Forest plant survey data rapid checking method and device

Abstract

The invention discloses a method and a device for quickly checking forest plant survey data in the technical field of quick data checking. The method comprises the following steps of: 1, establishing a standard database of forest plants; step 2, establishing a geographic three-dimensional diagram of the forest investigation area; step 3, obtaining plant appearance characteristics and performing primary check on the plant appearance characteristics and plant appearance characteristic data in a standard database; step 4, acquiring plant physiological characteristics and performing double check on the plant physiological characteristics and plant physiological characteristic data in a standard database; 5, combining the data of the first check and the data of the second check, and performing information matching with a standard database; and step 6, when the variety of the plant to be checked is matched with the information in the standard database, annotating the plant according to the name of the plant in the standard database, the appearance characteristics of the plant and the geographical location information of the plant. According to the method, the plant physiological feature information is introduced and combined with the plant appearance features for judgment, so that the detection accuracy is improved while the forest plants are rapidly detected.

Description

Technical Field

[0001] The present invention relates to the technical field of rapid data verification, and particularly to a method and device for rapidly verifying forest plant survey data. Background Art

[0002] Forest plants, as an important part of forest resources, cover trees, vegetation, and wild animals, plants, and microorganisms that rely on them for survival. They play a crucial role in nature and have various backgrounds and functions.

[0003] On the one hand, the roots and humus of forest plants can improve soil properties, protect the ground from direct rain impact, reduce soil erosion, and play a role in conserving water. Moreover, the roots of forest plants can hold the soil and reduce soil erosion; at the same time, dead branches and leaves can increase organic matter and humus, effectively improve soil structure, and increase soil fertility. Forest plants can absorb carbon dioxide through photosynthesis and release a large amount of oxygen, which can effectively slow down the greenhouse effect. 90% of the carbon in the terrestrial ecosystem is naturally stored in forests. At the same time, they can also block, filter, and adsorb dust, absorb toxic substances in the air, reduce the content of toxic substances in the air, and volatilize chemical substances to kill bacteria. On the other hand, it is one of the most abundant ecosystems on Earth, providing habitats for countless animal and plant species and being an important treasure house of biodiversity. Protecting forests directly and indirectly protects biodiversity.

[0004] Furthermore, in order to better protect biodiversity and evaluate forest management effects, through forest plant data surveys, not only can we understand the distribution and diversity of plant species in the forest, but also we can understand the growth and extinction dynamic laws of forest resources and the effects of management measures, providing a scientific basis for adjusting and optimizing management strategies. However, during the forest plant survey process, due to differences in the knowledge levels and experience of the collectors, there may be situations where the plant names are mislabeled or the data records are inaccurate. Moreover, there are a large number of forest plant species, widely distributed, with a large and complex survey workload, a long survey efficiency and cycle. At the same time, there may be some different plant species with roughly the same appearance, which will also affect the accuracy of manual or existing image recognition rapid verification methods.

[0005] In summary, the present invention proposes a method and device for rapidly verifying forest plant survey data, which jointly judges by introducing plant physiological characteristic information and plant appearance characteristics, improving the accuracy of verification while rapidly verifying forest plants. Summary of the Invention

[0006] To solve the above problems, the present invention provides a method and device for quickly verifying forest plant survey data. By introducing plant physiological characteristic information and making a combined judgment with plant appearance characteristics, the accuracy of verification is improved while quickly verifying forest plants.

[0007] To achieve the above object, the technical solution of the present invention is as follows: A method and device for quickly verifying forest plant survey data, including:

[0008] S1: Initially organize the original data collected in forest resource surveys and monitoring activities, and establish a standard database of forest plants;

[0009] S2: Establish a three-dimensional geographical map of the forest survey area, and obtain the geographical location information of the plants that need data verification;

[0010] S3: Obtain the ordinary image information of the plants that need data verification, extract the plant appearance characteristics, obtain the first verification data, and perform a primary verification on the first verification data with the plant appearance characteristic data in the standard database;

[0011] S4: Obtain the multi-spectral image information of the plants that need data verification and the surrounding environment information, extract the plant physiological characteristics, obtain the second verification data, and perform a secondary verification on the second verification data with the plant physiological characteristic data in the standard database;

[0012] S5: Combine the data of the primary verification and the secondary verification, obtain the combined index, and perform information matching with the standard database;

[0013] S5 also includes a weight ratio process:

[0014] S5-1: Based on the primary verification data and the secondary verification data, establish a regression model, and evaluate the influence degree of them on the final result through the model parameters;

[0015] S5-2: According to the evaluation result, set the weight range of the primary verification and the secondary verification;

[0016] S5-3: Within the set weight range, allocate the weights of the primary verification and the secondary verification according to the specific situation;

[0017] S5-4: According to the allocated weights, perform weighted processing on the data of the primary verification and the secondary verification, and analyze the result after the weighted processing to evaluate its accuracy and reliability;

[0018] When the plant species to be verified matches the information in the standard database, name the plant to be verified according to the plant name in the standard database. Based on the appearance characteristics of the plant and the geographical location information of the plant, generate a corresponding three-dimensional plant model in the three-dimensional geographical map, add an annotation window for the plant on the three-dimensional plant model and annotate it according to the plant name in the standard database. The specific location and size of the plant are annotated in the form of text of the appearance characteristics of the plant extracted from the geographical location information and the ordinary image information, and the growth condition is annotated in the form of text of the plant physiological characteristics; then, count the data of each annotation window to obtain the biodiversity analysis data of the plants in the forest.

[0019] Further, the original data in S1 at least includes on-site records, observation data and image materials.

[0020] Further, the plant appearance characteristics in S3 at least include plant shape, size and color.

[0021] Further, the plant physiological characteristics in S4 at least include chlorophyll content, photosynthesis rate and respiration rate.

[0022] Further, the environmental information in S4 at least includes oxygen content and carbon dioxide content.

[0023] Further, the annotation window in S6 is used to display the prediction result, plant appearance characteristics, geographical location and physiological characteristics.

[0024] Further, S4 also includes the process of extracting chlorophyll content:

[0025] S4-1 Extract the spectral reflectance information of each pixel at different wavelengths. For a specific band, extract the gray value of the corresponding band in the multispectral image.

[0026] S4-2 Establish a mathematical model between chlorophyll content and spectral reflectance according to the spectral reflectance information of the sample leaves with known chlorophyll content.

[0027] S4-3 Input the spectral reflectance information or gray value in the extracted multispectral image into the established model to calculate the chlorophyll content of each pixel or the whole leaf.

[0028] Further, S6 also includes the prediction process when the plant species to be verified does not match the information in the standard database:

[0029] S6-1 Establish a correlation evaluation model according to the characteristics of the plant to be verified.

[0030] S6-2 Input the morphological and physiological characteristic data of the plant to be verified into the correlation evaluation model, start the correlation evaluation model, and calculate the correlation scores between the plant to be verified and each plant species in the standard database;

[0031] S6-3 Select the plant species with the highest score as the prediction result according to the correlation scores;

[0032] S6-4 On the interface of the geographical three-dimensional map, open the annotation window of the plant to be verified, and display the prediction result obtained from the correlation evaluation in the annotation window, including at least the plant name and prediction probability information;

[0033] S6-5 In the geographical three-dimensional map of the forest survey area, locate the three-dimensional model of the plant to be verified, highlight the located three-dimensional model, and at the same time highlight the annotation window.

[0034] Further, the device includes: a flight module, an appearance feature acquisition module, a physiological feature acquisition module, an environment detection module, a three-dimensional module, and a processing module. The appearance feature acquisition module, the physiological feature acquisition module, and the environment monitoring module are all mounted on the flight module;

[0035] The flight module is used to fly in the forest area that needs to be verified according to a predetermined route;

[0036] The appearance feature acquisition module is used to obtain the ordinary image information of the forest plants in the area to be verified;

[0037] The physiological feature acquisition module is used to obtain the multi-spectral image information of the forest plants in the area to be verified;

[0038] The environment monitoring module is used to obtain the oxygen content and carbon dioxide content in the area to be verified;

[0039] The three-dimensional module is used to establish the geographical three-dimensional map of the forest survey area and the three-dimensional model of the forest plants;

[0040] The processing module is used to perform image processing on the ordinary image information and multi-spectral image information, and perform data processing on the obtained oxygen and carbon dioxide data, obtain the corresponding plant appearance features and plant physiological characteristic data, and perform weight analysis and matching to obtain the biodiversity analysis data of the plants in the forest.

[0041] Further, the three-dimensional module is also used to implement human-computer interaction, and update the existing plant three-dimensional model data based on the verified plant appearance feature data.

[0042] Adopting the above solution has the following beneficial effects:

[0043] 1. The existing rapid verification method for forest plant survey data mainly collects the general image information of forest plant images, analyzes the characteristic data of the color, size, and shape of plants in the images, and compares and matches the above characteristic data with the existing standard plant database to finally obtain the corresponding results. However, in the forest, there are many plant species. For some plants with the same shape and color but different species, when performing image analysis using the above method, the obtained results are prone to confusion, reducing the accuracy of verification. In the present invention, by identifying the plant appearance characteristic data such as the color, size, and shape of plants in the general image as a primary comparison and analysis condition, and simultaneously collecting the spectral image of the same plant to analyze its chlorophyll content (chlorophyll is one of the most important pigments in plants, mainly present in chloroplasts, which can absorb light energy and convert it into chemical energy. The chlorophyll content varies among different plant species, which may be related to the genetic characteristics, growth environment, and physiological state of plants. Therefore, the chlorophyll content can be used as a reference index for identifying plant species) as a secondary comparison and analysis condition; and obtaining the oxygen (plants need oxygen during respiration, and the oxygen concentration not only affects the intensity of respiration but also determines the respiratory type of plants. Different plant species may have different demands for and adaptabilities to oxygen. Some special plants, such as mangroves, can grow their roots in an oxygen-deficient environment, which reflects their special adaptability to oxygen content. Therefore, the oxygen content around plants can also be used as a clue for identifying certain special plant species) and carbon dioxide content (the carbon dioxide content around plants affects the intensity and efficiency of photosynthesis. Different plant species may have different sensitivities to and utilization efficiencies of carbon dioxide; at the same time, the carbon dioxide compensation points and saturation points of different plant species for photosynthesis are different, which reflects the differences in their adaptabilities and utilization efficiencies to carbon dioxide content. Therefore, the carbon dioxide content around plants can also be used as a reference factor for identifying plant species) around the plants as auxiliary conditions for the secondary comparison and analysis condition, so as to quickly and efficiently verify the forest plant survey data from both aspects of plant physiological characteristics and appearance characteristics, and conduct double verification through two different aspects of plant characteristics to more accurately identify plant species and evaluate their growth status, reducing the possibility of misjudgment and missed judgment of some plants with the same shape and color but different species, and improving the accuracy and reliability of the data.

[0044] 2. In this solution, the three-dimensional module can establish the geographical three-dimensional map of the forest survey area and the three-dimensional model of forest plants, realizing the visualization of data. This not only helps to intuitively understand the spatial distribution and growth of forest plants, but also can update the three-dimensional model of plants through accurate image information for each verification, better reflecting the changes in forest plants, thus providing strong support for scientific research and management decisions.

[0045] The 3D module also supports human-computer interaction, allowing users to interact with the device through the interface to view and analyze data.

[0046] 3. In this solution, by integrating a flight module, an appearance feature acquisition module, a physiological feature acquisition module, etc., the automatic monitoring of forest plants is realized. It can not only improve work efficiency, but also reduce labor costs, and at the same time promote the investigation and verification of forest plants on a large scale.

[0047] Additional aspects and advantages of the present invention will be given in part in the following description, become apparent in part from the following description, or be learned through the practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 It is a schematic flowchart of the method for quickly verifying forest plant survey data in an embodiment of the present invention;

[0049] Figure 2 It is a schematic diagram of the chlorophyll content extraction process of the method for quickly verifying forest plant survey data in an embodiment of the present invention;

[0050] Figure 3 It is a schematic diagram of the weight ratio process of the method for quickly verifying forest plant survey data in an embodiment of the present invention;

[0051] Figure 4 It is a schematic diagram of the device of the method for quickly verifying forest plant survey data in an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0052] The technical solutions of the present invention will be described clearly and completely below with reference to the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of them. 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.

[0053] In the description of the present invention, it should be noted that the orientation or positional relationship indicated by the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc. is based on the orientation or positional relationship shown in the drawings. It is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the present invention. In addition, the terms "first", "second", "third" are only used for descriptive purposes and cannot be construed as indicating or implying relative importance.

[0054] In the description of the present invention, it should be noted that unless otherwise clearly specified and limited, the terms "installation", "connection", and "coupling" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, and it can be the communication inside two components. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0055] The following is a further detailed description through specific embodiments:

[0056] Embodiment 1:

[0057] As shown in Figure 1 , Figure 2 and Figure 3 : A method for quickly checking forest plant survey data, including:

[0058] S1 Preliminary collation of the original data collected in forest resource surveys and monitoring activities to establish a standard database of forest plants; the original data in S1 includes at least field records, observation data, and image materials.

[0059] S2 Establish a three-dimensional geographical map of the forest survey area to obtain the geographical location information of the plants that need data checking; the geographical location information mainly includes distribution areas, growth environments, and the natural zones to which they belong, etc., so that the established three-dimensional geographical map fits the actual situation to the greatest extent, facilitating subsequent staff to conduct corresponding forest plant diversity analysis surveys.

[0060] S3 Obtain the ordinary image information of the plants that need data checking, extract the plant appearance features to obtain the first verification data, and perform a first-level verification on the first verification data and the plant appearance feature data in the standard database; the plant appearance features in S3 include at least plant shape, size, and color. Among them, the extraction of plant appearance features is realized through existing image recognition technologies, such as: multi-scale convolutional neural networks, YOLOV8, DEEPLABV3+, and other plant leaf image segmentation technologies.

[0061] S4 obtains the multispectral image information of the plants that need data verification and the information of their surrounding environment, extracts the physiological characteristics of the plants to obtain the second verification data, and performs a double verification on the second verification data and the plant physiological characteristic data in the standard database; the plant physiological characteristics in S4 at least include chlorophyll content, photosynthesis rate, and respiration rate; the environmental information in S4 at least includes oxygen content and carbon dioxide content. Among them, for the extraction of chlorophyll content, it is achieved through hyperspectral image technology. Hyperspectral image technology can obtain the reflection or transmission information of plant leaves in different spectral bands, and these information reflect the internal physiological state and chemical composition of the leaves, including chlorophyll content. By collecting the hyperspectral images of the leaves with a hyperspectral imaging device and using algorithms of vegetation indices (such as ratio vegetation index RVI, normalized difference vegetation index NDVI, modified chlorophyll absorption ratio index MCARI, etc.), the plant canopy coverage and chlorophyll content can be measured. These algorithms utilize the reflectance differences in different spectral bands to calculate the vegetation indices related to chlorophyll content, thereby realizing the rapid estimation of chlorophyll content.

[0062] The chlorophyll content extraction process is also included in S4:

[0063] S4-1 Extracts the spectral reflectance information of each pixel point at different wavelengths. For a specific band, extracts the gray value of the corresponding band in the multispectral image.

[0064] S4-2 Based on the spectral reflectance information of the sample leaves with known chlorophyll content, establishes a mathematical model between chlorophyll content and spectral reflectance.

[0065] S4-3 Inputs the extracted spectral reflectance information or gray value in the multispectral image into the established model, and calculates the chlorophyll content of each pixel point or the entire leaf.

[0066] S5 combines the data of the first verification and the second verification, obtains the combined index, and performs information matching with the standard database.

[0067] The weight ratio process is also included in S5:

[0068] S5-1 Based on the first verification data and the second verification data, establishes a regression model, and evaluates the influence degree of them on the final result through the model parameters; if the first verification is already accurate enough, then its weight may be higher; if the second verification can provide more accurate or additional information, then its weight should also be increased accordingly.

[0069] S5-2 According to the evaluation results, sets the weight ranges of the first verification and the second verification.

[0070] Within the set weight range, allocate the weights for the first-level verification and the second-level verification according to the specific situation. After the weights are allocated, perform a verification to ensure that the sum of the weights is 100%.

[0071] According to the allocated weights, perform weighted processing on the data of the first-level verification and the second-level verification, and analyze the results after the weighted processing to evaluate its accuracy and reliability. For example, if the weight of the first-level verification is 60% and the weight of the second-level verification is 40%, the final result can be expressed as: Final result = First-level verification data × 60% + Second-level verification data × 40%. At the same time, if the result is consistent with the expectation, it indicates that the weight ratio is reasonable; if the result has a large deviation from the expectation, it is necessary to re-consider the weight ratio principle and method.

[0072] When the plant species to be verified matches the information in the standard database, name the plant to be verified according to the plant name in the standard database. Based on the appearance characteristics of the plant and the geographical location information of the plant, generate a corresponding three-dimensional plant model in the geographical three-dimensional map, add an annotation window to the three-dimensional plant model and annotate it according to the plant name in the standard database. The specific location and size of the plant are annotated in the form of text of the plant appearance characteristics extracted from the geographical location information and the ordinary image information, and the growth situation is annotated in the form of text of the plant physiological characteristics; then, count the data of each annotation window to obtain the biodiversity analysis data of the plants in the forest; the annotation window in S6 is used to display the prediction result, plant appearance characteristics, geographical location and physiological characteristics. Among them, both information matching and relevance evaluation are realized through the deep learning model.

[0073] Among them, information matching refers to comparing the image information of the leaves of the plant to be detected with the information in the standard database to find the most similar matching item. In the chlorophyll content detection, the deep learning model can learn a large number of plant leaf image data, extract the key features in the image, such as color, texture, shape, etc., and compare these features with the known information in the database. Specifically, the deep learning model (such as the convolutional neural network CNN) can automatically learn and extract features from the image, and these features can well represent the content of the image. Then, the model will calculate the similarity between the image to be detected and each known image in the database, usually using distance measurement methods (such as Euclidean distance, cosine similarity, etc.) to evaluate the similarity. Finally, the model will select the known image with the highest similarity as the matching item, so as to provide preliminary information such as the species of the plant to be detected.

[0074] The S6 also includes the prediction process when the plant species to be verified does not match the information in the standard database:

[0075] S6-1 Establish a relevance evaluation model according to the characteristics of the plant to be verified;

[0076] S6-2 Input the morphological and physiological characteristic data of the plant to be verified into the relevance evaluation model, start the relevance evaluation model, and calculate the relevance scores of the plant to be verified and each plant species in the standard database;

[0077] S6-3 Select the plant species with the highest score as the prediction result according to the relevance scores;

[0078] S6-4 On the interface of the geographical 3D map, open the annotation window of the plant to be verified, and display the prediction result obtained from the relevance evaluation in the annotation window, including at least the plant name and prediction probability information;

[0079] S6-5 In the geographical 3D map of the forest survey area, locate the 3D model of the plant to be verified, highlight the located 3D model, and at the same time highlight the annotation window.

[0080] Among them, the relevance evaluation is based on information matching, and further evaluates the degree of association between the leaves of the plant to be detected and the known information in the standard database. This usually involves in-depth analysis and comparison of image features, as well as using machine learning or deep learning models for prediction and classification. In the chlorophyll content detection, the deep learning model can evaluate the degree of association between the leaves of the plant to be detected and the known chlorophyll content by analyzing features in the image, such as the greenness of the leaves and texture changes. The model can learn and extract features closely related to the chlorophyll content, and use these features to predict the chlorophyll content of the leaves of the plant to be detected. In addition, the deep learning model can also be trained and optimized to improve the accuracy of the relevance evaluation. During the training process, the model will learn how to predict the chlorophyll content based on image features, and continuously adjust the model parameters to reduce the prediction error, improve the accuracy of the relevance model evaluation, and further improve the accuracy of the rapid verification of forest plant survey data.

[0081] Embodiment 2

[0082] As shown in the appendix Figure 4 A rapid verification device for forest plant survey data includes: a flight module, an appearance feature acquisition module, a physiological feature acquisition module, an environmental monitoring module, a 3D module, and a processing module. The appearance feature acquisition module, the physiological feature acquisition module, and the environmental monitoring module are all loaded on the flight module;

[0083] The flight module is used to fly in the forest area to be verified according to a predetermined route.

[0084] Among them, the flight module is the core power and navigation part of the whole device, and it has highly intelligent autonomous flight capabilities. This module is built-in with an advanced flight control system and a navigation system, and can fly stably and precisely over the forest according to the preset route. Whether it is a complex mountainous terrain or a dense forest area, the flight module can respond flexibly to ensure covering all areas that need to be inspected. At the same time, the flight module also has strong endurance and can complete large-area forest inspection work in one flight mission, greatly improving work efficiency.

[0085] The appearance feature acquisition module is used to obtain the ordinary image information of forest plants in the area to be inspected.

[0086] Among them, the appearance feature acquisition module is preferably a high-resolution camera, which can capture clear images of plant leaves, branches, flowers and other parts, providing an important basis for subsequent plant classification and identification.

[0087] The physiological feature acquisition module is used to obtain the multi-spectral image information of forest plants in the area to be inspected. Among them, the multi-spectral imaging technology can capture the reflection and transmission information of plants under different spectra, thereby revealing the internal physiological state and health status of plants, and can accurately measure key physiological indicators such as chlorophyll content, water content, and nutritional status of plant leaves.

[0088] The environmental monitoring module is used to obtain the oxygen content and carbon dioxide content in the area to be inspected.

[0089] The 3D module is used to establish a geographical 3D map of the forest survey area and a 3D model of forest plants; it is also used to achieve human-computer interaction and update the existing plant 3D model data based on the plant appearance feature data that has passed the inspection.

[0090] The processing module is used to perform image processing on the ordinary image information and multi-spectral image information and data processing on the obtained oxygen and carbon dioxide data, obtain the corresponding plant appearance features and plant physiological feature data, and perform weight analysis and ratio matching to obtain the biodiversity analysis data of plants in the forest.

[0091] Obviously, the above embodiments are only examples given for clear illustration and not limitations on the implementation manners. For those of ordinary skill in the art, other different forms of changes or modifications can be made based on the above description. It is not necessary and impossible to list all implementation manners here. And the obvious changes or modifications derived therefrom are still within the protection scope of the present invention.

Claims

1. A method for rapid verification of forest plant survey data, characterized in that: include: S1 Preliminarily organize the original data collected in forest resource survey and monitoring activities and establish a standard database of forest plants; S2 establishes a three-dimensional geographic map of the forest survey area and obtains the geographic location information of plants that require data verification; S3: obtaining common image information of the plant that needs data verification, extracting plant appearance features, obtaining first verification data, and performing a double check on the first verification data with plant appearance feature data in a standard database; S4 obtains multispectral image information of the plant that needs data verification and its surrounding environment information, extracts plant physiological characteristics, obtains second verification data, and performs double verification on the second verification data and plant physiological characteristic data in the standard database; S5 combines the data of the first check and the second check to obtain the combined index and matches the information with the standard database; S5 also includes the weight matching process: S5-1 Establish a regression model based on the first-check data and the second-check data, and use the model parameters to evaluate their influence on the final results; S5-2 Set the weight range of the first and second checks based on the evaluation results; S5-3: Allocate the weights of the first check and the second check according to the specific situation within the set weight range; S5-4 performs weighted processing on the data of the first check and the second check according to the assigned weights, and analyzes the weighted processing results to evaluate their accuracy and reliability; S6 When the plant species to be inspected matches the information in the standard database, the plant to be inspected is named according to the plant name in the standard database, and a corresponding plant three-dimensional model is generated in the geographic three-dimensional map based on the plant appearance characteristics and the plant geographic location information. An annotation window of the plant is added to the plant three-dimensional model and annotated according to the plant name in the standard database. The specific location and size of the plant are annotated in the form of text using the plant appearance characteristics extracted from the geographic location information and ordinary image information, and the growth condition is annotated in the form of text using the plant physiological characteristics. The data in each annotation window is then statistically analyzed to obtain the biodiversity analysis data of the plants in the forest.

2. The method for rapid verification of forest plant survey data according to claim 1, characterized in that: The raw data in S1 at least includes on-site records, observation data and image data.

3. The method for rapid verification of forest plant survey data according to claim 1, characterized in that: The plant appearance characteristics in S3 at least include plant shape, size and color.

4. The method for rapid verification of forest plant survey data according to claim 1, characterized in that: The plant physiological characteristics in S4 at least include chlorophyll content, photosynthesis rate and respiration rate.

5. The method for rapid verification of forest plant survey data according to claim 1, characterized in that: The environmental information in S4 at least includes oxygen content and carbon dioxide content.

6. The method for rapid verification of forest plant survey data according to claim 1, characterized in that: The annotation window in S6 is used to display the prediction results, plant appearance characteristics, geographical location and physiological characteristics.

7. The method for rapid verification of forest plant survey data according to claim 1, characterized in that: The S4 also includes the chlorophyll content extraction process: S4-1 extracts the spectral reflectance information of each pixel at different wavelengths, and for a specific band, extracts the grayscale value of the corresponding band in the multispectral image; S4-2 establishes a mathematical model between chlorophyll content and spectral reflectance based on the spectral reflectance information of sample leaves with known chlorophyll content; S4-3 inputs the spectral reflectance information or grayscale value extracted from the multispectral image into the established model to calculate the chlorophyll content of each pixel or the entire leaf.

8. The method for rapid verification of forest plant survey data according to claim 1, characterized in that: The S6 also includes a prediction process when the plant species to be checked does not match the information in the standard database: S6-1 Establish a correlation evaluation model based on the characteristics of the nuclear plants to be tested; S6-2 inputs the morphological and physiological characteristic data of the plant to be tested into the correlation evaluation model, starts the correlation evaluation model, and calculates the correlation score between the plant to be tested and each plant species in the standard database; S6-3 selects the plant species with the highest score as the prediction result according to the correlation score; S6-4, on the interface of the geographic three-dimensional map, opening an annotation window of the plant to be checked, and displaying the prediction result obtained by the correlation evaluation in the annotation window, including at least the plant name and prediction probability information; S6-5 locates the three-dimensional model of the plant to be inspected in the geographical three-dimensional map of the forest survey area, highlights the located three-dimensional model, and also highlights the annotation window.

9. A rapid verification device for forest plant survey data, characterized in that: include: The flight module, the appearance feature acquisition module, the physiological feature acquisition module, the environment detection module, the three-dimensional module and the processing module are all loaded on the flight module; A flight module, used to fly in the forest area that needs to be checked according to a predetermined route; Appearance feature acquisition module, used to obtain common image information of forest plants in the area to be inspected; Physiological characteristics acquisition module, used to obtain multi-spectral image information of forest plants in the area to be inspected; Environmental monitoring module, used to obtain the oxygen content and carbon dioxide content in the nuclear area to be inspected; 3D module, used to build 3D geographical maps of forest survey areas and 3D models of forest plants; The processing module is used to perform image processing on ordinary image information and multispectral image information, as well as data processing on the acquired oxygen and carbon dioxide data, to obtain corresponding plant appearance characteristics and plant physiological characteristics data, and to perform weight analysis and matching to obtain biodiversity analysis data of plants in the forest.

10. The device for the rapid verification method of forest plant survey data according to claim 9, characterized in that: The three-dimensional module is also used to realize human-computer interaction and update the existing plant three-dimensional model data based on the verified plant appearance feature data.

Images