CN-122024107-A - Method for evaluating forest land quality by unmanned aerial vehicle remote sensing technology

CN122024107ACN 122024107 ACN122024107 ACN 122024107ACN-122024107-A

Abstract

The invention provides a method for evaluating forest site quality by using an unmanned aerial vehicle remote sensing technology, and belongs to the technical field of forestry information. The invention builds a systematic evaluation frame based on unmanned plane laser radar, which comprises the steps of firstly collecting high-precision point cloud data through an unmanned plane, carrying out pretreatment such as denoising, normalization and single wood segmentation, secondly automatically extracting key site factors such as tree height, stand density and gradient, and finally realizing quantitative evaluation of site quality based on a stand growth model and a potential productivity calculation method. Verification of the Guangdong fir pure forest is taken as an example, the method can efficiently and accurately extract the site information, the evaluation precision is remarkably improved, the tree height, the cross section and the accumulation amount determining coefficients of the constructed model respectively reach 0.6812, 0.8219 and 0.8597, and reliable technical support is provided for accurate forestry management.

Inventors

ZHAO XIAODI
LU XIANGFEI
LIU BOYUAN
LIU ZIYANG
CHENG RUI
WANG QINGJIE
Tian Shize
ZHOU JIANHUA

Assignees

中国林业科学研究院林业科技信息研究所

Dates

Publication Date: 20260512
Application Date: 20260210

Claims (12)

1. The method for evaluating the forest floor quality by using the unmanned aerial vehicle remote sensing technology is characterized by comprising the following steps of: S1, data acquisition and three-dimensional information acquisition, namely performing aerial survey on a target forest area by utilizing a remote sensing sensor carried by an unmanned aerial vehicle to acquire observation data representing a three-dimensional space structure of the area; s2, stand information extraction, namely automatically extracting a stand structure parameter set for describing the forest growth condition and a site factor set for describing the environmental characteristics based on the observed data; s3, modeling the site level and the growth rule, namely dividing different site levels based on the influence of the site factor set on the growth of the stand, and respectively establishing mathematical models for describing the growth of the stand according to the different site levels; s4, calculating potential productivity, namely calculating the maximum productivity which can be achieved by the stand under the conditions of each site grade and different stand ages based on the mathematical model describing the stand growth, and taking the maximum productivity as a potential productivity reference under the conditions; And S5, estimating the standing quality, namely acquiring the corresponding potential productivity standard calculated in the step S4 according to the corresponding standing grade and the stand age of the real stand to be estimated, and comparing the real productivity of the real stand with the potential productivity standard to obtain the quantitative evaluation of the standing quality of the real stand.
2. The method for evaluating forest floor quality by using unmanned aerial vehicle remote sensing technology according to claim 1, wherein in the step S1, the remote sensing sensor is a laser radar, and the observation data is laser radar point cloud data.
3. The method for evaluating forest stand quality by using unmanned aerial vehicle remote sensing technology according to claim 1, wherein in the step S2, the forest stand structure parameter set at least comprises one or more of a forest stand average height, a forest stand density index and a forest breaking area, and the extraction process comprises preprocessing, single wood segmentation and parameter calculation of the laser radar point cloud data.
4. The method of claim 3, wherein the preprocessing comprises denoising and normalization, and wherein the single wood segmentation uses a local maximum watershed algorithm based on a canopy height model.
5. The method according to claim 1 or 2, wherein in step S2, the set of ground factors comprises at least five or more of altitude, grade, slope, grade, soil thickness, and canopy density.
6. The method according to claim 1, wherein in step S3, the method of classifying the ground level comprises: s31, grading each factor in the site factor set, and combining to form a plurality of site type units; S32, analyzing the difference of forest stand growth among units of different site types by taking the average height of the forest stand as a response variable; s33, based on the growth difference, clustering the site type units with similar growth rules to form different site grades.
7. The method according to claim 6, characterized in that in step S32, the growth difference is analyzed by fitting a stand average height growth curve based on the age of the stand.
8. The method according to claim 1, characterized in that in step S3, the mathematical model describing the growth of the stand comprises at least a stand breaking area growth model and a stand accumulating growth model.
9. The method according to claim 8, characterized in that the forest cutting area growth model and/or the stand accumulation growth model adopts a whole-forest whole model of the form: Wherein Y represents a forest breaking area G or a forest stand accumulation V; SDI is a stand density index; T is the average age of the stand; b 1i is a parameter related to the ith site sub-level; b 2 ,b 3 ,b 4 is a general model parameter; Is an error term.
10. The method according to claim 1, wherein in the step S4, the calculation method of the potential productivity is that, for a given site grade and stand age, the maximum amount of stand accumulated annual growth is targeted, and optimization solution is performed within a preset stand density value range, so as to obtain the optimal stand density and the corresponding maximum accumulated annual growth amount, and the maximum accumulated annual growth amount is the accumulated potential productivity.
11. The method of claim 1, further comprising the step of outputting the results of the evaluation of step S5 in a visual manner to generate a site quality level distribution map or an evaluation report.
12. A forest resource information processing system, comprising a processor and a memory, the memory storing a computer program which, when executed by the processor, implements the method of forest stand quality assessment using drone telemetry as claimed in any one of claims 1 to 11.

Description

Method for evaluating forest land quality by unmanned aerial vehicle remote sensing technology Technical Field The invention relates to the fields of forestry information technology and remote sensing monitoring, in particular to a method for evaluating forest site quality by using an unmanned aerial vehicle remote sensing technology. Background Global forests play a vital role in coping with climate change, maintaining biodiversity and providing ecosystem services. However, forest resources are facing multiple pressures of agricultural expansion, excessive harvesting, fire and insect pests. Under the background, scientific and accurate forest site quality evaluation is a basis for realizing sustainable forest management. The forest site quality evaluation aims to comprehensively evaluate the influence of site conditions on the growth and potential productivity of forest vegetation, and is a core basis for forestry planning, operation decision and ecological restoration. Traditional forest site quality evaluation mainly depends on field manual investigation. The method has obvious limitations that firstly, the cost is high, the efficiency is low, a large number of professional technicians are needed, a large area or a region with complex topography is difficult to cover, secondly, the data updating period is long, dynamic real-time monitoring cannot be realized, thirdly, the investigation result is easily influenced by subjective experience and operation errors of personnel, and the data consistency and objectivity are difficult to guarantee. In recent years, remote sensing technology, in particular unmanned aerial vehicle remote sensing, brings revolutionary changes to forest resource monitoring. The unmanned aerial vehicle platform has the advantages of flexibility, high efficiency and relatively low cost, and can be used for carrying various sensors (such as RGB (red, green, blue) cameras, multispectral cameras, laser radars and the like) to acquire forest data with high space-time resolution. Unmanned aerial vehicle remote sensing is successfully applied to the fields of forest biomass estimation, carbon reserve estimation, tree species identification, forest stand structural parameter extraction and the like in the existing research. For example, researchers can utilize unmanned plane laser radar data to combine with ground investigation to realize accurate biomass estimation of single wood scale, and other researches can use unmanned plane images and deep learning models to complete high-precision tree classification and drawing of stand scale. Although the unmanned aerial vehicle remote sensing technology has made remarkable progress in a plurality of aspects of forestry, a set of systematic, strong-operability and deep fusion unmanned aerial vehicle remote sensing data and forest growth model stand-by quality evaluation method and framework are not available at present. The existing research focuses on the application of a single technology or a single parameter, and fails to effectively integrate multi-source and multi-dimensional data acquired by an unmanned aerial vehicle into a complete logic chain for standing quality evaluation, and particularly, a research blank exists in the aspects of how to utilize high-precision three-dimensional point cloud data to drive a stand growth model and quantitatively calculate potential productivity according to the high-precision three-dimensional point cloud data. Therefore, the invention aims to fill the blank, constructs a complete technical system from data acquisition, processing and information extraction to model evaluation, and systematically applies remote sensing technologies such as unmanned plane laser radar and the like to forest land quality evaluation so as to improve the automation, precision and intelligence level of evaluation. Disclosure of Invention First, the technical problem to be solved The technical problem to be solved by the invention is to overcome the defects of low efficiency, high cost, strong subjectivity and difficult dynamic update of the traditional forest site quality assessment method and the defects of scattered application and lack of system model support in the site quality assessment of the existing unmanned aerial vehicle remote sensing technology. The forest site quality evaluation method is efficient, accurate and quantifiable. (II) technical scheme In order to solve the technical problems, the invention provides a method for evaluating forest land quality by using unmanned aerial vehicle remote sensing technology, which comprises the following steps: S1, data acquisition and three-dimensional information acquisition, namely performing aerial survey on a target forest area by utilizing a remote sensing sensor carried by an unmanned aerial vehicle to acquire observation data representing a three-dimensional space structure of the area; s2, stand information extraction, namely automatically extracting a stand structure