Forest stand quality evaluation method and quality evaluation system based on enclosure domain

Abstract

The invention provides a forest stand quality evaluation method and system based on an enclosure domain, and relates to the technical field of forest stand quality evaluation. The quality assessment method provided by the invention comprises the following steps: acquiring geographic information of a to-be-assessed forest stand based on a low-altitude technology, and modeling to obtain a global model; randomly selecting a plurality of enclosed domains in the global model, and performing parting based on the low-altitude visuality character combination to obtain a parting combination; investigating investigation values of the evaluation indexes in the enclosed domain under the same parting combination; weighting the survey value based on the typing combination to obtain a typing unit evaluation value; and performing global evaluation on the global model based on the typing unit evaluation value to obtain a forest stand quality evaluation result. According to the method provided by the invention, the heterogeneity characteristics in the forest stand can be fully considered, and the spatial heterogeneity deviation is eliminated, so that the accuracy and convenience of forest stand quality evaluation are effectively improved, and rapid quality evaluation and analysis in a complex forest region are facilitated.

Description

Technical Field

[0001] The present invention relates to the technical field of stand quality assessment, and particularly to a stand quality assessment method and a quality assessment system based on an enclosed domain. Background Art

[0002] Stand quality is the core index for measuring the comprehensive state of forest ecosystems, which includes tree species structure, functional characteristics and health level. From the analysis of tree species structure, it includes key parameters such as tree species composition, forest layer configuration, diameter class distribution and canopy density. From the analysis of functional characteristics, it includes carbon sequestration capacity, soil and water conservation, water source conservation, windbreak and sand fixation, soil ecological restoration, forest health care, landscape recreation, etc. And from the analysis of health level, it includes pest and disease infection rate, natural regeneration ability and stress resistance, etc. In addition, by accurately assessing specific functions in the stand, it is possible to effectively improve the introduction and implementation rate of targeted scientific measures in the process of forest management, and increase the scientific nature and sustainable development ability of stand management. For example, assessing the soil and water conservation ability can effectively identify high-risk areas of soil erosion, thus providing a targeted location for subsequent restoration.

[0003] Currently, the stand quality assessment technology has formed a multi-level system combining ground surveys, remote sensing detection and model analysis. Based on ground surveys and combined with remote sensing technology to invert stand quality information, and then combined with machine learning algorithms to fuse multi-source data in order to improve the accuracy of quality assessment results. In the prior art, when conducting ground surveys, sampling points are often set in the target stand by the random sampling method, sample plots with an area between 0.06 hectares and 1 hectare are set by the standard plot method, total stations, laser rangefinders and other measuring devices are used to locate the sample plots and set the sample plot ranges, and measuring tools such as measuring rods, rulers, and girth tapes are used to measure the physical parameters of trees one by one, and auxiliary indicators such as soil physical and chemical properties are measured by randomly taking points in the sample plot.

[0004] Although the random sampling method can ensure spatial representativeness in terms of statistics, there are still many defects in actual applications. When facing the quality assessment of stands in complex terrain areas, the accessibility of the sample plots obtained by random sampling is poor, and the range random sampling is likely to ignore the heterogeneity characteristics inside the stand. In addition, the position differences (aspect, soil type, stand density) during the layout of the sample plots will lead to inconsistent assessment results, which greatly affects the accuracy and convenience of stand quality assessment. Therefore, it is urgent to provide a solution to improve the above problems. Summary of the Invention

[0005] The purpose of the present invention is to provide a stand quality assessment method and a quality assessment system based on an enclosed domain, which can fully consider the heterogeneity characteristics inside the stand, eliminate the spatial heterogeneity deviation, and thus effectively improve the accuracy and convenience of stand quality assessment, and is conducive to rapid quality assessment and analysis in complex forest areas.

[0006] In a first aspect, a method for evaluating stand quality based on an enclosed area provided by the present invention includes: obtaining geographical information of a stand to be evaluated based on low-altitude technology and modeling it to obtain a global model; randomly selecting a plurality of enclosed areas within the global model and performing classification based on low-altitude visible trait combinations to obtain a classification combination; investigating the survey values of evaluation indicators within the enclosed areas under the same classification combination; obtaining a classified unit evaluation value by weighting the survey values based on the classification combination; performing a global evaluation on the global model based on the classified unit evaluation value to obtain a stand quality evaluation result; the evaluation indicators include canopy density, mixture ratio, volume, foliage biomass, content of plant VOCs in near-surface air, carbon sequestration capacity, water conservation capacity, nitrogen fixation capacity, mixture ratio, and negative oxygen ion concentration.

[0007] Optionally, the low-altitude visible trait combination includes one of deciduousness and evergreenness, fast-growing and non-fast-growing, broad-leaved and coniferous of arbors.

[0008] Optionally, when obtaining the geographical information of the stand to be evaluated based on low-altitude technology, the geographical information includes arbor coordinates, the low-altitude visible trait information of the arbors, and the environmental information of the stand to be evaluated.

[0009] Optionally, the low-altitude technology includes low-altitude unmanned aerial vehicles.

[0010] Optionally, the stand to be evaluated includes artificial mixed forests or natural forests.

[0011] Optionally, when randomly selecting a plurality of enclosed areas within the global model, the number of the enclosed areas is:

[0012] wherein, is the total number of enclosed areas within the global model, is the orthographic projection area of the global model, , are both adjustment coefficients, and is greater than 1.

[0013] Optionally, the enclosed area is formed by enclosing three adjacent arbors that are not all collinear, and there are no other arbors within the enclosed area.

[0014] Optionally, when selecting the enclosed area, two adjacent reference arbors are pre-selected and candidate arbors are marked around the reference arbors. Based on the minimum sum of adjacent interior angle differences within the enclosed area, the enclosing arbors are selected from the candidate arbors to form the enclosed area.

[0015] Optionally, when performing classification based on low-altitude visible trait combinations to obtain a classification combination, the types of the classification combination are from 1 to 4.

[0016] Optionally, when obtaining the typing combinations based on the low-altitude visual trait combinations, the quantity of each of the said typing combinations is greater than or equal to 10%.

[0017] Optionally, when performing typing, it includes manual annotation typing or typing using machine learning.

[0018] Optionally, based on the low-altitude visual trait combinations and according to the evaluation direction, type the enclosed domain.

[0019] Optionally, when investigating the survey values of the evaluation indicators within the enclosed domain under the same typing combination, enter the stand to be evaluated for manual investigation.

[0020] Optionally, when weighting the survey values based on the typing combinations, it includes:[[]]

[0021] wherein,[[]] is the evaluation value of the typing unit of the enclosed domain under the th typing combination,[[]] is the total quantity of the enclosed domain under the th typing combination,[[]] is the survey value of the evaluation indicator of the th enclosed domain under the th typing combination,[[]] is the orthographic projection area of the th enclosed domain under the th typing combination.

[0022] Optionally, when performing global evaluation on the global model based on the evaluation value of the typing unit, it includes:[[]]

[0023] wherein,[[]] is the total forest stand quality evaluation quantity of the evaluation indicator of the stand to be evaluated,[[]] is the evaluation value of the typing unit of the enclosed domain under the th typing combination,[[]] is the total quantity of the enclosed domain under the th typing combination,[[]] is the quantity of the typing combinations,[[]] is the total quantity of the enclosed domains within the global model,[[]] is the total area of the stand to be evaluated.

[0024] Optionally, when typing the enclosed domain based on the deciduousness and evergreenness of the arbors, the typing combinations include: three deciduous trees; two deciduous trees and one evergreen tree; one deciduous tree and two evergreen trees; three evergreen trees.

[0025] Optionally, when classifying the enclosed area based on the fast-growing and non-fast-growing characteristics of arbors, the classification combinations include: three fast-growing trees; two fast-growing trees and one non-fast-growing tree; one fast-growing tree and two non-fast-growing trees; three non-fast-growing trees.

[0026] Optionally, when classifying the enclosed area based on the broad-leaved and coniferous characteristics of arbors, the classification combinations include: three broad-leaved trees; two broad-leaved trees and one coniferous tree; one broad-leaved tree and two coniferous trees; three coniferous trees.

[0027] In a second aspect, the present invention also provides a stand quality assessment system for implementing any of the above optional quality assessment methods, including: A map construction module that models the geographical information of the stand to be evaluated based on low-altitude technology to obtain a global model; A selection and classification module that randomly selects multiple enclosed areas within the global model and classifies them based on low-altitude visible trait combinations to obtain classification combinations; An index investigation module that investigates the measured values of the evaluation indexes within the enclosed area under the same classification combination, and the evaluation indexes include canopy density, mingling degree, volume, foliage biomass, content of plant VOCs in the near-surface air, carbon sink amount, water conservation amount, nitrogen fixation amount, mingling ratio, negative oxygen ion concentration; A weighted analysis module that conducts weighted scoring on the measured values based on the classification combination to obtain the evaluation value of the classification unit; A global evaluation module that conducts a global evaluation on the global model based on the evaluation value of the classification unit to obtain the stand quality assessment result. Description of the Drawings

[0028] Figure 1 It is a flowchart of a stand quality assessment method based on an enclosed area provided by the present invention; Figure 2 It is a schematic diagram of the global model constructed and obtained by the present invention in step S1; Figure 3 It is a schematic diagram of generating all enclosed areas within the global model by the present invention in step S2; Figure 4 It is a schematic diagram of the structure of a stand quality assessment system provided by the present invention. Detailed Embodiments

[0029] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be described clearly and completely below. Apparently, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention. Unless otherwise defined, the technical terms or scientific terms used herein shall have the ordinary meanings understood by those of ordinary skill in the art belonging to the field of the present invention.

[0030] Referring to Figure 1 , the present invention provides a stand quality assessment method based on an enclosed domain, including the following steps: S1. Obtain a global model by building a geographical information model of the stand to be evaluated based on low-altitude technology; S2. Randomly select multiple enclosed domains within the global model and perform typing based on low-altitude visible trait combinations to obtain a typing combination; S3. Investigate the survey values of the evaluation indicators within the enclosed domain under the same typing combination; S4. Based on the typing combination, perform weighted scoring on the survey values to obtain a typing unit evaluation value; S5. Based on the typing unit evaluation value, perform a global evaluation on the global model to obtain a stand quality assessment result.

[0031] In fact, the stand quality assessment method provided by the present invention, by combining low-altitude technology modeling and trait typing sampling, can not only improve the safety and comprehensiveness of information collection, but also fully consider the heterogeneity characteristics within the stand, correct the position differences, and thus is beneficial to improving the accuracy and convenience of quality assessment. In addition, when using the enclosed domain for evaluation and investigation, the interspecific relationships of the stand and the associated factors between the stand and the environment can be fully considered, which can further improve the investigation accuracy and the relevance of the evaluation and investigation results to environmental factors.

[0032] In some embodiments, the low-altitude technology used in step S1 includes low-altitude unmanned aerial vehicles. In fact, by using low-altitude technology to map the stand to be evaluated, the geographical information of the stand to be evaluated can be obtained, and by processing the geographical information, the global model corresponding to the stand to be evaluated can be obtained, which is beneficial to randomly selecting enclosed domains on the computer side.

[0033] Furthermore, the geographical information obtained by using low-altitude technology in step S1 includes the terrain undulation changes, vegetation coverage, environmental characteristics (such as slopes, water flows, etc.) in the stand to be evaluated, as well as the low-altitude visualization traits and positions of the arbors in the stand to be evaluated. In fact, by using low-altitude technology for mapping, the accuracy and convenience of data acquisition can be improved, and the comprehensiveness of information acquisition can be effectively improved from a low-altitude perspective.

[0034] Specifically, after obtaining the geographical information of the stand to be evaluated based on low-altitude technology in step S1, a global model can be obtained by using the commonly used map construction method in the art for modeling. In fact, when modeling to obtain the global model, the terrain changes in the stand to be evaluated can be ignored, and the focus can be on the position information of the arbors in the stand to be evaluated to construct a simple model, which is beneficial to reducing the method complexity and improving the modeling efficiency.

[0035] In some embodiments, the simple global model obtained by modeling after obtaining the geographical information based on low-altitude technology in step S1 can be Figure 2 the schematic diagram shown in Figure 2 where the circles represent an arbor in the stand to be evaluated, and the size of the circles can intuitively represent the projected area size of each arbor from the top to the bottom. In addition, the mutual position relationship between the arbors in the stand to be evaluated can be intuitively expressed through the global model.

[0036] In some embodiments, the stand to be evaluated in step S1 is an artificial mixed forest or a natural forest. In fact, by conducting a stand quality survey on the artificial mixed forest, the operation status of the artificial mixed forest can be effectively known, which is beneficial to adjusting the mixing strategy during artificial mixing. In addition, when conducting a stand survey on the natural forest, the stand quality of the stand in a balanced state under natural conditions can be known, which can provide a basis and direction for stand management.

[0037] In fact, when randomly selecting multiple enclosing domains within the global model in step S2, the formed enclosing domains are formed by enclosing three adjacent arbors that are not all collinear, and there are no other arbors within the enclosing domains. Specifically, the enclosing domains are triangular. By randomly selecting enclosing domains within the global model for sampling survey, the evaluation accuracy can be effectively improved. At the same time, according to the classification results, the visual traits of the enclosing domains can be associated with the evaluation results, effectively improving the relevance between the results and the stand information.

[0038] In some embodiments, when randomly selecting multiple enclosing domains through the global model in step S2, all the arbors can be associated in the global model in advance to form enclosing domains, and the required number of enclosing domains can be randomly selected from all the enclosing domains for sampling survey, as Figure 3 shown. In fact, generating all enclosing domains is beneficial to improving the comprehensiveness during sampling.

[0039] In some other embodiments, when selecting an enclosed area within the global model in step S2, two adjacent reference arbors can be randomly selected in the global model in advance, and the candidate arbors can be marked around the reference arbors. Based on the minimum sum of the adjacent interior angle differences within the enclosed area, the enclosing arbors are selected from the candidate arbors to form the enclosed area. In fact, forming the enclosed area based on the minimum sum of the adjacent interior angle differences can preferentially obtain an enclosed area with three acute angles, which can better reflect the interaction relationship between the arbors.

[0040] In some embodiments, when randomly selecting an enclosed area within the global model in step S2, the number of enclosed areas is positively correlated with the total area of the stand to be evaluated. In fact, the more enclosed areas are set within the global model, the higher the theoretical accuracy, but the accompanying survey workload will also increase exponentially. Therefore, the number of enclosed areas within the global model can be expressed as:

[0041] Wherein, is the total number of enclosed areas within the global model, is the orthographic projection area of the global model (which can be directly measured by low-altitude technology in step S1), , are both adjustment coefficients, and is greater than 1.

[0042] In some embodiments, when classifying based on the low-altitude visible trait combination in step S2, the low-altitude visible trait combination is a combination of arbor traits that can be directly observed with the naked eye. Specifically, it can be directly obtained by low-altitude technology in step S1 and directly labeled through machine learning combined with a neural network model. More specifically, the low-altitude visible trait combination can be expressed as A and a (A and a are a pair of relative traits), and further can include one of the deciduousness and evergreenness, fast-growing and non-fast-growing, broad-leaved and needle-leaved of the arbor.

[0043] In a further embodiment, after classifying based on the low-altitude visible trait combination in step S2, the number of types of classification combinations is from 1 to 4, specifically can be "AAA", "AAa", "Aaa", and "aaa". Further, when classifying a combination within the enclosed area based on deciduousness and evergreenness, fast-growing and non-fast-growing, broad-leaved and needle-leaved, the types of enclosed areas are shown in Table 1 below. When classifying based on these two combinations of deciduousness and evergreenness, broad-leaved and needle-leaved, the types of arbors are shown in Table 2, and the types of enclosed areas are shown in Table 3 below.

[0044] Table 1 Types of Enclosed Areas When Classifying Based on One Low-Altitude Visible Trait Combination

[0045]

[0046] Table 2 Arbor types when classified based on deciduousness and evergreenness, broad-leavedness and needle-leavedness

[0047]

[0048] Table 3 Types of enclosing domains when classified based on the arbor types in Table 2

[0049]

[0050] In some embodiments, when obtaining the classification combinations based on the low-altitude visible traits in step S2, the quantity of each classification combination is greater than or equal to 10% of the total quantity. In fact, by controlling the lower limit value of the quantity of enclosing domain types, the mutual relationships of arbors under various relative traits in a mixed forest (artificial mixed forest or natural forest) can be evaluated more comprehensively, the contribution degree of the mutual relationships between arbors to the evaluation result can be quantified, and the problem of evaluation bias can be improved.

[0051] In some embodiments, after selecting the enclosing domains in the global model and classifying them, if the quantity of the enclosing domains of one or several types does not reach the preset standard, it is necessary to re-select the enclosing domains in the global model. Further, when it is difficult to reach the preset standard after selecting the enclosing domains in the global model multiple times (the upper limit is 5 to 10 times), the classification result of the enclosing domains of the last time is used for investigation.

[0052] In some embodiments, when classifying based on the low-altitude visible trait combinations in step S2, a machine learning combined with a neural network model can be used to classify the combinations in the selected enclosing domains. Specifically, a binary classification method can be adopted. For example, if a certain trait is represented as "1" and the corresponding shape is represented as "0", then the types of enclosing domains can be represented as "111", "110", "100", "000".

[0053] In some embodiments, in step S2, the enclosing domains can be classified based on the low-altitude visible trait combinations and according to the evaluation direction. In fact, the evaluation direction is related to the evaluation index to determine which type of index should be concerned when evaluating the stand quality. When based on the trait combination of arbor deciduousness and evergreenness, the corresponding index investigation of the enclosing domains can be carried out according to different evaluation directions.

[0054] In some embodiments, steps S1 and S2 can be combined. After obtaining the geographical information and modeling using the low-altitude technology, directly select the enclosing domains in the global model and classify the traits, so as to obtain the classification result of the enclosing domains, and send the coordinates of the selected enclosing domains to forestry personnel for forestry personnel to enter the forest to be evaluated to conduct artificial forestry investigations on the area enclosed by the enclosing domains.

[0055] In fact, when investigating the evaluation indicators within the enclosed domain under the same classification combination in step S3, common investigation methods in the art can be adopted, so as to obtain the investigation values of different enclosed domains under the same evaluation indicator under the same classification combination. Specifically, by pre-sampling to obtain the enclosed domain and investigating the selected enclosed domain, the unevenness of sampling in traditional investigations can be improved. In addition, when investigating the enclosed domains under the same classification combination, the internal indicators have a certain degree of convergence, which can effectively reduce the number of sample plots.

[0056] In some embodiments, the evaluation indicators in step S3 include canopy density, mixture degree, volume, foliage biomass, content of plant VOCs in near-surface air, carbon sink amount, water conservation amount, nitrogen fixation amount, mixture ratio, and negative oxygen ion concentration. By investigating the relevant evaluation indicators within the enclosed domain, it can then be extended to the entire stand to be evaluated, effectively improving the accuracy and efficiency of quality investigation for large-area or complex stands.

[0057] Specifically, after weighting the investigation values based on the classification combination in step S4, the investigation values of multiple enclosed domains with different shapes under the same classification combination can be integrated, thereby obtaining the evaluation value of the classification unit of the enclosed domain under this classification combination. In fact, even for the enclosed domains of the same classification combination, due to factors such as stand growth, natural elimination, or artificial elimination, the shapes and enclosed areas of the enclosed domains are not the same. Weighting and integrating multiple enclosed domains can effectively improve the accuracy.

[0058] In some embodiments, when weighting the investigation values based on the classification combination in step S4, it can be expressed as:

[0059] Where, is the evaluation value of the classification unit of the enclosed domain under the th classification combination, is the total number of enclosed domains under the th classification combination, is the investigation value of the evaluation indicator of the th enclosed domain under the th classification combination, is the projected area of the th enclosed domain under the th classification combination.

[0060] In fact, in step S5, based on the evaluation value of the classification unit, the global model can be globally evaluated to obtain the stand quality evaluation result. Specifically, it can be expressed as:

[0061] Where, The total stand quality assessment, which is an assessment index for the stand to be evaluated is the assessment value of the classification unit of the enclosure domain under the th classification combination, is the number of classification combinations, is the total number of enclosure domains in the global model, is the total area of the stand to be evaluated.

[0062] See Figure 4 , the present invention provides a stand quality assessment system for implementing any of the above optional quality assessment methods, including: A map construction module 100 that models the geographical information of the stand to be evaluated based on low-altitude technology to obtain a global model; A selection and classification module 200 that randomly selects multiple enclosure domains in the global model and classifies them based on low-altitude visual trait combinations to obtain classification combinations; An index investigation module 300 that investigates the investigation values of the evaluation indexes within the enclosure domains under the same classification combination; wherein, the evaluation indexes include canopy density, mixing degree, stock volume, foliage biomass, content of plant VOCs in near-surface air, carbon sequestration amount, water conservation amount, nitrogen fixation amount, mixing ratio, negative oxygen ion concentration; A weighted analysis module 400 that performs weighted scoring on the investigation values based on the classification combinations to obtain the evaluation values of the classification units; A global evaluation module 500 that globally evaluates the global model based on the evaluation values of the classification units to obtain the stand quality assessment result.

[0063] Although the embodiments of the present invention have been described in detail above, it is obvious to those skilled in the art that various modifications and changes can be made to these embodiments. However, it should be understood that such modifications and changes are all within the scope and spirit of the present invention described in the claims. Moreover, the present invention described herein may have other embodiments and can be implemented or realized in various ways.

Claims

1. A forest stand quality assessment method based on enclosure domain, characterized in that: include: The global model is obtained by acquiring the geographic information of the forest stands to be evaluated based on low-altitude technology; Randomly select multiple enclosed domains in the global model and perform typing based on the combination of low-altitude visible traits to obtain typing combinations; investigate the survey values ​​of evaluation indicators in the enclosed domain under the same typing combination; and weight the survey values ​​based on the typing combination to obtain typing unit evaluation values; The forest stand quality assessment result is obtained by conducting a global assessment of the global model based on the classification unit assessment value; the assessment indicators include canopy density, intermixing degree, accumulation, branch and leaf biomass, VOCs content of near-surface air plants, carbon sink, water conservation, nitrogen fixation, intermixing ratio, and negative oxygen ion concentration.

2. The quality assessment method according to claim 1, characterized in that: The combination of low-altitude visible traits includes one of the deciduous and evergreen properties, fast-growing and non-fast-growing properties, broad-leaved and coniferous properties of trees; and / or, when obtaining the geographic information of the forest stand to be evaluated based on low-altitude technology, the geographic information includes the coordinates of the trees, the low-altitude visible trait information of the trees, and the environmental information of the forest stand to be evaluated; and / or, the low-altitude technology includes low-altitude drones; and / or, the forest stand to be evaluated includes an artificial mixed forest or a natural forest.

3. The quality assessment method according to claim 1, characterized in that: When multiple enclosed domains are randomly selected in the global model, the number of the enclosed domains is: ； in, is the total number of enclosed domains in the global model, is the orthographic projection area of ​​the global model, , are adjustment factors, and Greater than 1; And / or, the enclosed domain is formed by three adjacent and non-collinear trees, and there are no other trees in the enclosed domain; and / or, when selecting the enclosed domain, two adjacent reference trees are pre-selected and trees to be selected are calibrated around the reference trees, and based on the minimum sum of adjacent internal angle differences in the enclosed domain, enclosed trees are selected from the trees to be selected to form the enclosed domain.

4. The quality assessment method according to claim 1, characterized in that: When typing is performed based on the combination of low-altitude visible traits to obtain typing combinations, the types of the typing combinations are 1 to 4; and / or, when typing is performed based on the combination of low-altitude visible traits to obtain typing combinations, the number of each typing combination is greater than or equal to 10%; and / or, typing includes manual annotation typing or typing using machine learning; and / or, typing the enclosed domain based on the low-altitude visible trait combination and according to the evaluation direction.

5. The quality assessment method according to claim 1, characterized in that: When investigating the survey values ​​of the evaluation indicators in the enclosed area under the same classification combination, enter the forest stand to be evaluated to conduct manual investigation; and / or, when weighting survey values ​​based on a combination of subtypes, including: ； in, For the The evaluation value of the typing unit of the enclosed domain under the typing combination, For the The total number of enclosed domains under the classification combination, For the The following classification combinations The survey value of the evaluation index of the enclosed domain, For the The following classification combinations The orthographic projection area of ​​the enclosed domain.

6. The quality assessment method according to claim 1, characterized in that: When conducting a global assessment of the global model based on the classification unit assessment value, it includes: ； in, is the total amount of stand quality assessment of the assessment indicators of the stand to be assessed, For the The evaluation value of the typing unit of the enclosed domain under the typing combination, For the The total number of enclosed domains under the classification combination, is the number of typing combinations, is the total number of enclosed domains in the global model, is the total area of ​​the forest stand to be evaluated.

7. The quality assessment method according to claim 1, characterized in that: When the enclosed domain is classified based on the deciduousness and evergreenness of the trees, the classification combinations include: three deciduous trees; two deciduous trees and one evergreen tree; one deciduous tree and two evergreen trees; three evergreen trees; and / or, when the enclosed domain is classified based on the fast-growing and non-fast-growing nature of the trees, the classification combinations include: three fast-growing trees; two fast-growing trees and one non-fast-growing tree; one fast-growing tree and two non-fast-growing trees; three non-fast-growing trees; and / or, when the enclosed domain is classified based on the broad-leaved and coniferous nature of the trees, the classification combinations include: three broad-leaved trees; two broad-leaved trees and one coniferous tree; one broad-leaved tree and two coniferous trees; three coniferous trees.

8. A forest stand quality assessment system for implementing the quality assessment method according to any one of claims 1 to 7, characterized in that: include: The map construction module uses low-altitude technology to obtain the geographic information of the forest stands to be evaluated and build a global model; Select a typing module, randomly select multiple enclosed domains in the global model, and perform typing based on the combination of low-altitude visible traits to obtain a typing combination; The index survey module investigates the survey values ​​of the evaluation indicators in the enclosed area under the same classification combination, and the evaluation indicators include canopy density, mixed degree, accumulation, branch and leaf biomass, VOCs content of near-surface air plants, carbon sink, water conservation, nitrogen fixation, mixed ratio, and negative oxygen ion concentration; Weighted analysis module, which weights the survey values ​​based on the classification combination to obtain the classification unit evaluation value; The global assessment module conducts a global assessment of the global model based on the classification unit assessment values ​​to obtain the forest stand quality assessment results.

Images