Forest wetland environment transition zone identification method and system for unmanned aerial vehicle

Abstract

The invention relates to the technical field of computer vision and image processing, in particular to a forest wetland environment transition zone identification method and system for an unmanned aerial vehicle. The method comprises the steps of obtaining various prior data layers of a detection area and fluctuation scales corresponding to the prior data layers, taking the fluctuation scales as weights, obtaining similarity degrees between area boundaries of different prior data layers, and identifying similar boundary groups based on the similarity degrees; obtaining the environment fusion degree of the similar boundary groups; and based on the environment fusion degree, identifying an environment transition zone and an unmanned aerial vehicle flight focus region. By fusing prior data layers of DEM, hydrology, vegetation and the like, analyzing boundary similarity according to a fluctuation scale and identifying an environment fusion area, an environment transition zone and an unmanned aerial vehicle key monitoring area are further determined, targeted monitoring and identification of a key interface of a complex ecological crisscross zone are realized, the precision and efficiency of complex environment monitoring are effectively improved, and the method is suitable for popularization and application. And meanwhile, terrain and water area flight risks are avoided.

Description

Technical Field

[0001] This application belongs to the field of computer vision and image processing technology, specifically relating to a method and system for identifying the transition zone between forest and wetland environments for unmanned aerial vehicles (UAVs). Background Technology

[0002] By analyzing the spatial distribution differences and coupling relationships of multidimensional data such as Digital Elevation Model (DEM), hydrology, and vegetation, heterogeneous regions with dramatic topographic relief, complex hydrological conditions, or abrupt changes in vegetation communities within the forest-wetland transition zone can be identified.

[0003] However, traditional environmental transition zone identification methods only reflect the complex changes in a region based on the division of the environmental area. They often do not provide a comprehensive analysis of the environmental complexity of the region and are not accurate enough in identifying the environmental transition zone. This results in inaccurate acquisition of the key areas of focus for UAV flights, affecting flight safety. Summary of the Invention

[0004] To address the aforementioned issues, this application provides a method and system for identifying forest-wetland transition zones using unmanned aerial vehicles (UAVs).

[0005] According to a first aspect of the embodiments of this application, a method for identifying forest-wetland transition zones for unmanned aerial vehicles (UAVs) is provided, the method comprising: Multiple prior data layers of the detection area are acquired, including a digital elevation model (DEM) data layer, a hydrological data layer, and a vegetation data layer. Obtain the fluctuation scale corresponding to the prior data layer, whereby the fluctuation scale is used to characterize the range of data fluctuation allowed by the prior data layer when performing regional division. Using the fluctuation scale as a weight, the similarity between the regional boundaries of different prior data layers is obtained, and similar boundary groups are identified based on the similarity. Obtain the degree of environmental integration of the similar boundary groups; Based on the degree of environmental integration, environmental transition zones and key areas of focus for UAV flights are identified.

[0006] In one implementation, obtaining the fluctuation scale corresponding to the prior data layer includes: Obtain the acceptable fluctuation scale of the relevant objects in the prior data layer; Based on the acceptable level of the fluctuation scale, the fluctuation scale corresponding to the prior data layer is obtained.

[0007] In one implementation, obtaining the acceptable fluctuation scale of the relevant objects in the prior data layer includes: Obtain the first average value of the distribution area of ​​the relevant objects in the detection area; Obtain the second mean of the maximum area of ​​all objects in the detection region; Obtain the total area of ​​the detection region; Based on the first mean, the second mean, and the total area, the acceptable fluctuation scale of the relevant objects in the prior data layer is obtained.

[0008] In one implementation, obtaining the fluctuation scale corresponding to the prior data layer based on the acceptable fluctuation scale includes: The acceptable fluctuation scale is subjected to maximum and minimum normalization to obtain the normalized acceptable fluctuation scale. Obtain the preset reference fluctuation scale; Based on the normalized fluctuation scale acceptability and the reference fluctuation scale, the fluctuation scale corresponding to the prior data layer is obtained.

[0009] In one implementation, the step of obtaining the similarity between regional boundaries of different prior data layers using the fluctuation scale as weight, and identifying similar boundary groups based on the similarity, includes: Obtain the initial similarity between the region boundaries of different prior data layers; Based on the fluctuation scale and the initial similarity, a corrected similarity is obtained, and the corrected similarity is used as the similarity between the regional boundaries of different prior data layers; The similarity is processed by maximum and minimum normalization to obtain the normalized similarity. If the normalized similarity is greater than the first preset threshold, it is determined that the regional boundaries of different prior data layers are in the same similar boundary region. Similar boundary regions are determined for all boundaries of the detection area to obtain multiple similar boundary groups.

[0010] In one implementation, obtaining the initial similarity between the region boundaries of different prior data layers includes: Obtain the minimum distance between any point on the boundary of the first prior data layer and the nearest point on the boundary of the second prior data layer, and obtain the sum of all minimum distances; Obtain the Fraser distance between the region boundaries of the first prior data layer and the region boundaries of the second prior data layer; Obtain the minimum Fraser distance between the boundaries of all regions within the detection area; Based on the sum of all minimum distances, the Fraser distance, and the minimum Fraser distance, the initial similarity between the region boundaries of different prior data layers is obtained.

[0011] In one implementation, obtaining the environmental blending degree of the similar boundary grouping includes: Obtain the maximum enclosed area formed by all boundaries in the group of similar boundaries; Obtain the third mean of the similarity between all pairwise region boundaries in the similar boundary group; Obtain the maximum and minimum values ​​of the similarity between all pairwise region boundaries in the similar boundary group; The degree of environmental integration of the similar boundary grouping is obtained based on the maximum enclosing area, the third mean, the maximum value of the similarity, and the minimum value of the similarity.

[0012] In one implementation, identifying environmental transition zones and key areas of interest for UAV flight based on the degree of environmental integration includes: If the degree of environmental integration is greater than a second preset threshold, the similar boundary group is determined to be an environmental transition zone; Based on the degree of environmental integration, the flight attention level of each location point in the environmental transition zone is obtained; Based on the stated level of flight attention, the key areas of focus for UAV flight are identified.

[0013] In one implementation, obtaining the flight attention level of each location point in the environmental transition zone based on the degree of environmental integration includes: The saliency results map for each environment type is obtained based on the saliency detection algorithm; Obtain the fourth mean of the values ​​at any point in the environmental transition zone across all significance result maps; Obtain the absolute value of the difference between the value of any location point on any of the saliency result maps and the fourth mean; Obtain the maximum value of the environmental blending degree in the detection area; The degree of flight attention at any location point in the environmental transition zone is obtained based on the degree of environmental integration in the environmental transition zone, the maximum value of the degree of environmental integration, and the absolute value.

[0014] According to a second aspect of the embodiments of this application, a forest-wetland environment transition zone identification system for unmanned aerial vehicles (UAVs) is provided, the system comprising a server, the server comprising: A memory on which computer programs are stored; A processor for executing the computer program in the memory to implement the steps of the method of any one of the first aspects.

[0015] The embodiments of this application have the following beneficial effects: This application provides a method for identifying forest-wetland environmental transition zones using unmanned aerial vehicles (UAVs), including: acquiring multiple prior data layers of the detection area and the fluctuation scales corresponding to the prior data layers; using the fluctuation scales as weights to obtain the similarity between the regional boundaries of different prior data layers, and identifying similar boundary groups based on the similarity; obtaining the environmental integration degree of the similar boundary groups; and identifying the environmental transition zone and the key areas of focus for UAV flight based on the environmental integration degree. By fusing prior data layers such as DEM, hydrology, and vegetation, and analyzing boundary similarity based on fluctuation scales, environmental integration zones are identified, thereby determining the environmental transition zone and the key monitoring areas for UAVs. This achieves targeted monitoring and identification of key interfaces in complex ecological ecotones, effectively improving the accuracy and efficiency of complex environmental monitoring, while avoiding the risks of terrain and water area flight.

[0016] Other features and advantages of this application will be described in detail in the following detailed description section. Attached Figure Description

[0017] To more clearly illustrate the implementation schemes of this application, the accompanying drawings used in the implementation schemes will be briefly introduced below. It should be understood that the accompanying drawings only show some implementation schemes of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained from the accompanying drawings without creative effort.

[0018] Figure 1 This is a flowchart illustrating a method for identifying the transition zone between forest and wetland environments for unmanned aerial vehicles (UAVs) according to an exemplary embodiment. Figure 2 This is a flowchart illustrating a method for obtaining the fluctuation scale corresponding to a priori data layer according to an exemplary embodiment; Figure 3 This is a flowchart illustrating a method for obtaining the acceptable fluctuation scale of a relevant object in a priori data layer, according to an exemplary embodiment. Figure 4 This is a flowchart illustrating a method for obtaining the fluctuation scale corresponding to a priori data layer based on the fluctuation scale acceptability, according to an exemplary embodiment. Figure 5 This is a flowchart illustrating an exemplary embodiment of a method for obtaining the similarity between regional boundaries of different prior data layers using fluctuation scale as weight, and identifying similar boundary groups based on the similarity. Figure 6 This is a flowchart illustrating a method for obtaining the initial similarity between region boundaries of different prior data layers according to an exemplary embodiment; Figure 7This is a flowchart illustrating a method for obtaining the degree of environmental integration of similar boundary groups according to an exemplary embodiment; Figure 8 This is a flowchart illustrating a method for identifying environmental transition zones and key areas of interest for drone flights based on the degree of environmental integration, according to an exemplary embodiment. Figure 9 This is a flowchart illustrating a method for obtaining the degree of flight attention at each location point in an environmental transition zone based on the degree of environmental integration, according to an exemplary embodiment. Figure 10 This is a block diagram illustrating a forest-wetland environment transition zone identification system for unmanned aerial vehicles (UAVs) according to an exemplary embodiment. Figure 11 This is a block diagram illustrating a server according to an exemplary embodiment. Detailed Implementation

[0019] The specific embodiments of this application will be described in detail below with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are for illustration and explanation only and are not intended to limit this application.

[0020] It should be understood that the term "comprising" and its variations as used herein are open-ended, meaning "including but not limited to". The term "based on" means "at least partially based on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". Definitions of other terms will be given in the following description.

[0021] It should be noted that the concepts of "first" and "second" mentioned in this application are only used to distinguish different devices, modules, or units, and are not used to limit the order of functions performed by these devices, modules, or units or their interdependencies. The modifications of "one" and "multiple" mentioned in this application are illustrative rather than restrictive, and those skilled in the art should understand that, unless explicitly stated in the context, they should be understood as "one or more". In the description of this application, unless otherwise stated, "multiple" means two or more, and other quantifiers are similar; "at least one item", "one item or multiple items", or similar expressions refer to any combination of these items, including any combination of single or multiple items.

[0022] Although operations or steps are described in a specific order in the accompanying drawings in the embodiments of this application, this should not be construed as requiring these operations or steps to be performed in the specific order or serial order shown, or requiring all of the shown operations or steps to be performed to obtain the desired result. In the embodiments of this application, these operations or steps may be performed serially; they may be performed in parallel; or a portion of these operations or steps may be performed.

[0023] The names of the messages or information exchanged between multiple devices in the embodiments of this application are for illustrative purposes only and are not intended to limit the scope of these messages or information. It is understood that before using the technical solutions of the various embodiments of this application, the type, scope of use, and usage scenarios of the personal information involved in this application should be disclosed to the user and their authorization obtained in accordance with relevant laws and regulations through appropriate means.

[0024] Figure 1 This is a flowchart illustrating a method for identifying a forest-wetland transition zone for an unmanned aerial vehicle (UAV) according to an exemplary embodiment. Figure 1 As shown in the figure, this application provides a method for identifying the forest-wetland transition zone for unmanned aerial vehicles (UAVs), which may include the following steps: In step S10, multiple prior data layers of the detection area are acquired, including a digital elevation model (DEM) data layer, a hydrological data layer, and a vegetation data layer.

[0025] In this step, multiple prior data layers of the detection area are acquired, including a Digital Elevation Model (DEM) data layer, a hydrological data layer, and a vegetation data layer. For example, oblique photography or lidar scanning can be performed using a drone equipped with a high-precision PPK / RTK positioning module to generate a high-resolution DEM that accurately reflects topographic relief. Using multispectral or hyperspectral sensors, the unique absorption characteristics of water bodies in the near-infrared band are analyzed to accurately identify water boundaries and distribution, and to monitor parameters such as water depth and quality. Multispectral cameras acquire information in the red-edge and near-infrared bands, and vegetation indices such as NDVI (Normalized Difference Vegetation Index) and NDRE (Normalized Red Edge Index) are calculated to effectively invert vegetation cover, species distribution, and biomass status. Prior data such as DEM data, hydrological data, and vegetation data of the detection area are acquired in this way, and relevant image data are retained as reference data for subsequent transition zone identification.

[0026] In step S20, the fluctuation scale corresponding to the prior data layer is obtained. The fluctuation scale is used to characterize the range of data fluctuation allowed by the prior data layer when performing regional division.

[0027] In this step, the fluctuation scale corresponding to the prior data layer is obtained. The fluctuation scale is used to characterize the range of data fluctuation allowed by the prior data layer when performing region division. For example, the acceptable fluctuation scale of the relevant objects in the prior data layer can be obtained first, and then the fluctuation scale corresponding to the prior data layer can be obtained based on the acceptable fluctuation scale.

[0028] In step S30, the similarity between the regional boundaries of different prior data layers is obtained using the fluctuation scale as the weight, and similar boundary groups are identified based on the similarity.

[0029] In this step, the similarity between the regional boundaries of different prior data layers is obtained using the fluctuation scale as a weight, and similar boundary groups are identified based on the similarity. For example, an initial similarity between the regional boundaries of different prior data layers can be obtained first. Then, based on the fluctuation scale and the initial similarity, a corrected similarity can be obtained, and this corrected similarity can be used as the similarity between the regional boundaries of different prior data layers. Next, the similarity is subjected to max-min normalization to obtain a normalized similarity. If the normalized similarity is greater than a first preset threshold, it is determined that the regional boundaries of different prior data layers are in the same similar boundary region. Finally, similar boundary regions are determined for all regional boundaries of the detected region, resulting in multiple similar boundary groups.

[0030] In step S40, the degree of environmental integration of the similar boundary grouping is obtained.

[0031] In this step, the degree of environmental integration of similar boundary groups is obtained. For example, the maximum enclosing area formed by all boundaries in the similar boundary group can be obtained first, then the third mean of the similarity between all pairwise region boundaries in the similar boundary group can be obtained, then the maximum and minimum values ​​of the similarity between all pairwise region boundaries in the similar boundary group can be obtained, and finally the degree of environmental integration of the similar boundary group is obtained based on the maximum enclosing area, the third mean, the maximum value of similarity, and the minimum value of similarity.

[0032] In step S50, based on the degree of environmental integration, the environmental transition zone and the key areas of focus for UAV flight are identified.

[0033] In this step, based on the degree of environmental blending, environmental transition zones and key areas of interest for UAV flight are identified. For example, if the degree of environmental blending is greater than a second preset threshold, similar boundaries can be grouped into environmental transition zones. Then, based on the degree of environmental blending, the flight interest level of each location point in the environmental transition zone is obtained. Finally, based on the flight interest level, the key areas of interest for UAV flight are obtained.

[0034] This process achieves targeted coverage of key interfaces in complex ecological ecotones, effectively improving the accuracy and efficiency of monitoring complex environments, while avoiding the risks of flying over terrain and water.

[0035] Figure 2 This is a flowchart illustrating a method for obtaining the fluctuation scale corresponding to a priori data layer according to an exemplary embodiment. Figure 2 As shown, obtaining the fluctuation scale corresponding to the prior data layer may include the following steps: In step S201, the acceptable fluctuation scale of the relevant objects in the prior data layer is obtained.

[0036] In this step, the acceptable scale of fluctuation of the relevant objects in the prior data layer is obtained. For example, a first mean of the distribution area of ​​the relevant objects in the detection region can be obtained first, then a second mean of the maximum area of ​​all objects in the detection region can be obtained, then the total area of ​​the detection region can be obtained, and finally the acceptable scale of fluctuation of the relevant objects in the prior data layer can be obtained based on the first mean, the second mean, and the total area.

[0037] In step S202, the fluctuation scale corresponding to the prior data layer is obtained based on the acceptable fluctuation scale.

[0038] In this step, the fluctuation scale corresponding to the prior data layer is obtained based on the acceptable fluctuation scale. For example, the acceptable fluctuation scale can first be processed by max-min normalization to obtain the normalized acceptable fluctuation scale, then a preset reference fluctuation scale can be obtained, and finally the fluctuation scale corresponding to the prior data layer can be obtained based on the normalized acceptable fluctuation scale and the reference fluctuation scale.

[0039] This process matches the fluctuation scale with the characteristics of the data itself and the requirements for detection accuracy, providing a precise weighting basis for subsequent regional boundary similarity analysis and ensuring the reliability of the subsequent analysis results.

[0040] Figure 3 This is a flowchart illustrating a method for obtaining the acceptable fluctuation scale of relevant objects in a priori data layer, according to an exemplary embodiment. Figure 3 As shown, obtaining the acceptable fluctuation scale of relevant objects in the prior data layer may include the following steps: In step S2011, the first average value of the distribution area of ​​the relevant object in the detection area is obtained.

[0041] In this step, the first mean value of the distribution area of ​​the objects related to prior data layer i in the detection region is obtained. For example, object recognition methods based on deep learning (such as YOLO, Faster R-CNN, etc.) can extract image features from prior data through convolutional neural networks and use classification and regression techniques to directly locate and identify relevant objects (water flow, vegetation, etc.) in the image.

[0042] In step S2012, the second mean of the maximum area of ​​all objects in the detection area is obtained.

[0043] In this step, a single object has multiple distinct area distributions within the detection region. The area with the highest number of pixels is selected as the maximum area of ​​the corresponding object within the detection region. Then, the second mean of the maximum areas of all objects within the detection region is calculated, denoted as... .

[0044] In step S2013, the total area of ​​the detection region is obtained.

[0045] In this step, the total area S of the detection region is obtained.

[0046] In step S2014, the acceptable fluctuation scale of the relevant objects in the prior data layer is obtained based on the first mean, the second mean, and the total area.

[0047] In this step, based on the first mean Second mean And the total area S, to obtain the acceptable fluctuation scale of the relevant objects in the prior data layer i. For example, the acceptable scale of fluctuations in the relevant objects of prior data layer i. It can be obtained from the following formula: Formula 1 in, This represents a non-zero constant, used to avoid the denominator of a fraction being zero. Its empirical value can be 0.01 or 0.001.

[0048] The first mean of the distribution area of ​​the relevant object i in the detection region The larger the ratio of the total area S of the detection region, and the greater the ratio of the second mean, the better. Differences The smaller the size, the larger its area, which allows for identification at lower resolutions, and the greater the acceptable fluctuation scale.

[0049] This process accurately captures the tolerance of the distribution characteristics of objects in the prior data layer to fluctuations, laying the foundation for the scientific calculation of subsequent fluctuation scales and ensuring that the fluctuation scales are consistent with the actual situation of objects in the prior data layer.

[0050] Figure 4 This is a flowchart illustrating a method for obtaining the fluctuation scale corresponding to a priori data layer based on the acceptability of the fluctuation scale, according to an exemplary embodiment. Figure 4 As shown, obtaining the fluctuation scale corresponding to the prior data layer based on the acceptable fluctuation scale may include the following steps: In step S2021, the acceptable fluctuation scale is subjected to maximum and minimum normalization to obtain the normalized acceptable fluctuation scale.

[0051] In this step, the acceptable scale of fluctuation is... Perform max-min normalization to obtain the acceptable normalized fluctuation scale. .

[0052] In step S2022, a preset reference fluctuation scale is obtained.

[0053] In this step, a preset reference fluctuation scale B is obtained.

[0054] In step S2023, the fluctuation scale corresponding to the prior data layer is obtained based on the normalized fluctuation scale acceptability and the reference fluctuation scale.

[0055] In this step, the acceptable level of normalized fluctuation scale is considered. And with reference fluctuation scale B, obtain the fluctuation scale corresponding to the prior data layer i. For example, the fluctuation scale corresponding to the prior data layer i. It can be obtained from the following formula: Formula 2 First, the acceptable fluctuation scale obtained in the previous calculation is normalized by minimax to unify its numerical range. The acceptable normalized fluctuation scale is obtained within a certain range. Then, based on the actual needs of the detection task and industry standards, a reference fluctuation scale is preset. Finally, the final fluctuation scale corresponding to the prior data layer is obtained by calculating the acceptable normalized fluctuation scale and the reference fluctuation scale. Normalization eliminates differences in data dimensions, making the fluctuation tolerance of different objects comparable. Combined with the preset reference value, the practicality of the fluctuation scale is further improved, ensuring its adaptability to subsequent regional boundary analysis.

[0056] The system segments relevant objects in different prior data layers using the mean of the fluctuation scales corresponding to those layers, obtaining the region boundaries of relevant objects in a single prior data layer within the image. For complex and intersecting regions, a boundary-aware segmentation algorithm is used to re-optimize the segmentation boundaries. Then, a buffer is constructed and topological checks are performed to eliminate microscopic gaps and overlaps, resulting in refined region boundaries. These refined region boundaries are then placed within the same image as the detection region for subsequent analysis and recognition of similar boundary groups.

[0057] Figure 5 This is a flowchart illustrating, according to an exemplary embodiment, a method for obtaining the similarity between regional boundaries of different prior data layers using fluctuation scale as weight, and identifying similar boundary groups based on the similarity. Figure 5 As shown, the step of obtaining the similarity between regional boundaries of different prior data layers using the fluctuation scale as weight, and identifying similar boundary groups based on the similarity, may include the following steps: In step S301, the initial similarity between the region boundaries of different prior data layers is obtained.

[0058] In this step, the initial similarity between the region boundaries of different prior data layers is obtained. For example, the minimum distance between any point on the region boundary of the first prior data layer and the corresponding nearest point on the region boundary of the second prior data layer can be obtained first, and the sum of all minimum distances can be obtained. Then, the Fraser distance between the region boundaries of the first and second prior data layers can be obtained. Next, the minimum Fraser distance between all region boundaries of the detection region can be obtained. Finally, based on the sum of all minimum distances, the Fraser distance, and the minimum Fraser distance, the initial similarity between the region boundaries of different prior data layers is obtained.

[0059] In step S302, a corrected similarity is obtained based on the fluctuation scale and the initial similarity, and the corrected similarity is used as the similarity between the regional boundaries of different prior data layers.

[0060] In this step, based on the prior data layer Corresponding fluctuation scale Prior data layer Corresponding fluctuation scale Similarity to the initial Obtain the corrected similarity And will correct the similarity level as For different prior data layers and The degree of similarity between region boundaries. For example, correcting the similarity. It can be obtained from the following formula: Formula 3 in, For normalization, it can specifically be a maximum-minimum normalization function.

[0061] When the value is large, the current boundary similarity representation is larger than the actual boundary similarity representation, so its representation of the original boundary should be increased.

[0062] In step S303, the similarity is subjected to maximum and minimum normalization to obtain a normalized similarity. If the normalized similarity is greater than a first preset threshold, it is determined that the regional boundaries of different prior data layers are in the same similar boundary region.

[0063] In this step, the similarity will be... Perform min-max normalization to obtain the normalized similarity. and in terms of normalized similarity If the value exceeds a first preset threshold (e.g., 0.8), different prior data layers are determined. and The boundaries of the regions are located within the same similar boundary region. The first preset threshold is used to compare similar cases, and its value ranges from 0 to 1. To filter out boundaries with a high degree of similarity, the first preset threshold can be set to 0.8, which can be adjusted by the implementer according to specific circumstances.

[0064] In step S304, similar boundary regions are determined for all region boundaries of the detection area to obtain multiple similar boundary groups.

[0065] In this step, similar boundary regions are determined for all boundaries of the detection area, resulting in multiple similar boundary groups.

[0066] This process fully considers the impact of data fluctuation characteristics on boundary similarity, accurately identifies boundary clusters with similar characteristics, and provides clear analytical units for subsequent analysis of environmental integration.

[0067] Figure 6 This is a flowchart illustrating a method for obtaining the initial similarity between region boundaries of different prior data layers, according to an exemplary embodiment. Figure 6 As shown, obtaining the initial similarity between the region boundaries of different prior data layers may include the following steps: In step S3011, the minimum distance between any point on the region boundary of the first prior data layer and the nearest point on the region boundary of the second prior data layer is obtained, and the sum of all minimum distances is obtained.

[0068] In this step, the first prior data layer is obtained. Any point on the boundary of the region and the second prior data layer Find the minimum distance to the nearest point on the boundary of the region, and obtain the sum of all minimum distances. .

[0069] In step S3012, the Fraser distance between the region boundaries of the first prior data layer and the region boundaries of the second prior data layer is obtained.

[0070] In this step, the first prior data layer is obtained. Region boundaries and the second prior data layer Fraser distance of the regional boundary Fraser distance The smaller the value, the higher the first prior data layer. Region boundaries and the second prior data layer The greater the similarity in shape between the regional boundaries.

[0071] In step S3013, the minimum Fraser distance between the boundaries of all regions in the detection area is obtained.

[0072] In this step, the minimum Fraser distance between the boundaries of all regions in the detection area is obtained. .

[0073] In step S3014, the initial similarity between the region boundaries of different prior data layers is obtained based on the sum of all minimum distances, the Fraser distance, and the minimum Fraser distance.

[0074] In this step, based on the sum of all minimum distances... Frechet distance and minimum Frescher distance Obtain different prior data layers and Initial similarity between region boundaries For example, the initial similarity. It can be obtained from the following formula: Formula 4 in, This represents a non-zero constant, used to avoid the denominator of a fraction being zero. Its empirical value can be 0.01.

[0075] This process comprehensively considers boundary similarity from two dimensions: the sum of all minimum distances at a location and shape similarity. It also references historical minimum Fraser distances for calibration to ensure the accuracy and objectivity of the initial similarity calculation.

[0076] Figure 7 This is a flowchart illustrating a method for obtaining the degree of environmental integration of similar boundary groupings according to an exemplary embodiment. Figure 7 As shown, obtaining the environmental blending degree of the similar boundary grouping may include the following steps: In step S401, the maximum enclosing area formed by all boundaries in the similar boundary group is obtained.

[0077] In this step, the maximum enclosing area formed by all boundaries in the similar boundary group g is obtained. For example, the maximum enclosing area of ​​a group g of single similar boundaries can be obtained using the convex hull algorithm. To obtain.

[0078] In step S402, the third mean of the similarity between all pairs of region boundaries in the similar boundary group is obtained.

[0079] In this step, the third mean of the similarity between all pairwise region boundaries in the similarity boundary group g is obtained. .

[0080] In step S403, the maximum and minimum values ​​of the similarity between all pairs of region boundaries in the similar boundary group are obtained.

[0081] In this step, the maximum similarity between all pairwise region boundaries in the similarity boundary group g is obtained. and minimum value .

[0082] In step S404, the environmental integration degree of the similar boundary grouping is obtained based on the maximum enclosing area, the third mean, the maximum value of the similarity degree, and the minimum value of the similarity degree.

[0083] In this step, based on the maximum enclosing area Third mean The maximum value of similarity and the minimum value of similarity Obtain the degree of environmental integration of similar boundary grouping g. For example, the degree of environmental integration of similar boundary grouping g. It can be obtained from the following formula: Formula 5 in, The maximum value of the maximum enclosed area of ​​all similar boundary groups obtained in history. Not zero, Not zero; This represents a non-zero constant, used to avoid the denominator of a fraction being zero. Its empirical value can be 0.01.

[0084] When the maximum enclosing area of ​​similar boundary grouping g Acquisition from history maximum value ratio Larger, third mean The smaller the value, the higher the similarity. Minimum similarity to historical boundaries The difference The smaller the size, the more chaotic the distribution of environmentally related things within the range of the similar boundary group g, and the greater the degree of environmental integration. The larger the value, the larger the corresponding transition zone.

[0085] This process integrates the spatial range of similar boundary groupings and the similarity features of internal boundaries, accurately depicting the fusion state of different environments within the region, and providing core quantitative indicators for the identification of environmental transition zones.

[0086] Figure 8 This is a flowchart illustrating a method for identifying environmental transition zones and key areas of interest for UAV flight based on the degree of environmental integration, according to an exemplary embodiment. Figure 8 As shown, identifying the environmental transition zone and the key areas of interest for UAV flight based on the degree of environmental integration may include the following steps: In step S501, if the degree of environmental integration is greater than a second preset threshold, the similar boundary group is determined to be an environmental transition zone.

[0087] In this step, the degree of environmental integration If the similarity boundary grouping is greater than the second preset threshold T, the grouping is determined to be an environmental transition zone. For example, the second preset threshold T can be 0.7. The value of the environmental integration degree ranges from 0 to 1. In order to select the environmental transition zone, it is necessary to compare the environmental integration degree and the second preset threshold. The larger the second preset threshold is set, the more reliable the selected environmental transition zone is. Therefore, the second preset threshold can be set to 0.7.

[0088] In step S502, the flight attention level of each location point in the environmental transition zone is obtained based on the degree of environmental integration.

[0089] In this step, the flight attention level of each location point in the environmental transition zone is obtained based on the degree of environmental integration. For example, first, a saliency result map of each environmental type can be obtained based on a saliency detection algorithm. Then, the fourth mean of the values ​​of any point in the environmental transition zone on all saliency result maps is obtained. Next, the absolute value of the difference between the value of any location point on any saliency result map and the fourth mean is obtained. Then, the maximum value of the environmental integration degree of the detection area is obtained. Finally, based on the environmental integration degree of the environmental transition zone, the maximum value of the environmental integration degree, and the absolute value, the flight attention level of any location point in the environmental transition zone is obtained.

[0090] In step S503, based on the flight attention level, the key attention area of ​​the UAV flight is obtained.

[0091] In this step, the flight attention level is based on location point c. This allows us to identify the key areas of focus during drone flight. For example, the min-max normalization method can be used to... Normalization is performed to obtain the normalized flight attention level. Its range is [0,1]. When When the third preset threshold (e.g., 0.8) is set, location point c is determined as the key area of ​​focus for the flight. The set of all location points that meet the above conditions is defined as the key area of ​​focus for the UAV flight. The third preset threshold, used for comparative analysis of flight focus levels, is selected between 0 and 1. A higher flight focus level indicates that the corresponding location point is more likely to be a key area of ​​focus for the flight; therefore, the third preset threshold is set relatively high, such as 0.8.

[0092] In one possible implementation, key areas of interest for the drone flight can be marked in the image and visualized on the staff's work page, so that the staff can make reasonable arrangements for the drone flight path based on the corresponding information of the key areas of interest.

[0093] Based on key areas of interest, monitoring priorities are determined. A terrain-following trajectory is constructed using DEM data to ensure flight safety. Hydrological data is used to avoid water area risks and optimize river corridor patrol paths. Flight altitude is adjusted with reference to vegetation data to ensure identification accuracy. An optimal patrol route that balances monitoring efficiency, flight safety, and data acquisition quality is generated through a multi-constraint path planning algorithm.

[0094] The process first clarifies the scope of the transition zone, and then refines it to specific key locations, providing a precise basis for UAV flight path planning and ensuring UAV flight safety and environmental monitoring quality.

[0095] Figure 9This is a flowchart illustrating a method for obtaining the degree of flight attention at various locations within an environmental transition zone based on the degree of environmental integration, according to an exemplary embodiment. Figure 9 As shown, obtaining the flight attention level of each location point in the environmental transition zone based on the degree of environmental integration may include the following steps: In step S5021, saliency result maps for each environment type are obtained based on the saliency detection algorithm.

[0096] In this step, saliency result maps for each environmental type are obtained based on a saliency detection algorithm. For example, saliency result maps for each environmental type (forest vegetation, wetland invasive species, bird habitat) can be obtained using a saliency detection algorithm. The more pronounced the overall environmental integration within the transition zone, and the smaller the numerical differences between different saliency result maps for a single location, the more complex the environmental characteristics at that location, and this area should be given special attention to ensure flight safety.

[0097] In step S5022, the fourth mean of the values ​​of any point in the environmental transition zone on all significance result maps is obtained.

[0098] In this step, the fourth mean of the values ​​of any point c in the environmental transition zone on all significance result plots is obtained. .

[0099] In step S5023, the absolute value of the difference between the value of any location point on any of the saliency result maps and the fourth mean is obtained.

[0100] In this step, the numerical value of any location point c on any saliency result map a is obtained. Compared with the fourth mean The absolute value of the difference .

[0101] In step S5024, the maximum value of the environmental blending degree of the detection area is obtained.

[0102] In this step, the maximum value of the environmental blending degree in the detection area is obtained. .

[0103] In step S5025, the flight attention level of any location point in the environmental transition zone is obtained based on the degree of environmental integration of the environmental transition zone, the maximum value of the degree of environmental integration, and the absolute value.

[0104] In this step, the degree of environmental integration in the environmental transition zone is considered. The maximum value of environmental integration and absolute value Obtain the flight attention level of any location point c in the environmental transition zone. For example, the degree of flight attention at any location point c in the environmental transition zone. It can be obtained from the following formula: Formula 6 in, Let 'a' represent the total number of significance plots, and 'a' represent the a-th significance plot. Not zero.

[0105] This process meticulously captures the differences in performance of individual locations across different environmental types, accurately quantifies the complexity of the location environment, and provides strong support for the precise identification of key areas of interest.

[0106] This application also provides a computer-readable storage medium storing computer program instructions thereon, which, when executed by a processor, implement the steps of the forest-wetland environment transition zone identification method for unmanned aerial vehicles provided in this application.

[0107] Figure 10 This is a block diagram illustrating a forest-wetland environment transition zone identification system for unmanned aerial vehicles (UAVs) according to an exemplary embodiment. Figure 10 As shown in the figure, this application provides a forest wetland environment transition zone identification system 1000 for unmanned aerial vehicles, including a server 1100.

[0108] Figure 11 This is a block diagram illustrating a server according to an exemplary embodiment. (Refer to...) Figure 11 Server 1100 includes processor 1122, which further includes one or more processors, and memory resources represented by memory 1132 for storing instructions, such as applications, that can be executed by processor 1122. The applications stored in memory 1132 may include one or more modules, each corresponding to a set of instructions. Furthermore, processor 1122 is configured to execute instructions to perform the aforementioned method for identifying forest-wetland transition zones for unmanned aerial vehicles.

[0109] Server 1100 may also include a power supply component 1126 configured to perform power management of server 1100, a communication component 1150 configured to connect server 1100 to a network, and an input / output interface 1158. Server 1100 can operate on an operating system stored in memory 1132.

[0110] In another exemplary embodiment, a computer program product is also provided, comprising a computer program executable by a programmable electronic device, the computer program having a code portion for performing the above-described method for identifying forest-wetland transition zones for unmanned aerial vehicles when executed by the programmable electronic device.

[0111] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these modifications and improvements all fall within the protection scope of this application.

Claims

1. A forest wetland ecotone identification method for a UAV, characterized in that, The method comprises: acquiring a plurality of prior data layers of a detection area, the prior data layers comprising a digital elevation model DEM data layer, a hydrological data layer and a vegetation data layer; acquiring a fluctuation scale corresponding to the prior data layers, the fluctuation scale being used to represent a data fluctuation range allowed by the prior data layers when area division is performed; acquiring a similarity degree between region boundaries of different prior data layers by taking the fluctuation scale as a weight, and identifying a similar boundary grouping based on the similarity degree; acquiring an environmental blending degree of the similar boundary grouping; based on the environmental blending degree, identifying an environmental transition zone and a key attention area for unmanned aerial vehicle flight. 2.The forest wetland ecotone recognition method for a UAV according to claim 1, wherein, The acquiring of the fluctuation scale corresponding to the prior data layers comprises: acquiring a fluctuation scale acceptability of a related object of the prior data layers; acquiring the fluctuation scale corresponding to the prior data layers according to the fluctuation scale acceptability. 3.The forest wetland ecotone recognition method for a UAV according to claim 2, wherein, The acquiring of the fluctuation scale acceptability of the related object of the prior data layers comprises: acquiring a first mean value of a distribution area size of the related object in the detection area; acquiring a second mean value of a maximum area of all objects in the detection area; acquiring a total area of the detection area; acquiring the fluctuation scale acceptability of the related object of the prior data layers according to the first mean value, the second mean value and the total area. 4.The forest wetland ecotone recognition method for a UAV of claim 2, wherein, The acquiring of the fluctuation scale corresponding to the prior data layers according to the fluctuation scale acceptability comprises: performing maximum-minimum normalization processing on the fluctuation scale acceptability to acquire a normalized fluctuation scale acceptability; acquiring a preset reference fluctuation scale; acquiring the fluctuation scale corresponding to the prior data layers according to the normalized fluctuation scale acceptability and the reference fluctuation scale. 5.The forest wetland ecotone recognition method for a UAV of claim 1, wherein, The acquiring of the similarity degree between region boundaries of different prior data layers by taking the fluctuation scale as a weight, and the identifying of the similar boundary grouping based on the similarity degree, comprise: acquiring an initial similarity degree between region boundaries of different prior data layers; acquiring a corrected similarity degree based on the fluctuation scale and the initial similarity degree, and taking the corrected similarity degree as the similarity degree between region boundaries of different prior data layers; performing maximum-minimum normalization processing on the similarity degree to acquire a normalized similarity degree, and determining that region boundaries of different prior data layers are in a same similar boundary region when the normalized similarity degree is greater than a first preset threshold value; performing similar boundary region determination on all region boundaries of the detection area to obtain a plurality of similar boundary groupings. 6.The forest wetland ecotone recognition method for a UAV according to claim 5, wherein, The acquiring of the initial similarity degree between region boundaries of different prior data layers comprises: acquiring a minimum distance between any point on a region boundary of a first prior data layer and a corresponding nearest point on a region boundary of a second prior data layer, and acquiring a sum value of all minimum distances; acquiring a Fréchet distance between the region boundary of the first prior data layer and the region boundary of the second prior data layer; acquiring a minimum Fréchet distance between all region boundaries of the detection area; According to the sum of all minimum distances, the Frechet distance, and the minimum Frechet distance, an initial similarity degree between region boundaries of different prior data layers is obtained. 7.The forest wetland ecotone recognition method for a UAV of claim 1, wherein, The environment blending degree of the similar boundary group is obtained by: Obtaining a maximum enclosed area formed by all boundaries in the similar boundary group; Obtaining a third mean value of the similarity degrees between all pairwise region boundaries in the similar boundary group; Obtaining a maximum value and a minimum value of the similarity degrees between all pairwise region boundaries in the similar boundary group; The environment blending degree of the similar boundary group is obtained according to the maximum enclosed area, the third mean value, the maximum value of the similarity degrees, and the minimum value of the similarity degrees. 8.The forest wetland ecotone recognition method for a UAV of claim 1, wherein, The environment transition zone and the key attention region of the UAV flight are identified based on the environment blending degree, including: In a case where the environment blending degree is greater than a second preset threshold, the similar boundary group is determined as the environment transition zone; Based on the environment blending degree, a flight attention degree of each position point in the environment transition zone is obtained; Based on the flight attention degree, a key attention region of the UAV flight is obtained. 9.The forest wetland ecotone recognition method for a UAV of claim 8, wherein, The flight attention degree of each position point in the environment transition zone is obtained based on the environment blending degree, including: Based on a saliency detection algorithm, a saliency result map of each environment type is obtained; A fourth mean value of the numerical values of any point in the environment transition zone on all saliency result maps is obtained; An absolute value of the difference between the numerical value of any position point on any saliency result map and the fourth mean value is obtained; A maximum value of the environment blending degree of the detection region is obtained; The flight attention degree of any position point in the environment transition zone is obtained according to the environment blending degree of the environment transition zone, the maximum value of the environment blending degree, and the absolute value.

10. A forest wetland ecotone identification system for a drone, comprising: The system comprises a server, and the server comprises: A memory having a computer program stored thereon; A processor configured to execute the computer program in the memory to implement the steps of the method according to any one of claims 1-9.

Images