Method for measuring and monitoring carbon sink of single-plantation based on machine vision recognition

Abstract

The application discloses a single-plantation carbon sink measurement and monitoring method based on machine vision recognition and belongs to the technical field of geographic spatial data processing. The method comprises the following steps: S10, spatial distribution data of a preset monitoring sample plot is acquired, wherein the spatial distribution data comprises forest types and forest distribution; and monitoring deployment is carried out on the preset monitoring sample plot, wherein the monitoring deployment mode comprises arranging a plurality of observation cameras on the preset monitoring sample plot; and S20, video picture data of the forest is acquired based on the arranged observation cameras to acquire growth data information of the forest, wherein the growth data information comprises forest height, tree diameter and coverage. Through the machine vision recognition technology, the method realizes effective monitoring of the growth data of the forest, improves the accuracy and efficiency of carbon sink measurement, and can timely find the change rule of the growth of the forest by recording and analyzing the growth data information in different periods, thereby providing a scientific basis for dynamic monitoring of the carbon sink.

Description

Technical Field

[0001] This invention relates to the field of geospatial data processing technology, and in particular to a method for measuring and monitoring carbon sequestration in single-tree afforestation based on machine vision recognition. Background Technology

[0002] Carbon sequestration afforestation refers to afforestation activities with specific requirements carried out on land with a defined baseline, with the primary purpose of increasing carbon sequestration, and involving the measurement and monitoring of carbon sequestration during the afforestation and stand (tree) growth process.

[0003] Carbon sequestration measurement and monitoring are key technical aspects of carbon sequestration afforestation. This requires accurate measurement of carbon storage before and after afforestation, as well as continuous monitoring of carbon uptake during forest growth. Commonly used measurement methods include plot surveys, remote sensing technology, and biomass models.

[0004] To promote the development of carbon sequestration measurement and monitoring technologies, application document CN202310203806.X provides a dynamic carbon sequestration measurement method for carbon sequestration in afforestation projects. This technical solution includes the design and layout of IoT monitoring plots, automatic data collection from these plots, setting the data collection frequency, and completing the analysis and storage of the collected data. It also includes monitoring the carbon storage of afforested forests, including sequentially calculating the carbon storage in the monitoring area, calculating and verifying the accuracy of the carbon storage in the monitoring area, and deducting and correcting for the accuracy of the carbon storage in the monitoring area. This technical solution, by deploying IoT monitoring plots and automatically acquiring monitoring data at fixed periods to calculate forest carbon storage, not only reduces monitoring costs and survey errors but also improves the timeliness of carbon sequestration measurement and monitoring.

[0005] However, in reality, forestry projects have long growth cycles and are significantly affected by natural factors. Climate change, such as droughts, floods, and extreme weather events, can severely damage forest growth, thereby affecting carbon sequestration. For example, prolonged droughts can slow tree growth or even kill trees, and forest fires can instantly release large amounts of stored carbon, causing the loss of carbon sequestration function. This necessitates complex models and extensive data support for carbon sequestration measurement and monitoring. The aforementioned technical solutions, when conducting per-tree surveys, largely rely on acquired tree diameter data and reuse it through calculation software to reduce the probability of data recording and calculation errors during carbon sequestration measurement. They do not consider trees falling due to various reasons, resulting in a poor data foundation and difficulty in extracting key measurement parameters such as tree height, diameter, and cover. When existing monitoring technologies establish a sufficient number of sample plots in large-scale forests for on-site monitoring, errors can occur during the subsequent verification of measurement data, making it difficult to ensure the authenticity and reliability of the monitoring data. Summary of the Invention

[0006] In view of the problems existing in the field of geospatial data processing technology, the present invention is proposed.

[0007] Therefore, one of the objectives of this invention is to provide a method for measuring and monitoring carbon sequestration in single-tree afforestation based on machine vision recognition. Through machine vision recognition technology, it achieves effective monitoring of tree growth data, improves the accuracy and efficiency of carbon sequestration measurement, and by recording and analyzing growth data information in stages, it can promptly discover the changing patterns of tree growth, providing a scientific basis for the dynamic monitoring of carbon sequestration.

[0008] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0009] This invention provides a method for measuring and monitoring carbon sequestration in single-tree afforestation based on machine vision recognition, comprising the following steps:

[0010] Step S10: Obtain spatial distribution data of the preset monitoring plots, the spatial distribution data including forest type and forest distribution; and deploy monitoring in the preset monitoring plots, the monitoring deployment method including setting up a number of observation cameras in the preset monitoring plots;

[0011] Step S20: Based on the deployed observation cameras, acquire video and image data of the trees to obtain the growth data information of the trees, including tree height, tree diameter and coverage.

[0012] Step S30: Record growth data information in cycles, and calculate carbon sequestration based on the recorded growth data information. At the same time, calculate the carbon sequestration of a single plant in the preset monitoring plot based on the calculation results. The cycled recording of growth data information includes recording in cycles of 1 to 2 years.

[0013] Step S40: Analyze the regular changes in the growth data information in each cycle, calculate the differences in tree height, tree diameter and coverage based on the regular changes; at the same time, calculate the difference in carbon sink based on the differences to obtain the regular influence of the differences on the difference, and generate a database.

[0014] Step S50: During the cycle alternation, the coverage of the preset monitoring plots is divided based on the previous cycle; wherein, the division method includes dividing according to the coverage of arbor forests, shrub forests and bamboo forests, analyzing the changes in growth data information of different coverage, and the impact of the changes in growth data information on the carbon sink.

[0015] Step S60: Count the number of trees with different coverage, and based on the number, divide the tree height and diameter in the growth data information into intervals to obtain the regular changes in carbon sink of trees in different intervals, upload the regular changes to the database, and update the database.

[0016] In a preferred embodiment of the present invention, step S60 includes adjusting the tree height and diameter according to θ. 1 区间 θ 2 区间 and θ n 区间 The intervals are differentiated, where n represents the nth interval; and the difference in carbon sequestration between different intervals in the previous period is calculated using the following formula:

[0017] Where, δ i This represents the δ-th carbon sink acquired within the i-th cycle;

[0018] In the formula, C tree This indicates the total weight of trees in different sections, expressed in tons; △ tree t1 and t2 represent the average tree height and diameter of trees in different intervals; t1 and t2 represent the start and end times of the calculation.

[0019] In a preferred embodiment of the present invention: based on the calculation results, the corresponding period is divided into at least four measurement periods to obtain the highest carbon sink generated in different measurement periods; based on the highest carbon sink, and taking the atmospheric fluctuations of the preset monitoring site as a basis, the atmospheric fluctuations include wind speed; the eddy covariance method is used to calculate the regularity characteristics of the highest carbon sink fluctuations within the same measurement period and under the same interval, and the result is calculated according to the following formula:

[0020] Wherein, λ represents the time of the highest carbon sink fluctuation.

[0021] In the formula, w o This represents the w-th carbon sink amount collected from trees in the same interval during the o-th collection; q n represents the q-th carbon sink fluctuation value collected in the nth same interval of trees; k represents the wind speed fluctuation collected in the vertical direction of the same interval of trees.

[0022] In a preferred embodiment of the present invention, under the same interval within different measurement periods, at least 10 historically highest carbon sequestration values ​​are sorted by size. The changes in tree height and diameter are analyzed based on the sorted carbon sequestration data, and data is collected. Based on the collected data, the difference in tree height and diameter data within the same interval is calculated, and the difference is marked as a reference difference. If the reference difference shows an increasing trend and the carbon sequestration increases within the corresponding measurement period, the calculated carbon sequestration value is determined to be an anomaly; otherwise, no determination is made.

[0023] In a preferred embodiment of the present invention: when the carbon sink calculation is determined to be an outlier, a set of carbon sink data is collected to generate a dataset. Within the corresponding interval, the forest cover area within the interval is divided into high cover area, medium cover area, and low cover area according to the calculated reference data difference. The carbon sink of the trees with the largest tree height and diameter in the first third of each cover area is calculated separately, and the calculated value is marked as a reference value. If the carbon sink calculated value of the trees in a future period exceeds the reference value, the carbon sink calculation is determined to be a normal value; otherwise, no determination is made.

[0024] In a preferred embodiment of the present invention, the following steps are taken: among the trees with the largest height and diameter in the first third of the forest, the proportion of the trees to all trees in the corresponding coverage area is calculated, and the number of trees is obtained based on the proportion. The trees are then sorted by number, and a critical threshold is preset based on the carbon sequestration difference. If the carbon sequestration difference calculated based on the number of trees in a future period exceeds the critical threshold, the carbon sequestration calculation value is determined to be an abnormal value; otherwise, no abnormal value is determined.

[0025] In a preferred embodiment of the present invention, the carbon sequestration amount when it exceeds the critical threshold is recorded and stored; the average carbon sequestration amount of the trees is calculated based on the carbon sequestration amount; and the carbon sequestration amount of a single tree in the number of trees is calculated based on the average carbon sequestration amount. If the carbon sequestration amount of a certain tree is lower than the average carbon sequestration amount, the carbon sequestration amount generated by the tree is marked as a key monitoring carbon sequestration amount.

[0026] In a preferred embodiment of the present invention, the following steps are taken: a preset number of monitoring sessions is set based on the changes in carbon sink generated by the trees, and the carbon sink corresponding to each monitoring session is recorded. The trend of the carbon sink change is analyzed, and an ROC curve is plotted. At the same time, the preset number of monitoring sessions is given as a monitoring period. If the change trend shows a downward trend within the monitoring period, the trees are determined to be in an abnormal growth state and are removed from the monitoring objects.

[0027] A terminal includes a processor, an input interface, an output interface, and a memory, which are interconnected. The memory stores a computer program, which includes program instructions. The processor is configured to invoke the program instructions to execute a method for measuring and monitoring carbon sequestration in single-tree afforestation based on machine vision recognition.

[0028] A computer-readable storage medium storing a computer program, the computer program including program instructions that, when executed by a processor, cause the processor to perform a method for measuring and monitoring carbon sequestration in single-tree afforestation based on machine vision recognition.

[0029] 1. By setting up observation cameras in pre-set monitoring plots and using machine vision technology to acquire forest growth data, and recording the growth data at a given period, the changing patterns of forest growth can be captured in a timely manner, providing basic data for the accurate calculation of carbon sequestration.

[0030] 2. Based on the recorded growth data, the carbon sink of a single tree in the pre-set monitoring plots can be calculated, which can accurately assess the carbon sink contribution of each tree. At the same time, by analyzing the regular changes in growth data, the differences in tree height, diameter and coverage can be calculated, and the difference in carbon sink can be calculated based on these differences to generate a database and provide data support for subsequent research.

[0031] 3. During the periodic alternation, the coverage of the pre-set monitoring plots is divided, and the changes in growth data information under different coverage and their impact on carbon sink are analyzed, which helps to achieve classified management and precise policy implementation; by statistically analyzing the number of trees under different coverage and classifying tree height and diameter into intervals, the regular changes in carbon sink of trees in different intervals can be obtained, and management measures can be further refined.

[0032] 4. By analyzing the changing trends of the calculated carbon sink value, and combining indicators such as the difference in reference data and the historical highest carbon sink, outliers can be identified and eliminated in a timely manner, thereby improving the accuracy and reliability of carbon sink calculation. Attached Figure Description

[0033] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. Wherein:

[0034] Figure 1 This is a schematic diagram of the method flow according to an embodiment of the present invention;

[0035] Figure 2 This is a schematic diagram of the process structure of an embodiment of the present invention. Detailed Implementation

[0036] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the described embodiments of the present invention are within the scope of protection of the present invention.

[0037] Due to the shortcomings of existing technologies, the data foundation obtained in carbon sequestration calculation is poor, making it difficult to extract key measurement parameters such as tree height, tree diameter, and cover. When existing monitoring technologies set up a sufficient number of sample plots in large forests for on-site monitoring, errors may occur in the later verification process of measurement data, making it difficult to ensure the authenticity and reliability of monitoring data.

[0038] Based on this, the present invention proposes a method for carbon sequestration measurement and monitoring of single-tree afforestation based on machine vision recognition. Through machine vision recognition technology, it realizes effective monitoring of forest growth data, improves the accuracy and efficiency of carbon sequestration measurement, and by recording and analyzing growth data information in stages, it can promptly discover the changing patterns of forest growth, providing a scientific basis for dynamic monitoring of carbon sequestration.

[0039] The present solution will be further described in detail below through embodiments and in conjunction with the accompanying drawings.

[0040] Reference Figures 1 to 2 This is one embodiment of the present invention, which provides a method for measuring and monitoring carbon sequestration in single-tree afforestation based on machine vision recognition, including the following steps:

[0041] Step S10: Obtain spatial distribution data of the preset monitoring plots, including tree type and tree distribution; and deploy monitoring in the preset monitoring plots, including setting up several observation cameras in the preset monitoring plots, with a minimum of 6 cameras.

[0042] In this embodiment, the specific number of observation cameras to be deployed can be reasonably determined based on the area of ​​the preset monitoring plot and / or the actual number of trees planted. Through this machine vision technology, comprehensive monitoring of the trees can be achieved.

[0043] Step S20: Obtain video and image data of the trees based on the deployed observation cameras to obtain the growth data information of the trees, including tree height, tree diameter and coverage.

[0044] Step S30: Record growth data information in cycles, and calculate carbon sequestration based on the recorded growth data information. At the same time, calculate the carbon sequestration of a single plant in the preset monitoring plots based on the calculation results. Record growth data information in cycles, including recording in cycles of 1 to 2 years.

[0045] In this embodiment, the selection of the carbon sequestration measurement period needs to comprehensively consider multiple factors, including project type, tree species growth characteristics, monitoring costs, and data accuracy. For example, in afforestation projects, fast-growing tree species (such as eucalyptus and poplar) can reach a certain growth stage within 5 to 7 years, and the carbon sequestration changes more significantly, while slow-growing tree species (such as pine and fir) may need 10 to 15 years to achieve a good carbon sequestration effect. Therefore, the specific time period for recording in this embodiment depends on the actual situation.

[0046] Step S40: Analyze the regular changes in growth data information in each cycle, calculate the differences in tree height, tree diameter and coverage based on the regular changes; at the same time, calculate the difference in carbon sink based on the differences to obtain the regular influence of the differences on the difference, and generate a database.

[0047] In this embodiment, this step helps to detect abnormalities in the growth process of trees, providing a basis for subsequent monitoring and management;

[0048] Step S50: During the cycle alternation, the coverage of the preset monitoring plots is divided based on the previous cycle; the division method includes dividing according to the coverage of arbor forest, shrub forest and bamboo forest, and analyzing the changes in growth data information of different coverage, as well as the impact of changes in growth data information on carbon sink.

[0049] In this embodiment, different types of trees can be managed more precisely by dividing the coverage area;

[0050] Step S60: Count the number of trees with different coverage, and based on this number, divide the tree height and diameter in the growth data into intervals, obtain the regular changes of the carbon sink of trees in different intervals, upload the regular changes to the database, and update the database.

[0051] In step S60, the tree height and diameter are calculated according to θ. 1 区间 θ 2 区间 and θ n 区间 The intervals are differentiated, where n represents the nth interval; and the difference in carbon sequestration between different intervals in the previous period is calculated using the following formula:

[0052] Where, δ i This represents the δ-th carbon sink acquired within the i-th cycle;

[0053] In the formula, C tree This indicates the total weight of trees in different sections, expressed in tons; △ tree t1 and t2 represent the average tree height and diameter of trees in different intervals; t1 and t2 represent the start and end times of the calculation.

[0054] Based on the above, this embodiment divides the corresponding period into at least four measurement periods according to the calculation results, and obtains the highest carbon sink amount generated in different measurement periods. Based on the highest carbon sink amount, and taking the atmospheric fluctuations of the preset monitoring sample area as a basis, including wind speed, the eddy covariance method is used to calculate the regularity of the highest carbon sink amount fluctuations within the same measurement period and under the same interval conditions, and the result is calculated according to the following formula:

[0055] Wherein, λ represents the time of the highest carbon sink fluctuation.

[0056] In the formula, w o This represents the w-th carbon sink amount collected from trees in the same interval during the o-th collection; q n This represents the q-th carbon sink fluctuation value collected in the nth same interval of trees; k represents the wind speed fluctuation collected in the vertical direction of the same interval of trees.

[0057] In this embodiment, the four measurement periods are divided according to the four seasons. According to actual forest management experience, seasonal changes have a certain impact on carbon sinks, including photosynthetic intensity, temperature and precipitation, as well as vegetation type and growth stage. For example, in summer, the long hours and high intensity of sunlight lead to the most vigorous photosynthesis and the strongest carbon absorption capacity. In contrast, in winter, the short hours and low intensity of sunlight weaken photosynthesis and reduce carbon absorption capacity. For instance, the highest atmospheric carbon dioxide concentration in the Northern Hemisphere is in winter, partly due to increased fossil fuel use caused by human activities such as heating, and partly due to weaker plant photosynthesis.

[0058] For example, trees grow rapidly in spring and summer and have a strong carbon absorption capacity, but enter a dormant period in winter and their carbon absorption capacity decreases; while shrubs and herbaceous plants grow vigorously in spring and summer and have a strong carbon absorption capacity, but wither in autumn and winter and their carbon absorption capacity decreases significantly.

[0059] Based on the above, under the same intervals within different measurement periods, at least 10 historically highest carbon sink values ​​are sorted by size. The changes in tree height and diameter are analyzed based on the sorted carbon sink data, and data is collected. Based on the collected data, the difference between tree height and diameter data within the same interval is calculated, and the data difference is marked as the reference data difference. If the reference data difference shows an increasing trend within the corresponding measurement period and the carbon sink value increases, the calculated carbon sink value is determined to be an outlier; otherwise, no determination is made.

[0060] Furthermore, when the carbon sink calculation is determined to be an outlier, a set of carbon sink data is collected to generate a dataset. Within the corresponding interval, the forest cover area within that interval is divided into high-coverage, medium-coverage, and low-coverage areas according to the calculated reference data difference. The carbon sink of the trees with the largest tree height and diameter in the first third of each coverage area is calculated separately, and the calculated values ​​are marked as reference values. If the carbon sink calculation value for trees in a future period exceeds the reference value, the carbon sink calculation is determined to be a normal value; otherwise, no determination is made.

[0061] Meanwhile, among the trees with the largest height and diameter in the first third of the forest, the proportion of the trees to all trees in the corresponding coverage area is calculated, and the number of trees is obtained based on the proportion. The trees are then sorted by number, and a critical threshold is preset based on the difference in carbon sequestration. If the difference in carbon sequestration calculated based on the number of trees in a future period exceeds the critical threshold, the carbon sequestration calculation value is determined to be an abnormal value; otherwise, no judgment is made.

[0062] In this embodiment, the carbon sink amount when it exceeds the critical threshold is recorded and stored. The average carbon sink amount of the trees is calculated based on the carbon sink amount, and the carbon sink amount of a single tree in a number of trees is calculated based on the average carbon sink amount. If the carbon sink amount of a certain tree is lower than the average carbon sink amount, the carbon sink amount generated by the tree is marked as the key carbon sink amount to be monitored.

[0063] In this embodiment, based on forest management experience, it is shown that a single tree also generates carbon sequestration, and the amount of carbon sequestration generated depends on a variety of factors, including tree species, growing environment, tree age, and management measures.

[0064] Based on the above, this embodiment presets the number of monitoring times according to the changes in carbon sink generated by forest trees, records the carbon sink corresponding to each monitoring time, analyzes the trend of carbon sink change, and plots ROC curves; at the same time, the preset number of monitoring times is given as a monitoring period. If the change trend shows a downward trend within the monitoring period, the forest trees are determined to be in an abnormal growth state and are removed from the monitoring objects.

[0065] In this embodiment, the number of monitoring sessions is within the corresponding measurement period, including once every half month.

[0066] A terminal includes a processor, an input interface, an output interface, and a memory, which are interconnected. The memory stores a computer program, which includes program instructions. The processor is configured to invoke the program instructions to execute a method for measuring and monitoring carbon sequestration in single-tree afforestation based on machine vision recognition.

[0067] A computer-readable storage medium storing a computer program, the computer program including program instructions, which, when executed by a processor, cause the processor to perform a method for measuring and monitoring carbon sequestration in single-tree afforestation based on machine vision recognition.

[0068] In summary, this invention utilizes machine vision recognition technology to effectively monitor forest growth data, improving the accuracy and efficiency of carbon sequestration measurement. Furthermore, by recording and analyzing growth data in stages, it can promptly identify patterns of forest growth changes, providing a scientific basis for dynamic monitoring of carbon sequestration.

[0069] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for measuring and monitoring carbon sequestration in single-tree afforestation based on machine vision recognition, characterized in that, Includes the following steps: step S10: Obtain spatial distribution data of the preset monitoring plots, the spatial distribution data including forest type and forest distribution; and deploy monitoring in the preset monitoring plots, the monitoring deployment method including setting up a number of observation cameras in the preset monitoring plots; Step S20: Based on the deployed observation cameras, acquire video and image data of the trees to obtain the growth data information of the trees, including tree height, tree diameter and coverage. Step S30: Record growth data information in cycles, and calculate carbon sequestration based on the recorded growth data information. At the same time, calculate the carbon sequestration of a single plant in the preset monitoring plot based on the calculation results. The cycled recording of growth data information includes recording in cycles of 1 to 2 years. Step S40: Analyze the regular changes in the growth data information in each cycle, calculate the differences in tree height, tree diameter and coverage based on the regular changes; at the same time, calculate the difference in carbon sink based on the differences to obtain the regular influence of the differences on the difference, and generate a database. Step S50: During the cycle alternation, the coverage of the preset monitoring plots is divided based on the previous cycle; wherein, the division method includes dividing according to the coverage of arbor forests, shrub forests and bamboo forests, analyzing the changes in growth data information of different coverage, and the impact of the changes in growth data information on the carbon sink. Step S60: Count the number of trees with different coverage, and based on the number, divide the tree height and diameter of the trees in the growth data information into intervals, obtain the regular changes of the carbon sink of trees in different intervals, upload the regular changes to the database, and update the database. In step S60, the method of interval differentiation includes dividing the tree height and diameter according to θ. 1 区间 θ 2 区间 and θ n 区间 The intervals are differentiated, where n represents the nth interval; and the difference in carbon sequestration between different intervals in the previous period is calculated using the following formula: Where, δ i This represents the δ-th carbon sink acquired within the i-th cycle; In the formula, C tree This represents the total weight of trees in different sections, expressed in tons; Δ tree t1 and t2 represent the average tree height and diameter of trees in different intervals; t1 and t2 represent the start and end times of the calculation. Based on the calculation results of the carbon sink difference, the corresponding period is divided into at least four measurement periods to obtain the highest carbon sink amount generated in different measurement periods. Based on the highest carbon sink amount, and taking the atmospheric fluctuations of the preset monitoring sample sites as a basis, the atmospheric fluctuations include wind speed. The eddy covariance method is used to calculate the regularity characteristics of the highest carbon sink amount fluctuations within the same measurement period and under the same interval conditions, and the result is calculated according to the following formula: Wherein, λ represents the time of the highest carbon sink fluctuation. In the formula, w o This represents the w-th carbon sink amount collected from trees in the same interval during the o-th collection; q n represents the q-th carbon sink fluctuation value collected in the nth same interval of trees; k represents the wind speed fluctuation collected in the vertical direction of the same interval of trees.

2. The method for single-tree afforestation carbon sequestration measurement and monitoring based on machine vision recognition as described in claim 1, characterized in that, In the same interval within different measurement periods, at least 10 historically highest carbon sequestration values ​​are sorted by size. The changes in tree height and diameter are analyzed based on the sorted carbon sequestration data, and data is collected. Based on the collected data, the difference in tree height and diameter within the same interval is calculated, and the difference is marked as a reference difference. If the reference difference shows an increasing trend and the carbon sequestration increases within the corresponding measurement period, the calculated carbon sequestration value is determined to be an outlier. Conversely, no judgment is made.

3. The method for single-tree afforestation carbon sequestration measurement and monitoring based on machine vision recognition as described in claim 2, characterized in that, When the carbon sink calculation is determined to be an outlier, a set of carbon sink data is collected to generate a dataset. Within the corresponding interval, the forest cover area within the interval is divided into high cover area, medium cover area and low cover area according to the calculated reference data difference. The carbon sink of the trees with the largest tree height and diameter in the first third of each cover area is calculated separately, and the calculated value is marked as the reference value. If the calculated carbon sequestration of the trees in the future exceeds the reference value, the carbon sequestration is determined to be a normal value. Conversely, no judgment is made.

4. The method for single-tree afforestation carbon sequestration measurement and monitoring based on machine vision recognition as described in claim 3, characterized in that, Among the trees with the largest height and diameter in the first third of the forest, the proportion of the trees to all trees in the corresponding coverage area is calculated, and the number of trees is obtained based on the proportion. The trees are then sorted by number, and a critical threshold is preset based on the difference in carbon sequestration. If the difference in carbon sequestration calculated based on the number of trees in a future period exceeds the critical threshold, the calculated carbon sequestration value is determined to be an abnormal value. Conversely, no judgment is made.

5. The method for single-tree afforestation carbon sequestration measurement and monitoring based on machine vision recognition as described in claim 4, characterized in that, Record and store the carbon sequestration amount when it exceeds the critical threshold, calculate the average carbon sequestration amount of the trees based on the carbon sequestration amount, and calculate the carbon sequestration amount of a single tree in the number of trees based on the average carbon sequestration amount. If the carbon sequestration amount of a certain tree is lower than the average carbon sequestration amount, then the carbon sequestration amount generated by the tree is marked as the carbon sequestration amount to be monitored.

6. The method for single-tree afforestation carbon sequestration measurement and monitoring based on machine vision recognition as described in claim 5, characterized in that, The number of monitoring sessions is preset based on the changes in carbon sink generated by the trees, and the carbon sink corresponding to each monitoring session is recorded. The trend of the carbon sink change is analyzed, and an ROC curve is plotted. At the same time, the preset number of monitoring sessions is given as a monitoring period. If the change trend shows a downward trend within the monitoring period, the trees are determined to be in an abnormal growth state and are removed from the monitoring objects.

7. A terminal, characterized in that, The device includes a processor, an input interface, an output interface, and a memory, which are interconnected. The memory is used to store a computer program, which includes program instructions. The processor is configured to call the program instructions to execute the method as described in any one of claims 1 to 6.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, the computer program including program instructions that, when executed by a processor, cause the processor to perform the method as described in any one of claims 1 to 6.

Images