trees, set the interval length of each environmental factor data according to the degree of change of each environmental factor data, disassemble the range of each environmental factor data by the interval length to obtain multiple interval segments of each environmental factor data, combine different interval segments of multiple environmental factor data into a growth environment scenario, and determine the environmental impact coefficient corresponding to the growth environment scenario;

[0008] Acquire biological factor data of camphor tree, and construct a camphor tree growth rate change curve chart based on the biological factor data to determine the basic growth rate of camphor tree;

[0009] The environmental factor data, biological factor data and point cloud data of camphor trees over a period of time are obtained, and the corresponding relationship between the environmental factor data and the point cloud data of camphor trees is determined based on the biological factor data of camphor trees, and a growth model of camphor trees is trained. The biomass of camphor trees is evaluated and predicted through the growth model of camphor trees.

[0010] In some embodiments of the present application, the information loss degree is determined by comparing the difference between the feature points in the point cloud data and the feature points in the image data, including:

[0011] According to the matching relationship between the feature points in the point cloud data and the feature points in the image data, the feature points are divided into two categories: matching feature points and non-matching feature points;

[0012] According to the types of feature point attributes in the point cloud data and the feature point attributes in the image data, the feature point attributes are divided into common attributes;

[0013] Normalize the common attributes of the matching feature points in the point cloud data and the image data, and perform difference analysis on the common attributes of the two feature points to obtain the difference of each common attribute, retain the common attributes whose difference exceeds the corresponding difference threshold, and integrate the differences of the retained common attributes to determine the overall difference of the common attributes of the matching feature points;

[0014] For non-matching feature points in the point cloud data and the image data, the same common attributes of all non-matching feature points are integrated to obtain the comprehensive values ​​of multiple common attributes of the non-matching feature points in the point cloud data and the comprehensive values ​​of multiple common attributes of the non-matching feature points in the image data;

[0015] The information loss degree is determined based on the overall difference of the common attributes of the matching feature points, the comprehensive value of multiple common attributes of the non-matching feature points in the point cloud data, and the comprehensive value of multiple common attributes of the non-matching feature points in the image data.

[0016] In some embodiments of the present application, the information loss degree is determined based on the overall difference of the common attributes of the matching feature points, the comprehensive value of multiple common attributes of the non-matching feature points in the point cloud data, and the comprehensive value of multiple common attributes of the non-matching feature points in the image data, including:

[0017]

[0018] Among them, Z is the information loss degree, α1 is the loss weight of the matching feature points, E1 is the overall difference of the common attributes of the matching feature points, α2 is the loss weight of the non-matching feature points, n is the number of common attributes, E2 i is the comprehensive value of the common attribute of the non-matching feature point in the point cloud data, E3i is the comprehensive value of the i-th common attribute of the non-matching feature points in the image data, m2 is the number of non-matching feature points in the point cloud data, m3 is the number of non-matching feature points in the image data, and k is a preset constant.

[0019] In some embodiments of the present application, segmenting the image data according to the information loss degree includes:

[0020] An initial feature similarity threshold is set by the feature distribution of the image data, the information loss degree is mapped to obtain a feature similarity correction coefficient, and the initial feature similarity threshold is corrected by the feature similarity correction coefficient to obtain a feature similarity threshold;

[0021] The pixels in the image are integrated into multiple regions based on the feature similarity threshold to complete the segmentation of the image data.

[0022] In some embodiments of the present application, the interval length of each environmental factor data is set according to the degree of change of each environmental factor data, including:

[0023] Determine the seasonal cycle near the area where the camphor tree is located, find the time period with the largest change in environmental factor data within the seasonal cycle, and decompose the seasonal cycle into multiple time periods based on the length of this time period;

[0024] Calculate the coefficient of variation and average rate of change of environmental factor data in each period, and calculate the degree of change of each environmental factor data within the seasonal cycle through the coefficient of variation and average rate of change in each period;

[0025]

[0026] Among them, τ is the degree of change of environmental factor data, b is the number of time periods in the seasonal cycle, σ t is the coefficient of variation in the tth period, a t is the average rate of change in the tth period, and v is a preset constant;

[0027] The length of the data interval is determined according to the degree of change of the data of each environmental factor, and different degrees of change correspond to different data interval lengths.

[0028] In some embodiments of the present application, a growth rate change curve of camphor tree is constructed based on biological factor data to determine the basic growth rate of camphor tree, including:

[0029] A camphor tree simulation model under an ideal environment is established, the biological factor data of camphor tree is input into the camphor tree simulation model, and the structural parameters of camphor tree are output, so as to construct a camphor tree growth rate change curve chart, and the average value of the camphor tree growth rate in the camphor tree growth rate change curve chart is taken as the basic growth rate of camphor tree.

[0030] In some embodiments of the present application, the corresponding relationship between the environmental factor data of camphor and the point cloud data of camphor is determined based on the biological factor data of camphor, including:

[0031] The basic growth rate of camphor tree is determined by the biological factor data of camphor tree, and the basic growth rate of camphor tree is compared with the basic growth rate interval of standard camphor tree to obtain the growth rate deviation;

[0032] The frontier time scale is determined according to the growth rate deviation, and the starting time node of the environmental factor data of camphor is supplemented according to the size of the frontier time scale, so as to re-determine the starting time node of the environmental factor data of camphor;

[0033] The time nodes of the environmental factor data segment of camphor tree and the point cloud data segment of camphor tree are corresponded as the corresponding relationship between the two.

[0034] In some embodiments of the present application, training a growth model of Cinnamomum camphora includes:

[0035] Split the data segment of the correspondence between the environmental factor data of camphor and the point cloud data of camphor into multiple parts, and insert the biological factor data of camphor into each part in chronological order to form multiple samples;

[0036] Determine the structural parameters, environmental impact coefficient and basic growth rate of camphor tree according to the corresponding data in each sample, and mark the structural parameters, environmental impact coefficient and basic growth rate of camphor tree on each sample;

[0037] Multiple samples are divided into training sets or test sets respectively, and the growth model of Cinnamomum camphora is trained according to the training sets or test sets.

[0038] Correspondingly, the present application also provides a Cinnamomum camphora biomass assessment system, comprising:

[0039] An extraction module is used to obtain point cloud data of camphor tree, pre-process the point cloud data, extract feature points in the point cloud data, convert the point cloud data into image data, extract feature points in the image data, and determine the information loss degree by comparing the difference between the feature points in the point cloud data and the feature points in the image data, segment the image data according to the information loss degree, and extract the structural parameters of camphor tree from the segmented image data;

[0040] A construction module is used to obtain the environmental factor data of camphor, set the interval length of each environmental factor data according to the degree of change of each environmental factor data, disassemble the range of each environmental factor data by the interval length to obtain multiple interval segments of each environmental factor data, combine different interval segments of multiple environmental factor data into a growth environment scenario, and determine the environmental impact coefficient corresponding to the growth environment scenario;

[0041] The simulation module is used to obtain the biological factor data of camphor tree, and to construct a curve chart of the growth rate change of camphor tree according to the biological factor data, so as to determine the basic growth rate of camphor tree;

[0042] The evaluation module is used to obtain the environmental factor data, biological factor data and point cloud data of camphor trees over a period of time, and determine the corresponding relationship between the environmental factor data and the point cloud data of camphor trees based on the biological factor data of camphor trees, train the growth model of camphor trees, and evaluate and predict the biomass of camphor trees through the growth model of camphor trees.

[0043] By applying the above technical scheme, the point cloud data of camphor tree is obtained, the point cloud data is preprocessed, the feature points in the point cloud data are extracted, the point cloud data is converted into image data, the feature points in the image data are extracted, and the information loss degree is determined by comparing the difference between the feature points in the point cloud data and the feature points in the image data, the image data is segmented according to the information loss degree, and the structural parameters of camphor tree are extracted from the segmented image data; the environmental factor data of camphor tree is obtained, the interval length of each environmental factor data is set according to the degree of change of each environmental factor data, and the range of each environmental factor data is disassembled by the interval length to obtain each environmental factor data. Multiple intervals of factor data, different intervals of multiple environmental factor data are combined into a growth environment scenario, and the environmental impact coefficient corresponding to the growth environment scenario is determined; the biological factor data of camphor tree is obtained, and a camphor tree growth rate change curve is constructed based on the biological factor data to determine the basic growth rate of camphor tree; the environmental factor data, the biological factor data and the point cloud data of camphor tree within a period of time are obtained, and the corresponding relationship between the environmental factor data and the point cloud data of camphor tree is determined based on the biological factor data of camphor tree, the growth model of camphor tree is trained, and the biomass of camphor tree is evaluated and predicted through the growth model of camphor tree. The present application segments the image data according to the degree of information loss, thereby ensuring the accuracy of image segmentation, thereby improving the accuracy of feature extraction, and being able to accurately obtain the structural parameters of camphor tree. By defining different growth environment scenarios, the impact of the environment on camphor tree is determined, the growth model of camphor tree is trained, the impact of environmental factors and the impact of biological factors are combined, the evaluation accuracy of camphor tree biomass is improved, the future biomass of camphor tree can be predicted, and the reliability of camphor tree biomass evaluation is ensured. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0045] Figure 1 A schematic flow chart of a method for evaluating Cinnamomum camphora biomass proposed in an embodiment of the present invention is shown;

[0046] Figure 2 A structural schematic diagram of a Cinnamomum camphora biomass assessment system proposed in an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0047] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0048] The present application embodiment provides a method for evaluating the biomass of camphor trees. Figure 1 As shown, the method comprises the following steps:

[0049] Step S101, obtaining point cloud data of camphor tree, preprocessing the point cloud data, extracting feature points in the point cloud data, converting the point cloud data into image data, extracting feature points in the image data, and determining the degree of information loss by comparing the difference between the feature points in the point cloud data and the feature points in the image data, segmenting the image data according to the degree of information loss, and extracting the structural parameters of camphor tree from the segmented image data.

[0050] In this embodiment, laser scanners, drones and other equipment are used to collect point cloud data of camphor trees. Key feature points, such as the position and shape of trunks, branches and leaves, can be extracted from the point cloud data by calculating the geometric properties of the point cloud, such as curvature and curvature change rate. Converting three-dimensional point cloud data into two-dimensional images is to facilitate the use of traditional computer vision and image processing techniques to process data. Although three-dimensional point cloud data contains rich spatial information, many existing image processing algorithms and tools are designed for two-dimensional images, and the three-dimensional data requires large computing resources and large error fluctuations in specific analysis and processing. In the process of converting point cloud data into image data, due to algorithm or resource limitations, there may be many unmatched feature points. The degree of information loss in the conversion process is defined by combining the differences in matching feature points and the situation of unmatched feature points. Extracting feature points from a two-dimensional image may include edge detection, contour extraction and other methods.

[0051] In this embodiment, the degree of information loss during the conversion process can help to perform more accurate image segmentation, and then extract the corresponding structural parameters of Cinnamomum camphora from the segmented regions, including height, crown width, crown volume, etc.

[0052] In some embodiments of the present application, the information loss degree is determined by comparing the difference between the feature points in the point cloud data and the feature points in the image data, including:

[0053] According to the matching relationship between the feature points in the point cloud data and the feature points in the image data, the feature points are divided into two categories: matching feature points and non-matching feature points;

[0054] According to the types of feature point attributes in the point cloud data and the feature point attributes in the image data, the feature point attributes are divided into common attributes;

[0055] Normalize the common attributes of the matching feature points in the point cloud data and the image data, and perform difference analysis on the common attributes of the two feature points to obtain the difference of each common attribute, retain the common attributes whose difference exceeds the corresponding difference threshold, and integrate the differences of the retained common attributes to determine the overall difference of the common attributes of the matching feature points;

[0056] For non-matching feature points in the point cloud data and the image data, the same common attributes of all non-matching feature points are integrated to obtain the comprehensive values ​​of multiple common attributes of the non-matching feature points in the point cloud data and the comprehensive values ​​of multiple common attributes of the non-matching feature points in the image data;

[0057] The information loss degree is determined based on the overall difference of the common attributes of the matching feature points, the comprehensive value of multiple common attributes of the non-matching feature points in the point cloud data, and the comprehensive value of multiple common attributes of the non-matching feature points in the image data.

[0058] In this embodiment, the feature points are divided into two categories: matching feature points and non-matching feature points according to the matching relationship between the feature points in the point cloud data and the feature points in the image data. The matching relationship can be determined by a matching algorithm, such as FLANN (Fast Library for Approximate Nearest Neighbors, which is an efficient approximate nearest neighbor search algorithm that can be used to find matching feature point pairs in point cloud data and image data.

[0059] In this embodiment, the common attributes of feature points, such as position, shape, and direction, exist in both point cloud data and image data. Position: feature points have clear spatial positions, which allows them to be used to describe the shape and structure of objects. Shape: the area around the feature point may have specific shape features, such as the round shape of a tree trunk or the branch shape of a branch. Direction: feature points may have specific directional information, such as the growth direction of a branch or the arrangement direction of leaves.

[0060] In this embodiment, the common attributes whose difference exceeds the corresponding difference threshold are retained, and the differences of the retained common attributes are integrated to determine the overall difference of the common attributes of the matching feature points. The differences of the common attributes between the matching feature points are compared, the types with larger differences are retained, and all the retained types are integrated (weighted sum) to obtain the overall difference. For non-matching feature points, the same common attributes of all feature points are integrated to determine the comprehensive value.

[0061] In some embodiments of the present application, the information loss degree is determined based on the overall difference of the common attributes of the matching feature points, the comprehensive value of multiple common attributes of the non-matching feature points in the point cloud data, and the comprehensive value of multiple common attributes of the non-matching feature points in the image data, including:

[0062]

[0063] Among them, Z is the information loss degree, α1 is the loss weight of the matching feature points, E1 is the overall difference of the common attributes of the matching feature points, α2 is the loss weight of the non-matching feature points, n is the number of common attributes, E2 i is the comprehensive value of the common attribute of the non-matching feature point in the point cloud data, E3 iis the comprehensive value of the i-th common attribute of the non-matching feature points in the image data, m2 is the number of non-matching feature points in the point cloud data, m3 is the number of non-matching feature points in the image data, and k is a preset constant.

[0064] In this embodiment, the information loss degree is defined by the overall difference of the common attributes of the matching feature points and the comprehensive value of the common attributes of the non-matching feature points. It indicates the correction of the difference between the number of non-matching feature points in the point cloud data and the number of non-matching feature points in the image data on the difference in the comprehensive value. The corresponding comprehensive value will be affected by the different number of non-matching feature points.

[0065] In some embodiments of the present application, segmenting the image data according to the information loss degree includes:

[0066] An initial feature similarity threshold is set by the feature distribution of the image data, the information loss degree is mapped to obtain a feature similarity correction coefficient, and the initial feature similarity threshold is corrected by the feature similarity correction coefficient to obtain a feature similarity threshold;

[0067] The pixels in the image are integrated into multiple regions based on the feature similarity threshold to complete the segmentation of the image data.

[0068] In this embodiment, the features of the image data, such as color, texture, brightness, etc., have different feature similarity correction coefficients corresponding to different information loss degrees, and the feature similarity threshold = feature similarity correction coefficient * initial feature similarity threshold. The pixels in the image are divided into different regions according to their similarity (such as color, texture, brightness, etc.), and similar pixels are gradually added to the current region until no new pixels can be added.

[0069] Step S102, obtain the environmental factor data of camphor tree, set the interval length of each environmental factor data according to the degree of change of each environmental factor data, decompose the range of each environmental factor data by the interval length to obtain multiple interval segments of each environmental factor data, combine the different interval segments of multiple environmental factor data into a growth environment scenario, and determine the environmental impact coefficient corresponding to the growth environment scenario.

[0070] In this embodiment, the environmental factor data of camphor, such as temperature, precipitation, light intensity, etc., can be obtained through a weather station or the like. The degree of change of the environmental factor data describes the fluctuation of each environmental factor data. If the fluctuation is large, a smaller interval length needs to be set for monitoring. Conversely, the same applies. The interval length refers to the numerical range of the environmental factor data. For example, the range of a certain environmental factor data is 0-100, and the interval length is 20. Then, according to the numerical range of 20, the range is divided into 5 interval segments, namely 0-20, 20-40, 40-60, 60-80, and 80-100. Different intervals of all kinds of environmental factor data are combined into a growth environment scenario, and matched according to the preset impact table to determine the environmental impact coefficient.

[0071] It should be noted that the preset impact table is the degree of influence of the overall environment on the growth of camphor trees, determined by historical data, mathematical models, etc. The table is configured with environmental impact coefficients under all growth environment scenarios.

[0072] In some embodiments of the present application, the interval length of each environmental factor data is set according to the degree of change of each environmental factor data, including:

[0073] Determine the seasonal cycle near the area where the camphor tree is located, find the time period with the largest change in environmental factor data within the seasonal cycle, and decompose the seasonal cycle into multiple time periods based on the length of this time period;

[0074] Calculate the coefficient of variation and average rate of change of environmental factor data in each period, and calculate the degree of change of each environmental factor data within the seasonal cycle through the coefficient of variation and average rate of change in each period;

[0075]

[0076] Among them, τ is the degree of change of environmental factor data, b is the number of time periods in the seasonal cycle, σ t is the coefficient of variation in the tth period, a t is the average rate of change in the tth period, and v is a preset constant;

[0077] The length of the data interval is determined according to the degree of change of the data of each environmental factor, and different degrees of change correspond to different data interval lengths.

[0078] In this embodiment, different seasons of camphor trees have different effects on their growth, so the degree of change of each environmental factor data is defined based on the seasonal cycle. The degree of change is defined by combining the class average value (greater than the average value) of the coefficient of variation and the average change rate in multiple time periods.

[0079] Step S103, obtaining biological factor data of Cinnamomum camphora, and constructing a Cinnamomum camphora growth rate change curve diagram according to the biological factor data, so as to determine the basic growth rate of Cinnamomum camphora.

[0080] In this embodiment, the biological factor data include tree species characteristic data (genetic characteristics, growth habits), pest and disease data, intraspecific and interspecific competition data, etc. The basic growth rate of camphor tree describes the growth rate of camphor tree under ideal conditions considering only biological factors.

[0081] In some embodiments of the present application, a growth rate change curve of camphor tree is constructed based on biological factor data to determine the basic growth rate of camphor tree, including:

[0082] A camphor tree simulation model under an ideal environment is established, the biological factor data of camphor tree is input into the camphor tree simulation model, and the structural parameters of camphor tree are output, so as to construct a camphor tree growth rate change curve chart, and the average value of the camphor tree growth rate in the camphor tree growth rate change curve chart is taken as the basic growth rate of camphor tree.

[0083] In this embodiment, the camphor tree simulation model simulates ideal environmental conditions, and the environmental parameters remain basically unchanged to determine the basic growth rate of camphor tree.

[0084] Step S104, obtaining the environmental factor data, biological factor data and point cloud data of camphor trees within a period of time, and determining the correspondence between the environmental factor data and the point cloud data of camphor trees based on the biological factor data of camphor trees, training a growth model of camphor trees, and evaluating and predicting the biomass of camphor trees through the growth model of camphor trees.

[0085] In this embodiment, three types of data, namely, environmental factor data of camphor tree, biological factor data of camphor tree, and point cloud data of camphor tree over a period of time, are obtained, the three types of data are integrated into samples, and the corresponding structural parameters, environmental impact coefficients, and basic growth rates of camphor tree are determined, and the growth model of camphor tree is trained through multiple samples.

[0086] In some embodiments of the present application, the corresponding relationship between the environmental factor data of camphor and the point cloud data of camphor is determined based on the biological factor data of camphor, including:

[0087] The basic growth rate of camphor tree is determined by the biological factor data of camphor tree, and the basic growth rate of camphor tree is compared with the basic growth rate interval of standard camphor tree to obtain the growth rate deviation;

[0088] The frontier time scale is determined according to the growth rate deviation, and the starting time node of the environmental factor data of camphor is supplemented according to the size of the frontier time scale, so as to re-determine the starting time node of the environmental factor data of camphor;

[0089] The time nodes of the environmental factor data segment of camphor tree and the point cloud data segment of camphor tree are corresponded as the corresponding relationship between the two.

[0090] In the present embodiment, because the camphor tree basic growth rate is the growth rate under the ideal environment, and the biological factor data fluctuation change is relatively small, if the camphor tree basic growth rate is located within the standard camphor tree basic growth rate interval, then the environment is relatively ideal, and there is no need to supplement the environmental factor data. If the camphor tree basic growth rate is not located within the standard camphor tree basic growth rate interval, then the actual environment and the ideal environment have deviations, and it is necessary to supplement the environmental data, so as to obtain more accurate environmental factor data and the corresponding relationship of point cloud data. Different growth rate deviations correspond to the frontier time scale, and the frontier time scale is a time length, which supplements the environmental factor data in the time length before the start time node.

[0091] It should be noted that, in step S104, the environmental factor data, biological factor data and point cloud data of camphor tree within a period of time are obtained, and the time nodes of the three types of data are exactly the same, so as to ensure that they are within the same period of time. Because the growth of camphor tree may be related to the environmental factor data before this time, the environmental factor data is supplemented forward in time.

[0092] In some embodiments of the present application, training a growth model of Cinnamomum camphora includes:

[0093] Split the data segment of the correspondence between the environmental factor data of camphor and the point cloud data of camphor into multiple parts, and insert the biological factor data of camphor into each part in chronological order to form multiple samples;

[0094] Determine the structural parameters, environmental impact coefficient and basic growth rate of camphor tree according to the corresponding data in each sample, and mark the structural parameters, environmental impact coefficient and basic growth rate of camphor tree on each sample;

[0095] Multiple samples are divided into training sets or test sets respectively, and the growth model of Cinnamomum camphora is trained according to the training sets or test sets.

[0096] In this embodiment, the biological factor data of camphor tree is inserted into each part in chronological order, and the time of the biological factor data is consistent with the time of the point cloud data, and the insertion is performed accordingly. Camphor tree growth rate = environmental impact coefficient * camphor tree basic growth rate, which helps the model to be trained and optimized, and the camphor tree biomass is evaluated and predicted through the camphor tree growth model.

[0097] By applying the above technical scheme, the point cloud data of camphor tree is obtained, the point cloud data is preprocessed, the feature points in the point cloud data are extracted, the point cloud data is converted into image data, the feature points in the image data are extracted, and the information loss degree is determined by comparing the difference between the feature points in the point cloud data and the feature points in the image data, the image data is segmented according to the information loss degree, and the structural parameters of camphor tree are extracted from the segmented image data; the environmental factor data of camphor tree is obtained, the interval length of each environmental factor data is set according to the degree of change of each environmental factor data, and the range of each environmental factor data is disassembled by the interval length to obtain each environmental factor data. Multiple intervals of factor data, different intervals of multiple environmental factor data are combined into a growth environment scenario, and the environmental impact coefficient corresponding to the growth environment scenario is determined; the biological factor data of camphor tree is obtained, and a camphor tree growth rate change curve is constructed based on the biological factor data to determine the basic growth rate of camphor tree; the environmental factor data, the biological factor data and the point cloud data of camphor tree within a period of time are obtained, and the corresponding relationship between the environmental factor data and the point cloud data of camphor tree is determined based on the biological factor data of camphor tree, the growth model of camphor tree is trained, and the biomass of camphor tree is evaluated and predicted through the growth model of camphor tree. The present application segments the image data according to the degree of information loss, thereby ensuring the accuracy of image segmentation, thereby improving the accuracy of feature extraction, and being able to accurately obtain the structural parameters of camphor tree. By defining different growth environment scenarios, the impact of the environment on camphor tree is determined, the growth model of camphor tree is trained, the impact of environmental factors and the impact of biological factors are combined, the evaluation accuracy of camphor tree biomass is improved, the future biomass of camphor tree can be predicted, and the reliability of camphor tree biomass evaluation is ensured.

[0098] Through the description of the above implementation methods, those skilled in the art can clearly understand that the present invention can be implemented by hardware, or by means of software plus a necessary general hardware platform. Based on such an understanding, the technical solution of the present invention can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.), including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each implementation scenario of the present invention.

[0099] In order to further explain the technical idea of ​​the present invention, the technical solution of the present invention is now described in combination with specific application scenarios.

[0100] Correspondingly, the present application also provides a system for evaluating the biomass of camphor trees, such as Figure 2 As shown, including:

[0101] An extraction module is used to obtain point cloud data of camphor tree, pre-process the point cloud data, extract feature points in the point cloud data, convert the point cloud data into image data, extract feature points in the image data, and determine the information loss degree by comparing the difference between the feature points in the point cloud data and the feature points in the image data, segment the image data according to the information loss degree, and extract the structural parameters of camphor tree from the segmented image data;

[0102] A construction module is used to obtain the environmental factor data of camphor, set the interval length of each environmental factor data according to the degree of change of each environmental factor data, disassemble the range of each environmental factor data by the interval length to obtain multiple interval segments of each environmental factor data, combine different interval segments of multiple environmental factor data into a growth environment scenario, and determine the environmental impact coefficient corresponding to the growth environment scenario;

[0103] The simulation module is used to obtain the biological factor data of camphor tree, and to construct a curve chart of the growth rate change of camphor tree according to the biological factor data, so as to determine the basic growth rate of camphor tree;

[0104] The evaluation module is used to obtain the environmental factor data, biological factor data and point cloud data of camphor trees over a period of time, and determine the corresponding relationship between the environmental factor data and the point cloud data of camphor trees based on the biological factor data of camphor trees, train the growth model of camphor trees, and evaluate and predict the biomass of camphor trees through the growth model of camphor trees.

[0105] Those skilled in the art will appreciate that the modules in the system in the implementation scenario can be distributed in the system of the implementation scenario according to the implementation scenario description, or can be changed accordingly and located in one or more systems different from the implementation scenario. The modules in the above implementation scenario can be combined into one module, or can be further split into multiple submodules.

[0106] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for evaluating the biomass of Cinnamomum camphora, characterized in that: include: Acquire point cloud data of camphor tree, pre-process the point cloud data, extract feature points in the point cloud data, convert the point cloud data into image data, extract feature points in the image data, and determine the information loss degree by comparing the difference between the feature points in the point cloud data and the feature points in the image data, segment the image data according to the information loss degree, and extract the structural parameters of camphor tree from the segmented image data; Acquire the environmental factor data of camphor trees, set the interval length of each environmental factor data according to the degree of change of each environmental factor data, disassemble the range of each environmental factor data by the interval length to obtain multiple interval segments of each environmental factor data, combine different interval segments of multiple environmental factor data into a growth environment scenario, and determine the environmental impact coefficient corresponding to the growth environment scenario; Acquire biological factor data of camphor tree, and construct a camphor tree growth rate change curve chart based on the biological factor data to determine the basic growth rate of camphor tree; Acquire the environmental factor data, biological factor data and point cloud data of camphor trees over a period of time, determine the corresponding relationship between the environmental factor data and the point cloud data of camphor trees according to the biological factor data of camphor trees, train the growth model of camphor trees, and evaluate and predict the biomass of camphor trees through the growth model of camphor trees; in, And according to the biological factor data, a curve of the growth rate change of camphor tree is constructed to determine the basic growth rate of camphor tree, including: Establish a camphor tree simulation model under an ideal environment, input the biological factor data of camphor tree into the camphor tree simulation model, output the structural parameters of camphor tree, thereby constructing a camphor tree growth rate change curve diagram, and use the average value of the camphor tree growth rate in the camphor tree growth rate change curve diagram as the camphor tree basic growth rate; And the corresponding relationship between the environmental factor data of camphor tree and the point cloud data of camphor tree is determined according to the biological factor data of camphor tree, including: The basic growth rate of camphor tree is determined by the biological factor data of camphor tree, and the basic growth rate of camphor tree is compared with the basic growth rate interval of standard camphor tree to obtain the growth rate deviation; The frontier time scale is determined according to the growth rate deviation, and the starting time node of the environmental factor data of camphor is supplemented according to the size of the frontier time scale, so as to re-determine the starting time node of the environmental factor data of camphor; The time nodes of the environmental factor data segment of camphor tree and the point cloud data segment of camphor tree are corresponded as the corresponding relationship between the two; Training the growth model of Cinnamomum camphora, including: Split the data segment of the correspondence between the environmental factor data of camphor and the point cloud data of camphor into multiple parts, and insert the biological factor data of camphor into each part in chronological order to form multiple samples; Determine the structural parameters, environmental impact coefficient and basic growth rate of camphor tree according to the corresponding data in each sample, and mark the structural parameters, environmental impact coefficient and basic growth rate of camphor tree on each sample; Multiple samples are divided into training sets or test sets respectively, and the growth model of Cinnamomum camphora is trained according to the training sets or test sets.

2. The method for evaluating the biomass of Cinnamomum camphora according to claim 1, wherein: And the information loss degree is determined by comparing the differences between the feature points in the point cloud data and the feature points in the image data, including: According to the matching relationship between the feature points in the point cloud data and the feature points in the image data, the feature points are divided into two categories: matching feature points and non-matching feature points; According to the types of feature point attributes in the point cloud data and the feature point attributes in the image data, the feature point attributes are divided into common attributes; Normalize the common attributes of the matching feature points in the point cloud data and the image data, and perform difference analysis on the common attributes of the two feature points to obtain the difference of each common attribute, retain the common attributes whose difference exceeds the corresponding difference threshold, and integrate the differences of the retained common attributes to determine the overall difference of the common attributes of the matching feature points; For non-matching feature points in the point cloud data and the image data, the same common attributes of all non-matching feature points are integrated to obtain the comprehensive values ​​of multiple common attributes of the non-matching feature points in the point cloud data and the comprehensive values ​​of multiple common attributes of the non-matching feature points in the image data; The information loss degree is determined based on the overall difference of the common attributes of the matching feature points, the comprehensive value of multiple common attributes of the non-matching feature points in the point cloud data, and the comprehensive value of multiple common attributes of the non-matching feature points in the image data.

3. The method for evaluating the biomass of Cinnamomum camphora according to claim 2, wherein: The information loss degree is determined based on the overall difference of the common attributes of the matching feature points, the comprehensive value of multiple common attributes of the non-matching feature points in the point cloud data, and the comprehensive value of multiple common attributes of the non-matching feature points in the image data, including: ； in, is the information loss degree, is the loss weight of the matching feature points, is the overall difference of the common attributes of the matching feature points, is the loss weight of non-matching feature points, is the number of shared attributes, is the non-matching feature point in the point cloud data The combined value of the common attributes, is the non-matching feature point in the image data The combined value of the common attributes, is the number of non-matching feature points in the point cloud data, is the number of non-matching feature points in the image data, is a preset constant.

4. The method for evaluating the biomass of Cinnamomum camphora according to claim 1, wherein: Segment the image data according to the degree of information loss, including: An initial feature similarity threshold is set by the feature distribution of the image data, the information loss degree is mapped to obtain a feature similarity correction coefficient, and the initial feature similarity threshold is corrected by the feature similarity correction coefficient to obtain a feature similarity threshold; The pixels in the image are integrated into multiple regions based on the feature similarity threshold to complete the segmentation of the image data.

5. The method for evaluating the biomass of Cinnamomum camphora according to claim 1, wherein: The interval length of each environmental factor data is set according to the degree of change of each environmental factor data, including: Determine the seasonal cycle near the area where the camphor tree is located, find the time period with the largest change in environmental factor data within the seasonal cycle, and decompose the seasonal cycle into multiple time periods based on the length of this time period; Calculate the coefficient of variation and average rate of change of environmental factor data in each period, and calculate the degree of change of each environmental factor data within the seasonal cycle through the coefficient of variation and average rate of change in each period; ； in, is the degree of change of environmental factor data, is the number of time periods in this seasonal cycle, For the The coefficient of variation for each period, For the The average rate of change in the period, is a preset constant; The length of the data interval is determined according to the degree of change of the data of each environmental factor, and different degrees of change correspond to different data interval lengths.

6. A Cinnamomum camphora biomass assessment system, characterized in that: For implementing the method according to any one of claims 1 to 5, the system comprises: An extraction module is used to obtain point cloud data of camphor tree, pre-process the point cloud data, extract feature points in the point cloud data, convert the point cloud data into image data, extract feature points in the image data, and determine the information loss degree by comparing the difference between the feature points in the point cloud data and the feature points in the image data, segment the image data according to the information loss degree, and extract the structural parameters of camphor tree from the segmented image data; A construction module is used to obtain the environmental factor data of camphor, set the interval length of each environmental factor data according to the degree of change of each environmental factor data, disassemble the range of each environmental factor data by the interval length to obtain multiple interval segments of each environmental factor data, combine different interval segments of multiple environmental factor data into a growth environment scenario, and determine the environmental impact coefficient corresponding to the growth environment scenario; The simulation module is used to obtain the biological factor data of camphor tree, and to construct a curve chart of the growth rate change of camphor tree according to the biological factor data, so as to determine the basic growth rate of camphor tree; The evaluation module is used to obtain the environmental factor data, biological factor data and point cloud data of camphor trees over a period of time, and determine the corresponding relationship between the environmental factor data and the point cloud data of camphor trees based on the biological factor data of camphor trees, train the growth model of camphor trees, and evaluate and predict the biomass of camphor trees through the growth model of camphor trees.

Images