Ecological landscape restoration region division method and device, and electronic equipment

Abstract

The application provides an ecological landscape restoration area division method and device and electronic equipment, and relates to the technical field of ecological restoration. The method comprises the following steps: determining a plurality of visual nodes based on activity data of a crowd in a target ecological landscape area; determining a plurality of key attention areas according to the plurality of visual nodes, and screening a plurality of initial restoration areas from the plurality of key attention areas based on landscape element data corresponding to each key attention area; calculating a comprehensive priority score of each initial restoration area according to activity data of the crowd and ecological environment background data corresponding to each initial restoration area; and determining the boundary of a target restoration area in the target ecological landscape area according to the comprehensive priority score of each initial restoration area. Whether from the screening of the restoration area or the determination of the boundary of the target restoration area, the visual experience and use willingness of the crowd are considered, and on this basis, the ecological index itself is balanced to improve the ecological landscape restoration effect.

Description

Technical Field

[0001] This invention relates to the field of ecological restoration technology, and in particular to a method, apparatus and electronic equipment for dividing ecological landscape restoration areas. Background Technology

[0002] With the deepening of ecological environmental protection and restoration efforts, ecological landscape restoration has become a key measure to improve the living environment and enhance ecosystem service functions. Currently, the delineation of ecological restoration areas largely relies on baseline ecological and environmental data, such as soil erosion levels, vegetation cover, and the importance of ecological functions, using single or composite ecological indicators to determine the restoration scope.

[0003] However, the inventors discovered that ecological landscapes possess both ecological functions and visual aesthetic value. Existing classification methods generally ignore the core role of people as the main subjects of landscape perception. That is, in large-scale areas with extensive human activity and long-distance visual extension, such as parks, waterfront areas, and mountain trails, the effect of landscape restoration directly affects people's visual experience and willingness to use the landscape.

[0004] The relevant technologies have the following main shortcomings: First, they lack quantitative analysis of the visual perception patterns of people, making it impossible to accurately locate the landscape areas that people focus on, resulting in a disconnect between the investment of restoration resources and the actual needs of people; Second, the existing restoration zones are mostly based on the subjective judgment and static division of designers, without considering the spatial differences in people's activities, such as the differences in the field of vision of different viewing platforms and walkway nodes, resulting in insufficient accuracy in identifying key restoration areas, which in turn affects the effect of ecological landscape restoration. Summary of the Invention

[0005] This invention provides a method, apparatus, and electronic device for dividing ecological landscape restoration areas, in order to solve the problem that the core role of people as the main subjects of landscape perception is not considered when dividing ecological landscape restoration areas, resulting in poor ecological landscape restoration effects.

[0006] In a first aspect, embodiments of the present invention provide a method for delineating ecological landscape restoration areas, including: Based on population activity data in the target ecological landscape area, multiple visual nodes are identified; Multiple key areas of concern are identified based on multiple viewpoints, and multiple initial restoration areas are selected from these key areas of concern based on the landscape element data corresponding to each key area of ​​concern. Based on the activity data of the population and the baseline data of the ecological environment corresponding to each initial restoration area, a comprehensive priority score for each initial restoration area is calculated. Based on the comprehensive priority score of each initial restoration area, the boundary of the target restoration area in the target ecological landscape area is determined.

[0007] In one possible implementation, multiple key areas of interest are determined based on multiple viewpoint nodes, including: In the virtual landscape model corresponding to the target ecological landscape area, a virtual landscape experiment is conducted for each visual field node to obtain eye-tracking data for each visual field node; the eye-tracking data includes the coordinates of the fixation point and the visual field range. For each visual field node, based on the gaze coordinates in the eye movement data of that visual field node, the eye movement data corresponding to that visual field node is mapped to the spatial location corresponding to the virtual landscape model. Based on the eye movement data corresponding to each spatial location under each visual field node, the spatial locations are filtered to obtain the key attention locations; Determine the key areas of focus based on the field of vision corresponding to the key locations.

[0008] In one possible implementation, a comprehensive priority score for each initial restoration area is calculated based on the activity data of the population and the baseline ecological environment data corresponding to each initial restoration area, including: Based on the activity data of the population, the baseline data of the ecological environment, and the visual attention of the population corresponding to each initial restoration area, the activity frequency score, restoration demand score, and visual attention score of each initial restoration area were calculated; among them, visual attention includes fixation duration and fixation times in eye movement data. The scores for population activity frequency, repair needs, and visual attention corresponding to each initial repair area are weighted and fused according to preset weights to obtain the comprehensive priority score of each initial repair area.

[0009] In one possible implementation, the boundary of the target restoration area within the target ecological landscape area is determined based on the comprehensive priority score of each initial restoration area, including: The comprehensive priority score of each initial restoration area is mapped to the landscape virtual model to generate a visualization map, and key restoration areas are identified in the visualization map; the visualization map is used to represent the priority of each initial restoration area that needs to be restored. In the visualization map, adjacent key restoration areas are merged to obtain the boundary of the target restoration area in the target ecological landscape area.

[0010] In one possible implementation, after determining the boundaries of the target restoration areas within the target ecological landscape area based on the comprehensive priority scores of each initial restoration area, the method further includes: In the virtual landscape model corresponding to the target ecological landscape area, restoration simulation is performed based on the boundary of the target restoration area; The ecological landscape in the virtual landscape model after restoration simulation is evaluated.

[0011] In one possible implementation, based on landscape element data corresponding to each key area of ​​concern, multiple initial restoration areas are selected from multiple key areas of concern, including: The landscape element data corresponding to each key area of ​​concern is input into the standard scoring system to obtain the comprehensive demand score corresponding to each key area of ​​concern. Areas with a comprehensive demand score greater than a preset threshold will be designated as the initial repair areas.

[0012] In one possible implementation, the activity data includes activity paths and the frequency of stay at various locations along the activity paths; based on the activity data of the population in the target ecological landscape, multiple visual nodes are identified, including: The overlap rate of activity paths among the population is statistically analyzed to obtain the target activity path; The frequency of dwell time at various locations along the target activity path is statistically analyzed, and locations with a dwell time frequency greater than a preset frequency are designated as visual nodes.

[0013] In one possible implementation, before determining multiple viewpoints based on population activity data in the target ecological landscape area, the method further includes: Collect image data of the target ecological landscape area; Based on image data, construct a virtual landscape model corresponding to the target ecological landscape area.

[0014] Secondly, embodiments of the present invention also provide an ecological landscape restoration area delineation device, comprising: The determination module is used to identify multiple viewpoints based on population activity data in the target ecological landscape area. The filtering module is used to determine multiple key areas of concern based on multiple view nodes, and to filter out multiple initial restoration areas from the multiple key areas of concern based on the landscape element data corresponding to each key area of ​​concern. The calculation module is used to calculate the comprehensive priority score of each initial restoration area based on the activity data of the population and the baseline data of the ecological environment corresponding to each initial restoration area; The partitioning module is used to determine the boundaries of the target restoration areas within the target ecological landscape area based on the comprehensive priority scores of each initial restoration area.

[0015] Thirdly, embodiments of the present invention provide an electronic device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the method described in the first aspect or any possible implementation thereof.

[0016] In this embodiment of the invention, the core role of the public as the main subjects of landscape perception is fully considered. Areas with higher public attention during activities are designated as key areas of focus. Based on these key areas selected according to public attention, ecological indicators, represented by landscape element types, are used to further filter and determine initial restoration areas. During the selection process, public attention is combined with actual ecological standards. Finally, considering that public activity data can reflect their visual experience and willingness to use the area to a certain extent, this embodiment of the invention comprehensively considers public activity data and the ecological indicators themselves to calculate a comprehensive priority score for each initial restoration area. Based on this comprehensive priority score, a multi-dimensional target restoration area division result is obtained, considering public visual experience, willingness to use, and ecological indicators. In summary, the method provided by this embodiment of the invention, whether in selecting restoration areas or determining the boundaries of target restoration areas, takes into account public visual experience and willingness to use the area. On this basis, it balances the ecological indicators themselves to improve the ecological landscape restoration effect. Attached Figure Description

[0017] Figure 1 This is a flowchart illustrating the implementation of the ecological landscape restoration area division method provided in this embodiment of the invention; Figure 2a , 2b These are real-scene images of the target ecological landscape area provided in the embodiments of the present invention; Figure 3a , 3b These are schematic diagrams of eye movement data acquisition provided in embodiments of the present invention; Figure 4a , 4b This is a schematic diagram of eye-tracking data mapping provided in an embodiment of the present invention; Figure 5a , 5b This is a schematic diagram of the key focus area filtering provided in the embodiments of the present invention; Figure 6 This is a schematic diagram of the visualized atlas provided in an embodiment of the present invention; Figure 7 This is a schematic diagram of the boundary of the target repair area provided in an embodiment of the present invention; Figure 8 This is a schematic diagram of the ecological landscape restoration area division device provided in an embodiment of the present invention. Detailed Implementation

[0018] The embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0019] Figure 1 This is a flowchart illustrating the implementation of the ecological landscape restoration area division method provided in this embodiment of the invention. Figure 1 As shown, the method may include: Step 110: Based on the activity data of people in the target ecological landscape area, determine multiple visual nodes.

[0020] In this embodiment, the target ecological landscape area may include urban parks, waterfront landscape areas, mountain or suburban trail areas, urban greenways, mountain restoration landscape areas, campus or community ecological landscape areas, etc. These target ecological landscape areas have actual human activities and usage needs, and there is a need for ecological landscape restoration.

[0021] In this embodiment, the activity data may include the activity path and the frequency of stay at various locations along each activity path; for example, the frequency of stay at viewing platforms, trail turns, rest nodes, etc.

[0022] To identify multiple visual nodes within a target ecological landscape area, the overlap rate of activity paths within the target ecological landscape is first calculated, and paths with high overlap rates are selected as target activity paths. Specifically, the overlap rate of each activity path can be calculated, and paths with overlap rates greater than a preset overlap rate are selected as target activity paths. The resulting target activity path must be at least one activity path.

[0023] The overlap rate of activity paths is used to characterize the degree of spatial sharing among multiple activity paths. The target ecological landscape area is divided into spatial units, and the number of times each spatial unit is traversed by different activity paths is counted to obtain the path traversal frequency distribution. The path segments corresponding to spatial units with a traversal frequency higher than a preset threshold are identified as overlapping paths. The path overlap rate is the proportion of the length of overlapping paths to the total length of all activity paths.

[0024] The frequency of stops at various locations within each activity path is counted independently for each path; that is, the frequency of stops at various locations within that path is counted based on the start and end points of the population. However, the target activity path is essentially a comprehensive path system formed by integrating overlapping sections from these independent activity paths. Therefore, after determining the target activity path, it is also necessary to integrate the frequency of stops at various locations on overlapping activity paths, that is, to count the frequency of stops at various locations on the target activity path and designate locations with a frequency of stops greater than a preset frequency as visual nodes.

[0025] Step 120: Determine multiple key areas of concern based on multiple viewpoints, and select multiple initial restoration areas from the multiple key areas of concern based on the landscape element data corresponding to each key area of ​​concern.

[0026] In this embodiment, the key focus area can be determined by statistically analyzing the fixation duration and number of fixation times under the visual field node. The area with the longer fixation time or the more fixation times is called the key focus area.

[0027] The key focus areas and the field of view nodes are not in a one-to-one correspondence. A field of view node may correspond to multiple key focus areas, or it may not include any key focus areas at all.

[0028] Since the selection of restoration areas also needs to be based on the perspective of ecological restoration, after identifying the key areas of concern, the landscape element data corresponding to each key area of ​​concern should be determined.

[0029] Based on the landscape element data corresponding to each key area of ​​concern, the degree of ecological restoration needs is determined, and key areas of concern with ecological restoration needs greater than the preset needs are selected as the initial restoration areas.

[0030] Step 130: Calculate the comprehensive priority score for each initial restoration area based on the activity data of the population and the baseline data of the ecological environment corresponding to each initial restoration area.

[0031] Although all the identified initial restoration areas have restoration needs, the restoration priorities of different areas are different. Based on the activity data of the population and the baseline data of the ecological environment corresponding to each initial restoration area, this embodiment of the invention evaluates the restoration priority of each initial restoration area and obtains a comprehensive priority score for each initial restoration area. The higher the score, the higher the corresponding restoration priority.

[0032] Step 140: Determine the boundary of the target restoration area in the target ecological landscape area based on the comprehensive priority score of each initial restoration area.

[0033] In this embodiment, there may be cases where the initial restoration areas are adjacent. Therefore, adjacent initial restoration areas with the same restoration priority can be merged, and the boundary of the merged area can be used as the boundary of the target restoration area. In this way, the boundary of the target restoration area in the target ecological landscape area can be determined.

[0034] In summary, the embodiments of this invention fully consider the core role of the public as the main subjects of landscape perception, identifying areas of higher public attention during activities as key areas of focus. Based on these key areas selected according to public attention, ecological indicators, represented by landscape element types, are then used to further filter and determine initial restoration areas. This selection process combines public attention with actual ecological standards. Finally, considering the public's visual experience, willingness to use, and the ecological indicators themselves, a comprehensive priority score is calculated for each initial restoration area. Based on this comprehensive priority score, a multi-dimensional target restoration area division result is obtained, incorporating public visual experience, willingness to use, and ecological indicators. Therefore, the method provided by these embodiments of the invention, whether in selecting restoration areas or determining the boundaries of target restoration areas, takes into account the public's visual experience and willingness to use. On this basis, it balances the ecological indicators themselves to improve the ecological landscape restoration effect.

[0035] In an optional embodiment, step 120, which involves determining multiple regions of interest based on multiple viewpoints, may include: In the virtual landscape model corresponding to the target ecological landscape area, a virtual landscape experiment is conducted for each visual field node to obtain eye movement data corresponding to each visual field node; the eye movement data includes the coordinates of the fixation point and the visual field range.

[0036] For each visual field node, based on the gaze coordinates in the eye movement data of that visual field node, the eye movement data corresponding to that visual field node is mapped to the corresponding spatial location in the virtual landscape model.

[0037] Based on the eye-tracking data corresponding to each spatial location under each visual field node, the spatial locations are filtered to obtain the key locations of focus.

[0038] Determine the key areas of focus based on the field of vision corresponding to the key locations.

[0039] While aerial remote sensing technology is currently used for ecological monitoring and eye-tracking analysis technology is used for landscape visual quality assessment, the two technologies have not yet been integrated. This makes it difficult to simulate the field of vision and focus of attention of people at different spatial locations, such as different locations and heights. To address this issue, this embodiment provides a multi-scale fusion method that combines aerial oblique photography with eye-tracking analysis technology to determine the boundary of the target restoration area within the target ecological landscape area.

[0040] Before fusion, embodiments of the present invention require the acquisition of image data of the target ecological landscape area. The image data acquisition stage can be implemented using aerial oblique photography technology.

[0041] For example, a drone equipped with an oblique photography camera can be used to conduct all-round aerial photography of the target ecological landscape area, including the main activity areas of the population, such as walkways, viewing platforms, and waterfront spaces, to obtain high-resolution image data from multiple angles, such as orthogonal, forward, side, and rear views.

[0042] Figure 2a , 2b These are real-scene images of the target ecological landscape areas provided in the embodiments of the present invention, showing real-scene images of two different target ecological landscape areas. After acquiring image data of these two target ecological landscape areas using aerial oblique photography technology, a three-dimensional point cloud model and texture mapping model of the area are generated using photogrammetry software, realizing a refined three-dimensional virtual reconstruction of the target ecological landscape area and obtaining a virtual landscape model of the target ecological landscape area.

[0043] To facilitate subsequent processing, the virtual landscape model can be simplified, retaining only key landscape elements such as vegetation, water bodies, topography, and man-made structures.

[0044] In the constructed virtual landscape model corresponding to the target ecological landscape area, virtual reality (VR) devices are used to conduct virtual landscape experiments on each visual field node to simulate the real viewing scene of each visual field node and obtain eye movement data corresponding to each visual field node. The eye movement data includes fixation point coordinates, visual field range, fixation duration, saccade path, pupil diameter changes, etc.

[0045] Figure 3a , 3b These are schematic diagrams of eye-tracking data acquisition provided in embodiments of the present invention; wherein, Figure 3a for Figure 2a A schematic diagram of eye movement data acquisition. Figure 3b for Figure 2b The diagram below illustrates eye-tracking data acquisition. The areas marked in the diagram represent the acquisition areas for eye-tracking data, which are defined by visual field nodes.

[0046] For each visual field node, the coordinates of all gaze points in the eye movement data under that visual field node are extracted as spatial registration references. Through coordinate system one and spatial mapping algorithm, the full amount of eye movement data collected for that visual field node, including gaze points, gaze duration, saccade path, etc., is accurately matched and mapped to the corresponding three-dimensional spatial position of the landscape virtual model.

[0047] The specific mapping process is as follows: The process of mapping eye-tracking data to spatial locations in a virtual landscape model includes: Obtain the coordinates of the two-dimensional gaze point acquired by the eye-tracking device. A projection model of the virtual camera is established to describe the mapping relationship between three-dimensional spatial points and two-dimensional image points, and its expression is:

[0048] in, These are the three-dimensional spatial coordinates in the virtual landscape model. This is the camera intrinsic parameter matrix; This is a rotation matrix used to represent the spatial orientation relationship of the observation direction relative to the coordinate system of the virtual landscape model; This is a translation vector used to represent the position of the observation point in three-dimensional space; is a scale factor used to represent the depth scaling relationship of a point in three-dimensional space when projected onto two-dimensional image coordinates.

[0049] The 2D gaze point is normalized, and then converted into a gaze direction vector in the camera coordinate system through the inverse operation of the camera intrinsic parameter matrix. Based on the rotation matrix and translation vector, the gaze direction vector is transformed from the camera coordinate system to the world coordinate system corresponding to the virtual landscape model, obtaining the corresponding spatial gaze. The position where the spatial gaze first intersects the surface of the virtual landscape model is calculated, and the intersection point is used as the mapped position of the eye-tracking gaze point in 3D space.

[0050] Figure 4a , 4b This is a schematic diagram of eye-tracking data mapping provided in an embodiment of the present invention; wherein, for ease of explanation, Figure 4a and Figure 4b yes Figure 2b A schematic diagram of eye-tracking data mapping corresponding to the provided real-world image of the target ecological landscape area.

[0051] Figure 4a This is a schematic diagram of eye-tracking data mapping in two-dimensional pixels. Figure 4b Based on Figure 4a The provided two-dimensional pixel mapping diagram and the derived three-dimensional landscape space eye-tracking data mapping diagram.

[0052] Figures 4a to 4b This can be achieved using a perspective projection algorithm.

[0053] The shades of color in the figure are determined by the visual field nodes and the visual field range in the eye-tracking data corresponding to each visual field node. That is, the more visual field nodes there are, the more overlapping the visual field ranges are, and the darker the color in the corresponding labeled area.

[0054] In this scenario, corresponding eye-tracking data is recorded at the relevant spatial locations. Indicators reflecting the degree of visual attention, such as attention frequency and attention duration percentage, are statistically analyzed from this data. Spatial locations with attention frequency exceeding a preset frequency or attention duration percentage exceeding a preset percentage are designated as key landscape focus locations. The preset frequency and percentage can be determined based on actual needs.

[0055] Based on all the key locations identified through screening, and combined with the spatial distribution characteristics of each location in the landscape virtual model, the corresponding actual field of view coverage is delineated, and the continuous or related field of view is integrated and defined as the key area of ​​focus.

[0056] In an optional embodiment, step 120, which involves selecting multiple initial restoration areas from multiple key concern areas based on the landscape element data corresponding to each key concern area, may include: The landscape element data corresponding to each key area of ​​concern is input into the standard scoring system to obtain the comprehensive demand score for each key area of ​​concern.

[0057] Areas with a comprehensive demand score exceeding a preset threshold will be designated as initial repair areas. The preset threshold is calculated based on historical data and set according to actual needs.

[0058] In this embodiment, image semantic segmentation technology is used to identify landscape element data within each key area of ​​interest. Landscape element data refers to the actual state, attributes, and characteristics of various core landscape components within the key area of ​​interest, covering four main categories: vegetation, water bodies, topography and soil, and artificial structures. For example, it may specifically include the following categories: Vegetation data: vegetation coverage, vegetation growth status, vegetation species composition, and degree of vegetation degradation, such as whether it is withered, exposed, or affected by pests and diseases; Water body data: water transparency, water turbidity, eutrophication status, and surface cleanliness, such as whether there is floating garbage; Topography and soil data: topography integrity, such as whether there is subsidence, damage, or slope slippage, soil moisture content, degree of soil desertification or compaction, and soil erosion; Man-made structure data: the degree of damage, integrity, and cleanliness of structures such as walkways, viewing platforms, landscape stones, and railings, as well as the degree of integration of the structures with the surrounding landscape.

[0059] Figure 5a This is a schematic diagram illustrating the selection of key areas of interest according to an embodiment of the present invention. It uses image semantic segmentation technology to identify and label landscape element data within each key area of ​​interest. Each key area of ​​interest is labeled according to the identified landscape element type. Figure 5a The types of landscape elements identified include topography, vegetation, structures, and trails.

[0060] To simplify the image semantic segmentation results, that is, to simplify Figure 5a This improves the efficiency of landscape element data extraction. Simultaneously, to adapt to the application scenarios of the comprehensive demand scoring system and align with actual restoration assessment logic, the comprehensive demand scoring system in this embodiment only quantitatively evaluates the state of landscape elements within key areas of concern, rather than assessing the overall terrain. The terrain exists only as a spatial carrier of landscape elements. When a portion of the terrain lacks landscape elements, it does not need to participate in the comprehensive demand scoring or initial restoration area selection. Therefore, for... Figure 5a The landscape element data identified was then reduced, and irrelevant terrain was removed to obtain the following result. Figure 5b Among them, irrelevant terrain is visible. Figure 5b The invalid scenarios marked in the document.

[0061] Figure 5b The boundaries of landscape elements are clearer and their distribution is more intuitive, which can reduce the operational costs of subsequent extraction of landscape element data and improve the efficiency and accuracy of data identification and extraction.

[0062] After the landscape element data is extracted, the landscape element data corresponding to each key area of ​​concern is input into the standard scoring system to obtain a comprehensive demand score for each key area of ​​concern. Key areas of concern with comprehensive demand scores greater than a preset threshold are then designated as initial restoration areas.

[0063] In this embodiment, the standard scoring system is obtained by integrating ecological restoration standards and visual appeal. Landscape element data for each key area of ​​concern are input into the standard scoring system one by one, and a comprehensive score for the restoration needs of each key area of ​​concern is obtained through quantitative evaluation and calculation. The following embodiment illustrates the standard scoring system and the calculation process for the comprehensive needs score: For example, this standard's scoring system uses ecological restoration standards and visual appeal as its two core evaluation dimensions, employing a unified quantitative scoring method. The weighted sum of the scores from both dimensions yields a comprehensive score for restoration needs. The core evaluation criteria, specific dimensions, and scoring rules for the two standards are as follows. Scoring can be based on actual project needs using a 0-10 point system, a percentage system, etc. The following explanation uses a 0-10 point system as an example: The ecological restoration standard focuses on the ecological function integrity and natural health status of landscape elements. The core evaluation is whether various landscape elements meet the basic ecological indicator requirements and whether there are ecological degradation problems. It is the core basis for judging the necessity of regional ecological restoration. It mainly uses three core dimensions of vegetation, water body and topography and soil to make quantitative scores. The scores of each dimension are accumulated to form the sub-score of the ecological restoration standard.

[0064] The better the ecological condition of a landscape element, the more complete its ecological functions, and the lower its degree of degradation, the lower its sub-item score; conversely, the worse the ecological condition and the more severe the degradation, the higher its sub-item score.

[0065] Visual appeal focuses on the visual presentation effect of landscape elements and the visual perception experience of the public. Combining the general visual aesthetic preferences of the public, the core evaluation of landscape elements is the visual form, color, cleanliness, and overall landscape harmony. It is the core basis for judging the necessity of visual restoration of a region. It is mainly quantitatively scored from four core dimensions: form integrity, color harmony, visual cleanliness, and landscape layering. The scores of each dimension are accumulated to form the visual appeal sub-score.

[0066] The more complete the visual form of a landscape element, the more harmonious the color scheme, the cleaner the visual appearance of the area, and the richer the overall landscape layers, the lower the score for this sub-item; conversely, the higher the score for a damaged form, jarring colors, visual clutter, or monotonous layers. For example, if a key area of ​​concern has well-formed and undamaged landscape elements, natural vegetation colors that blend with the surrounding environment, no litter, clean structures, and trees, shrubs, and grasses of varying heights creating rich visual layers, this area will score 2 points for visual appeal. However, if a certain area has extensively damaged landscape stones, withered vegetation interspersed with greenery resulting in chaotic colors, floating litter in the waterfront area, and only single herbaceous vegetation without layers, this area will only score 8.5 points for visual appeal.

[0067] Specifically, visual attractiveness is used to quantify the degree to which a target landscape area attracts the visual attention of a crowd. The calculation of visual attractiveness comprehensively considers whether a crowd fixates on the target landscape area and the duration of that fixation. The calculation formula is as follows:

[0068] in, For visual appeal; This represents the total sampling time. The total number of participants in the test; This represents the total number of samples. For the first The first subject was on the During the second sampling, whether to allocate visual attention to the target landscape area, with a value of [value missing]. ; For the first The first subject was on the The time interval between each sampling.

[0069] The comprehensive demand score is calculated by weighting the scores of the ecological restoration standard sub-item score and the visual attractiveness sub-item score according to preset weights. The weights can be flexibly adjusted according to the project positioning of the target ecological landscape area. The higher the overall demand score, the higher the repair demand for the corresponding key areas of concern; the lower the score, the lower the repair demand for the corresponding key areas of concern.

[0070] In an optional embodiment, step 130, which calculates a comprehensive priority score for each initial restoration area based on the activity data of the population and the baseline ecological environment data corresponding to each initial restoration area, may include: Based on the activity data of the population, the baseline data of the ecological environment, and the visual attention of the population corresponding to each initial restoration area, scores for population activity frequency, restoration needs, and visual attention were calculated for each initial restoration area. Visual attention included fixation duration and number of fixations from eye-tracking data.

[0071] The scores for population activity frequency, repair needs, and visual attention corresponding to each initial repair area are weighted and fused according to preset weights to obtain the comprehensive priority score of each initial repair area.

[0072] In this embodiment, activity data of the population corresponding to each initial restoration area, baseline ecological environment data, and visual attention can be constructed into a multi-dimensional data coupling matrix. Within this multi-dimensional data coupling matrix, a population activity frequency score is calculated based on the activity data of the population corresponding to each initial restoration area; a restoration need score is calculated based on the baseline ecological environment data; and a visual attention level score is calculated based on the visual attention level.

[0073] When calculating the comprehensive priority score, based on the population activity data corresponding to the initial restoration area, the core indicators such as the frequency of population activities and path overlap rate of the visual nodes in that area are statistically analyzed, and converted into a population activity frequency score with a unified dimension according to a preset quantification rule. The higher the score, the higher the population activity activity in that area. Based on the ecological environment baseline data corresponding to the initial restoration area, combined with the actual degradation state of landscape elements, the restoration demand score of that area is quantified. The higher the score, the stronger the necessity of ecological restoration in that area. Since the comprehensive demand score of each initial restoration area has already been given in the specific implementation of step 120, this embodiment can also directly use the comprehensive demand score as the restoration demand score.

[0074] The preset quantification rules are standardized calculation rules that convert population activity data, ecological environment baseline data, and visual attention into a unified dimensional score of 0-10, as detailed below: The pre-defined quantitative rules for crowd activity frequency scoring are based on the frequency of crowd activity and path overlap rate of the visual nodes in the initial repair area, assigning interval values ​​accordingly. For example, an activity frequency of ≤2 times / hour is scored as 0-2 points, 2-5 times / hour as 3-5 points, 5-8 times / hour as 6-8 points, and ≥8 times / hour as 9-10 points; a path overlap rate of ≤30% is scored as 0-2 points, 30%-60% as 3-5 points, 60%-90% as 6-8 points, and ≥90% as 9-10 points. The average of the two indicators is taken as the final crowd activity frequency score, with a higher score indicating higher crowd activity.

[0075] The pre-defined quantitative rules for the restoration demand scoring directly adopt the comprehensive demand score as the restoration demand score. The comprehensive demand score is calculated as follows: ecological restoration standard score × 40% + visual attractiveness score × 60% (for densely populated landscapes) or each × 50% (for landscapes that take into account both ecology and environment). Both the ecological restoration standard and visual attractiveness adopt a 0-10 score system. The more severe the ecological degradation and the worse the visual effect, the higher the score and the stronger the restoration demand.

[0076] The pre-defined quantitative rules for visual attention scoring are based on the percentage of fixation time and the average number of fixations in eye-tracking data: a fixation time percentage of ≤10% is scored as 0-2 points, 10%-30% as 3-5 points, 30%-50% as 6-8 points, and ≥50% as 9-10 points; an average number of fixations of ≤2 times is scored as 0-2 points, 2-4 times as 3-5 points, 4-6 times as 6-8 points, and ≥6 times as 9-10 points; the average of the two indicators is taken as the visual attention score, with a higher score indicating higher visual attention.

[0077] The comprehensive priority score weighted fusion rule sums the crowd activity frequency score, repair need score, and visual attention score in a weighted ratio of 3:3:4 to obtain the comprehensive priority score. The calculation formula is: Comprehensive priority score = Crowd activity frequency score × 30% + Repair need score × 30% + Visual attention score × 40%; The higher the score, the higher the repair priority of the corresponding initial repair area.

[0078] The visual attention score is based on the visual attention corresponding to the initial repair area. It extracts fixation duration and fixation frequency from eye movement data as core evaluation indicators. According to preset quantification rules, such as fixation duration ratio and fixation frequency average, it is converted into a visual attention score with a unified dimension. The higher the score, the higher the visual attention the area receives from the population.

[0079] Finally, each score is multiplied by its corresponding weight, and the products are summed to obtain the overall priority score of the initial repair area. The higher the score, the higher the repair priority of the area.

[0080] Optionally, the weighting can be flexibly adjusted according to the positioning of the target ecological landscape area. The standard configuration is 40% for visual attention score, 30% for restoration need score, and 30% for population activity frequency score. If it is a natural ecological landscape area, the weight of the restoration need score can be appropriately increased. If it is a densely populated landscape area such as an urban park or waterfront, the weight of the population activity frequency score can be appropriately increased.

[0081] In an optional embodiment, step 140, which determines the boundary of the target restoration area within the target ecological landscape area based on the comprehensive priority score of each initial restoration area, may include: The comprehensive priority score of each initial restoration area is mapped to the landscape virtual model to generate a visualization map, and key restoration areas are identified in the visualization map; the visualization map is used to characterize the priority of each initial restoration area that needs to be restored.

[0082] In the visualization map, adjacent key restoration areas are merged to obtain the boundary of the target restoration area in the target ecological landscape area.

[0083] Figure 6 This is a schematic diagram of the visualization atlas provided in an embodiment of the present invention. The diagram uses color to indicate the priority of each key repair area. It should be noted here that... Figure 6 Not based on Figure 2a or Figure 2b Generated.

[0084] Figure 6 In the diagram, red indicates extremely high priority, orange indicates high priority, yellow indicates medium priority, and green indicates low priority. Specifically, red and orange represent key areas for repair, green indicates areas that do not require repair, and yellow indicates areas that require repair but are not of high priority.

[0085] In this embodiment, besides using color to indicate the priority of the initial repair areas, a data table can also be used to mark the priority of each initial repair area. For example, numbers 1-3 can be used, with larger numbers indicating higher priorities. In this embodiment, the comparison... Figure 6 The areas corresponding to red and orange can be designated as key areas for repair.

[0086] In the visualized map, based on the priority of key restoration areas, a spatial clustering algorithm is used to merge adjacent key restoration areas to obtain the target restoration area and its boundary, as shown in the following example. Figure 7 As shown.

[0087] Figure 7 The adjacent red and orange areas were merged to form three main target repair areas.

[0088] After determining the boundary of the target repair area, this embodiment can also extract the boundary of the target repair area and output the area coordinates, area, core repair objects and priority to form a complete repair strategy.

[0089] In an optional embodiment, after determining the boundary of the target restoration area within the target ecological landscape area in step 140 based on the comprehensive priority score of each initial restoration area, the method further includes: In the virtual landscape model corresponding to the target ecological landscape area, restoration simulation is performed based on the boundary of the target restoration area.

[0090] The ecological landscape in the virtual landscape model after restoration simulation is evaluated.

[0091] To verify the accuracy of the target ecological landscape restoration area delineation, this embodiment constructs a matching ecological landscape restoration plan in the landscape virtual model corresponding to the target ecological landscape area, based on the delineated target restoration area boundary, combined with the ecological environment baseline data, landscape element degradation characteristics and restoration needs of the area, and simulates the restoration plan in the landscape virtual model to restore the landscape morphology, ecological element status and overall visual presentation effect of the target restoration area after restoration.

[0092] For the ecological landscape in the virtual model of the restored landscape, a comprehensive evaluation is conducted from two dimensions: ecological function and visual aesthetics. The ecological function dimension evaluates the degree of improvement of ecological indicators such as vegetation coverage, water transparency, and soil condition after restoration, while the visual aesthetics dimension evaluates the morphological integrity, color harmony, layering richness, and integration with the surrounding landscape after restoration. At the same time, the evaluation is conducted in conjunction with the visual perception patterns of the population to verify the effect of the restored target area on the visual experience of the population, thus forming a complete evaluation result of the restoration simulation effect.

[0093] The evaluation process can refer to the aforementioned comprehensive needs scoring process, which will not be repeated here. If the evaluation results of the restoration simulation effect meet the preset ecological restoration standards and visual appeal, the restoration plan and the boundary of the target restoration area are determined as the final plan; if the evaluation results do not meet the preset standards, the boundary range of the target restoration area is adjusted or the restoration plan is optimized based on the issues raised in the evaluation feedback, and the restoration simulation and effect evaluation are carried out again until the evaluation results meet the standards.

[0094] In summary, the embodiments of this invention fully consider the core role of the public as the main subjects of landscape perception, identifying areas of higher public attention during activities as key areas of focus. Based on these key areas selected according to public attention, ecological indicators, represented by landscape element types, are then used to further filter and determine initial restoration areas. This selection process combines public attention with actual ecological standards. Finally, considering the public's visual experience, willingness to use, and the ecological indicators themselves, a comprehensive priority score is calculated for each initial restoration area. Based on this comprehensive priority score, a multi-dimensional target restoration area division result is obtained, considering the public's visual experience, willingness to use, and ecological indicators. In conclusion, the method provided by these embodiments of the invention, whether in selecting restoration areas or determining the boundaries of target restoration areas, takes into account the public's visual experience and willingness to use. On this basis, it balances the ecological indicators themselves to improve the ecological landscape restoration effect.

[0095] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0096] The following are device embodiments of the present invention. For details not described in detail, please refer to the corresponding method embodiments described above.

[0097] Figure 8 A schematic diagram of the ecological landscape restoration area division device provided in an embodiment of the present invention is shown. For ease of explanation, only the parts related to the embodiment of the present invention are shown, and are described in detail below: like Figure 8 As shown, the ecological landscape restoration area delineation device 8 includes: Module 81 is used to determine multiple view nodes based on population activity data in the target ecological landscape area; The filtering module 82 is used to determine multiple key areas of concern based on multiple view nodes, and to filter out multiple initial restoration areas from the multiple key areas of concern based on the landscape element data corresponding to each key area of ​​concern. Calculation module 83 is used to calculate the comprehensive priority score of each initial restoration area based on the activity data of the population and the baseline data of the ecological environment corresponding to each initial restoration area; The partitioning module 84 is used to determine the boundary of the target restoration area in the target ecological landscape area based on the comprehensive priority score of each initial restoration area.

[0098] In one possible implementation, the filtering module 82 is specifically used for: In the virtual landscape model corresponding to the target ecological landscape area, a virtual landscape experiment is conducted for each visual field node to obtain eye-tracking data for each visual field node; the eye-tracking data includes the coordinates of the fixation point and the visual field range. For each visual field node, based on the gaze coordinates in the eye movement data of that visual field node, the eye movement data corresponding to that visual field node is mapped to the spatial location corresponding to the virtual landscape model. Based on the eye movement data corresponding to each spatial location under each visual field node, the spatial locations are filtered to obtain the key attention locations; Determine the key areas of focus based on the field of vision corresponding to the key locations.

[0099] In one possible implementation, the computation module 83 is specifically used for: Based on the activity data of the population, the baseline data of the ecological environment, and the visual attention of the population corresponding to each initial restoration area, the activity frequency score, restoration demand score, and visual attention score of each initial restoration area were calculated; among them, visual attention includes fixation duration and fixation times in eye movement data. The scores for population activity frequency, repair needs, and visual attention corresponding to each initial repair area are weighted and fused according to preset weights to obtain the comprehensive priority score of each initial repair area.

[0100] In one possible implementation, module 84 is used specifically for: The comprehensive priority score of each initial restoration area is mapped to the landscape virtual model to generate a visualization map, and key restoration areas are identified in the visualization map; the visualization map is used to represent the priority of each initial restoration area that needs to be restored. In the visualization map, adjacent key restoration areas are merged to obtain the boundary of the target restoration area in the target ecological landscape area.

[0101] In one possible implementation, module 84 is also used for: In the virtual landscape model corresponding to the target ecological landscape area, restoration simulation is performed based on the boundary of the target restoration area; The ecological landscape in the virtual landscape model after restoration simulation is evaluated.

[0102] In one possible implementation, the filtering module 82 is specifically used for: The landscape element data corresponding to each key area of ​​concern is input into the standard scoring system to obtain the comprehensive demand score corresponding to each key area of ​​concern. Areas with a comprehensive demand score greater than a preset threshold will be designated as the initial repair areas.

[0103] In one possible implementation, the activity data includes the activity path and the frequency of stay at each location along the activity path; the determination module 81 is specifically used for: The overlap rate of activity paths among the population is statistically analyzed to obtain the target activity path; The frequency of dwell time at various locations along the target activity path is statistically analyzed, and locations with a dwell time frequency greater than a preset frequency are designated as visual nodes.

[0104] In one possible implementation, the determining module 81 is further used for: Collect image data of the target ecological landscape area; Based on image data, construct a virtual landscape model corresponding to the target ecological landscape area.

[0105] This invention also provides an electronic device, including a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the method described in the above method embodiments.

[0106] In the above embodiments, the descriptions of each embodiment have their own emphasis. Parts not detailed or described in a particular embodiment can be referred to in the relevant descriptions of other embodiments. Unless otherwise specified or in conflict with logic, the terminology and / or descriptions between different embodiments are consistent and can be referenced interchangeably. Technical features in different embodiments can be combined to form new embodiments based on their inherent logical relationships.

[0107] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. A method for delineating ecological landscape restoration areas, characterized in that, include: Based on population activity data in the target ecological landscape area, multiple visual nodes are identified; Multiple key areas of concern are determined based on the multiple visual nodes, and multiple initial restoration areas are selected from the multiple key areas of concern based on the landscape element data corresponding to each key area of ​​concern. Based on the activity data of the population and the baseline data of the ecological environment corresponding to each initial restoration area, a comprehensive priority score for each initial restoration area is calculated. The boundaries of the target restoration areas within the target ecological landscape area are determined based on the comprehensive priority scores of each initial restoration area.

2. The method for dividing ecological landscape restoration areas according to claim 1, characterized in that, The determination of multiple key areas of interest based on the multiple viewpoints includes: In the virtual landscape model corresponding to the target ecological landscape area, a virtual landscape experiment is conducted for each visual field node to obtain eye movement data corresponding to each visual field node; wherein, the eye movement data includes the coordinates of the fixation point and the visual field range; For each visual field node, based on the gaze coordinates in the eye movement data of that visual field node, the eye movement data corresponding to that visual field node is mapped to the spatial location corresponding to the virtual landscape model. Based on the eye movement data corresponding to each spatial location under each visual field node, the spatial locations are filtered to obtain the key attention locations; Based on the field of view corresponding to the key focus location, determine the key focus area.

3. The method for dividing ecological landscape restoration areas according to claim 2, characterized in that, The process involves calculating a comprehensive priority score for each initial restoration area based on population activity data and ecological environment baseline data, including: Based on the activity data of the population, the baseline data of the ecological environment, and the visual attention of the population corresponding to each initial restoration area, the activity frequency score, restoration demand score, and visual attention score of each initial restoration area are calculated; wherein, the visual attention includes the fixation duration and the number of times the eyes are fixed in the eye movement data. The scores for population activity frequency, repair needs, and visual attention corresponding to each initial repair area are weighted and fused according to preset weights to obtain the comprehensive priority score of each initial repair area.

4. The method for dividing ecological landscape restoration areas according to claim 2, characterized in that, The step of determining the boundary of the target restoration area within the target ecological landscape area based on the comprehensive priority score of each initial restoration area includes: The comprehensive priority score of each initial restoration area is mapped to the landscape virtual model to generate a visualization map, and key restoration areas are identified in the visualization map; wherein, the visualization map is used to characterize the priority of each initial restoration area that needs to be restored; In the visualization map, adjacent key restoration areas are merged to obtain the boundary of the target restoration area in the target ecological landscape area.

5. The method for dividing ecological landscape restoration areas according to claim 1, characterized in that, After determining the boundary of the target restoration area within the target ecological landscape area based on the comprehensive priority score of each initial restoration area, the method further includes: In the virtual landscape model corresponding to the target ecological landscape area, restoration simulation is performed based on the boundary of the target restoration area; The ecological landscape in the virtual landscape model after restoration simulation is evaluated.

6. The method for dividing ecological landscape restoration areas according to claim 1, characterized in that, Based on the landscape element data corresponding to each key concern area, multiple initial restoration areas are selected from the multiple key concern areas, including: The landscape element data corresponding to each key area of ​​concern is input into the standard scoring system to obtain the comprehensive demand score corresponding to each key area of ​​concern. The key areas of concern with a comprehensive demand score greater than a preset threshold are designated as the initial repair areas.

7. The method for dividing ecological landscape restoration areas according to claim 1, characterized in that, The activity data includes the activity path and the frequency of stops at various locations along the activity path; Based on the activity data of people in the target ecological landscape, multiple visual nodes are identified, including: The overlap rate of activity paths among the population is statistically analyzed to obtain the target activity path; The frequency of dwell time at various locations along the target activity path is statistically analyzed, and locations with a dwell time frequency greater than a preset frequency are designated as visual nodes.

8. The method for dividing ecological landscape restoration areas according to claim 1, characterized in that, Before determining multiple visual nodes based on population activity data in the target ecological landscape area, the method further includes: Collect image data of the target ecological landscape area; Based on the image data, a virtual landscape model corresponding to the target ecological landscape area is constructed.

9. An ecological landscape restoration area delineation device, characterized in that, include: The determination module is used to identify multiple viewpoints based on population activity data in the target ecological landscape area. The filtering module is used to determine multiple key areas of concern based on the multiple visual nodes, and to filter out multiple initial restoration areas from the multiple key areas of concern based on the landscape element data corresponding to each key area of ​​concern. The calculation module is used to calculate the comprehensive priority score of each initial restoration area based on the activity data of the population and the baseline data of the ecological environment corresponding to each initial restoration area; The partitioning module is used to determine the boundaries of the target restoration areas in the target ecological landscape area based on the comprehensive priority score of each initial restoration area.

10. An electronic device, characterized in that, It includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the method as described in any one of claims 1 to 8.

Images