Landslide spatial position data processing method and system based on dynamic segmentation

Abstract

The invention discloses a landslide spatial position data processing method and system based on dynamic segmentation, and relates to the technical field of data processing, and the method comprises the steps: obtaining a target mountain slope and an observation line, obtaining a GIS module, obtaining a to-be-processed period, and obtaining a surface slope sequence; obtaining fluctuation indexes, and recording the positions of the collection points with the minimum fluctuation indexes corresponding to the first n collection points as ridge lines; absolute variables are obtained, collection points corresponding to the surface slope sequence with the absolute variables exceeding the natural slope fluctuation threshold value are obtained, the collection points serve as initial trailing edge points, and the initial trailing edge point with the highest altitude serves as a landslide trailing edge point; acquiring fluctuation indexes, and recording the position of the collection point with the minimum fluctuation index corresponding to the last n collection points as a valley line; and obtaining a space relative index according to the landslide trailing edge point, the ridge line and the valley line. The method has the advantages of interference resistance, space consistency and comparability.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, specifically to a method and system for processing landslide spatial location data based on dynamic segmentation. Background Technology

[0002] In the field of landslide disaster monitoring and early warning, accurately identifying the spatial location characteristics of landslides (such as ridgelines, landslide trailing edges, and valley lines) is key to assessing slope stability and landslide movement patterns.

[0003] Existing methods often rely on fixed rules (such as directly selecting two locations) or static thresholds for judgment. However, the actual terrain of mountain slopes is complex and varied, which leads to the following problems. Specifically, local surface interference can cause misjudgments of feature points. For example, the front end of the observation line (such as near the starting point) may contain depressions, ponds, or vegetated areas. The surface slope of these areas fluctuates drastically due to non-landslide factors (such as water erosion and root disturbance). If these points are directly used as spatial location features (such as the ridgeline at the top of the slope), noise interference will cause inaccurate positioning, thus distorting the landslide scale analysis. At the same time, the identification of landslide trailing points is affected by natural fluctuations, which usually manifest as abrupt slope changes. However, natural weathering and shallow erosion can also cause local slope fluctuations. Relying solely on a single absolute threshold can easily misjudge fluctuations caused by non-landslide deformation as trailing points, reducing the reliability of detection. Finally, the inconsistent range of feature points leads to statistical bias. The actual positions of ridgelines and valleys vary naturally in multiple landslide events. If an excessively large dynamic segmentation range is used (such as too many collection points before and after), although local interference can be avoided, it will result in excessive dispersion of the feature point positions of each landslide, making it impossible to perform regular statistical analysis under a unified spatial benchmark (such as analysis of concentrated landslide occurrence intervals). Conversely, while fixed endpoints ensure position comparability, they cannot avoid terrain interference and noise. Summary of the Invention

[0004] To address the technical problem that existing technologies lack methods that balance resistance to local interference and spatial comparability, making it difficult to stably generate landslide spatial distribution statistical maps suitable for lateral comparison under complex terrain conditions, this invention provides a landslide spatial location data processing method and system based on dynamic segmentation.

[0005] A method for processing landslide spatial location data based on dynamic segmentation includes: acquiring a target mountain slope and an observation line extending downwards from the top of the target mountain slope; acquiring a GIS module for collecting the slope surface slope at each collection point on the observation line; acquiring the data collection period at which the landslide occurred on the target mountain slope and recording it as the processing period; acquiring the surface slope sequence corresponding to each collection point on the observation line within the processing period based on the GIS module; acquiring the fluctuation index corresponding to each of the first n collection points on the observation line based on the surface slope sequence corresponding to each of the first n collection points on the observation line; and recording the location of the collection point with the smallest fluctuation index among the first n collection points as the ridgeline; acquiring... The absolute variables between adjacent surface slope data in the surface slope sequence of each collection point are used to identify the collection points corresponding to the surface slope sequences with absolute variables exceeding the natural slope fluctuation threshold. These collection points are then used as initial trailing edge points. The initial trailing edge point with the highest elevation is selected from multiple initial trailing edge points and used as the landslide trailing edge point. The fluctuation index corresponding to the last n collection points on the observation line is obtained based on their respective surface slope sequences. The location of the collection point with the smallest fluctuation index is recorded as the valley line. Spatial relative indexes are obtained based on the landslide trailing edge point, ridge line, and valley line. A landslide spatial location statistical map is generated based on the spatial relative indexes corresponding to multiple landslides.

[0006] Optionally, obtaining the spatial relative index corresponding to this landslide based on the landslide trailing edge point, ridgeline, and valleyline includes: obtaining the first distance from the landslide trailing edge point to the ridgeline and the second distance from the landslide trailing edge point to the valleyline; adding the first distance and the second distance to obtain the total distance of the target mountain slope area; and dividing the second distance by the total distance of the target mountain slope area to obtain the spatial relative index.

[0007] Optionally, generating a landslide spatial location statistical map based on the spatial relative indicators corresponding to multiple landslides includes: dividing the range from 0 to 1 into multiple uniform value ranges, and marking the value ranges into which the spatial relative indicators corresponding to multiple landslides fall, so as to form a landslide spatial location statistical map.

[0008] Optionally, obtaining the absolute variable between adjacent surface slope data in the surface slope sequence of each collection point includes: obtaining the difference between adjacent surface slope data in the surface slope sequence of the i-th collection point, taking the absolute value of the difference, and obtaining the absolute variable.

[0009] Optionally, obtaining the fluctuation index corresponding to each of the first n collection points on the observation line based on the surface slope sequence of each of the first n collection points on the observation line includes: defining each of the first n collection points on the observation line as a front-end collection point; dividing each absolute variable in the surface slope sequence of the j-th front-end collection point by the average value of the surface slope data in the surface slope sequence of the j-th front-end collection point, and obtaining each fluctuation ratio in the surface slope sequence of the j-th front-end collection point; and summing all the fluctuation ratios in the surface slope sequence of the j-th front-end collection point to obtain the fluctuation index corresponding to the j-th front-end collection point.

[0010] Optionally, obtaining the fluctuation index corresponding to each of the last n collection points on the observation line based on their respective surface slope sequences includes: defining each of the last n collection points on the observation line as a back-end collection point; dividing each absolute variable in the surface slope sequence of the j-th back-end collection point by the average value of the surface slope data in the surface slope sequence of the j-th back-end collection point, and obtaining each fluctuation ratio in the surface slope sequence of the j-th back-end collection point; summing all the fluctuation ratios in the surface slope sequence of the j-th back-end collection point to obtain the fluctuation index corresponding to the j-th back-end collection point.

[0011] A landslide spatial location data processing system based on dynamic segmentation is also provided. The system includes: an acquisition module for acquiring the target mountain slope and an observation line extending downwards within the target mountain slope, and acquiring a GIS module for collecting the slope surface slope at each collection point on the observation line, acquiring the data acquisition period at which the landslide occurred and recording it as the processing period, and acquiring the surface slope sequence corresponding to each collection point on the observation line within the processing period based on the GIS module; a first data processing module for acquiring the fluctuation index corresponding to each of the first n collection points on the observation line based on the surface slope sequence corresponding to each of the first n collection points on the observation line, and recording the location of the collection point with the smallest fluctuation index corresponding to each of the first n collection points as the ridgeline; and a second data processing module. The first module is used to obtain the absolute variables between adjacent surface slope data in the surface slope sequence of each collection point, identify the collection points corresponding to the surface slope sequences with absolute variables exceeding the natural slope fluctuation threshold, and use these collection points as initial trailing edge points. The module then selects the initial trailing edge point with the highest elevation from multiple initial trailing edge points and uses it as the landslide trailing edge point. The second module is used to obtain the fluctuation index corresponding to the last n collection points on the observation line based on their respective surface slope sequences, and to record the location of the collection point with the smallest fluctuation index as the valley line. The third module is used to obtain spatial relative indicators based on the landslide trailing edge point, ridge line, and valley line, and to generate a landslide spatial location statistical map based on the spatial relative indicators corresponding to multiple landslides.

[0012] Optionally, the spatial statistics module is also used to: obtain a first distance from the landslide trailing edge to the ridgeline, and a second distance from the landslide trailing edge to the valley line; add the first distance and the second distance to obtain the total distance of the target mountain slope area; divide the second distance by the total distance of the target mountain slope area to obtain a spatial relative index.

[0013] Optionally, the spatial statistics module is also used to: divide the range from 0 to 1 into multiple uniform value ranges, and mark the value ranges into which the spatial relative indicators corresponding to multiple landslides fall, so as to form a landslide spatial location statistics map.

[0014] Optionally, the second data processing module is further configured to: obtain the difference between adjacent surface slope data in the surface slope sequence of the i-th sampling point, calculate the absolute value of the difference, and obtain an absolute variable.

[0015] The beneficial effects of this invention are reflected in: In the entire landslide spatial location data processing method based on dynamic segmentation, firstly, by limiting the number of small collection points before / after and introducing a fluctuation index to quantify stability (calculating the relative fluctuation ratio), the sensitivity of the fixed endpoint method to local terrain noise (such as the starting depression and the ending alluvial area) is overcome, ensuring that the ridgeline / valleyline is always anchored in the stable area near the top / toe of the slope (within a range of approximately 50 meters / 30 meters), thus eliminating statistical bias caused by differences in the spatial span of feature points from the source; furthermore, a two-level filtering strategy of temporal and spatial levels is adopted. To identify the trailing edge of a landslide, the initial point of a slope change is first screened by using a natural slope fluctuation threshold calibrated based on historical data. Then, secondary deformation interference is eliminated based on the physical laws of the highest altitude, significantly improving the identification accuracy of the top boundary of the main sliding surface under complex terrain. Furthermore, the cross-slope data is normalized by using spatial relative indicators (distance from the trailing edge to the valley line / total slope length), converting absolute positional differences into standard values ​​of 0 to 1. Combined with zonal statistical maps (e.g., R=0.6~0.8 for the middle and upper parts of the mountain), the regional landslide distribution pattern is intuitively revealed. Attached Figure Description

[0016] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the accompanying drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. In all the drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, the elements or parts are not necessarily drawn to scale.

[0017] Figure 1 This is a partial flowchart of S1 to S5 in the landslide spatial location data processing method based on dynamic segmentation of the present invention. Figure 2 This is a partial flowchart of S5 in the landslide spatial location data processing method based on dynamic segmentation of the present invention; Figure 3 This is a schematic diagram illustrating the steps of the landslide spatial location data processing method based on dynamic segmentation of the present invention; Figure 4 This is a schematic diagram of a portion of step S2 in the landslide spatial location data processing method based on dynamic segmentation of the present invention; Figure 5 This is a schematic diagram of a portion of step S4 in the landslide spatial location data processing method based on dynamic segmentation of the present invention; Figure 6 This is a schematic diagram of part of step S5 in the landslide spatial location data processing method based on dynamic segmentation of the present invention; Figure 7 This is a landslide spatial location comparison map in the landslide spatial location data processing method based on dynamic segmentation of the present invention. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0019] Therefore, the following detailed description of the embodiments of the invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the invention without inventive effort are within the scope of protection of the invention.

[0020] It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, the terms "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0021] like Figures 1 to 3 , Figure 7 As shown, a method for processing landslide spatial location data based on dynamic segmentation is provided. In one embodiment, the method includes: S1. Obtain the target mountain slope and the observation line located within the target mountain slope and extending from top to bottom, and obtain the GIS module used to collect the slope surface slope of each collection point on the observation line. Obtain the data collection period when the landslide occurred on the target mountain slope and record it as the processing period. Based on the GIS module, obtain the surface slope sequence corresponding to each collection point on the observation line within the processing period. S2. Based on the surface slope sequence corresponding to the first n collection points on the observation line, obtain the fluctuation index corresponding to each of the first n collection points on the observation line, and record the location of the collection point with the smallest fluctuation index corresponding to each of the first n collection points as the ridge line. S3. Obtain the absolute variables between adjacent surface slope data in the surface slope sequence of each collection point, obtain the collection point corresponding to the surface slope sequence with absolute variables exceeding the natural slope fluctuation threshold, and take the collection point as the initial trailing edge point. Select the initial trailing edge point with the highest altitude from multiple initial trailing edge points and take it as the landslide trailing edge point. S4. Based on the surface slope sequence corresponding to the last n collection points on the observation line, obtain the fluctuation index corresponding to each of the last n collection points on the observation line, and record the location of the collection point with the smallest fluctuation index corresponding to each of the last n collection points as the valley line. S5. Obtain spatial relative indicators based on the landslide trailing edge, ridgeline, and valleyline, and generate a landslide spatial location statistical map based on the spatial relative indicators corresponding to multiple landslides.

[0022] In this embodiment, it should be noted that in S1, the landslide trailing edge point is identified—that is, the boundary between the top of the landslide body and the stable slope. This point is usually characterized by a sharp abrupt change in the surface slope (such as a sudden steepening caused by rock and soil fracturing).

[0023] Its implementation relies on the dynamic analysis of the surface slope sequence generated in the previous steps: First, in stage S1, the GIS module acquires the surface slope data of each collection point on the observation line at continuous time points within a specific period (the processing period), forming a temporal change sequence for each point. In stage S3, based on this, the absolute variable (i.e., the absolute value of the slope change between two adjacent collections) of the surface slope sequence of each collection point is calculated independently for each collection point. This variable reflects the intensity of fluctuations in the local slope morphology over a short period. This design directly addresses the problem of terrain noise interference—while natural weathering or shallow erosion may cause small fluctuations in slope, the changes between adjacent time points are usually small and discrete; however, rock shearing or collapse during the formation of the landslide trailing edge can trigger significant, discontinuous slope jumps between adjacent collection points. Therefore, the absolute variable is filtered through a preset natural slope fluctuation threshold (the upper limit of normal environmental fluctuations obtained based on historical data analysis): only collection points where at least one adjacent variable exceeds this threshold are marked as initial trailing edge points. This mechanism quantifies "slope mutations" into comparable physical indicators, avoiding misjudgments caused by traditional methods that rely solely on single slope values ​​or static segmentation rules (such as mistaking local depressions caused by animal burrows for the rear edge of a landslide).

[0024] Furthermore, the initial trailing edge point selection results may include multiple candidate points (e.g., rock blocks from different locations sliding down in stages within the same landslide event), thus requiring further optimization based on spatial topological relationships. The requirement is to select the highest elevation among all initial trailing edge points as the final landslide trailing edge point. This stems from the physical laws of landslide movement: the trailing edge point, as the top boundary of the landslide body, is usually located at a relatively high position on the slope under the influence of gravity (distinguishing it from atypical deformation points formed in the middle and lower parts due to local erosion or secondary sliding).

[0025] For example, in a bedrock landslide, the top of the slope may generate the first sampling point A where the absolute variable exceeds the threshold due to the expansion of the main crack, while a sampling point B may be generated somewhere in the middle or lower part due to the peeling of surface soil (although it meets the initial trailing edge condition, it is actually a secondary deformation). By comparing the elevation of A and B, the higher A is determined to be the true trailing edge point.

[0026] This operation implicitly incorporates two key design elements: First, by coupling temporal variation characteristics (absolute variable exceeding the threshold) with spatial location attributes (altitude), atypical interference signals in low-lying areas (such as surface soil creep caused by rainfall) are eliminated. Second, the determination of the trailing edge point ensures spatial consistency with the ridgeline and valley line determined by S2 / S4—the trailing edge point must be located in a gravity-sensitive area below the ridgeline (stable slope top) and above the valley line (stable slope toe), so that the spatial relative index (the proportion of the distance from the landslide trailing edge to the valley line) generated in S5 can accurately reflect the relative position of the landslide on the slope profile (such as sliding in the upper middle or lower part), laying the foundation for horizontal comparison of statistical charts. This layered filtering strategy (initial screening based on temporal fluctuation characteristics, followed by fine screening based on spatial location) significantly improves the accuracy of identifying the true trailing edge of a landslide in complex terrain.

[0027] In S2, the ridgeline representing the stable boundary of the slope top is determined. Its design directly addresses the shortcomings of the fixed endpoint method: if the starting point of the observation line is directly selected as the ridgeline, the slope sequence may fluctuate drastically due to local terrain interference (such as the presence of seasonal waterlogged depressions at the starting point), thereby distorting the actual position of the slope top.

[0028] To address this, S2 introduces a dynamic segmentation strategy—limiting the search range to a finite number of acquisition points at the top of the observation line (typically the first 2-3 points), and evaluating the stability of each point within this small range using a fluctuation index. This index essentially quantifies the noise sensitivity of the slope sequence: first, based on the surface slope sequence generated by S1, the absolute variable of the slope change between adjacent times for each acquisition point is calculated (reflecting the intensity of short-term fluctuations). Then, all absolute variables are compared with the mean of the slope sequence at that point to calculate a series of relative fluctuation ratios (eliminating biases caused by different slope base values). Finally, these ratios are summed to generate the fluctuation index; the smaller the value, the lower the relative amplitude of the slope change over time at that point (i.e., the stronger the anti-interference ability).

[0029] For example, on an observation line extending along a ridge, if the first sampling point is located in a shallow depression prone to water accumulation, the repeated wet and dry cycles during the rainy season cause its slope sequence to oscillate wildly between 20° and 5° (extremely high volatility index); while the third sampling point is located on a stable slope crest with exposed bedrock, the slope sequence consistently fluctuates within a narrow range of 32° to 35° (extremely low volatility index). By comparing the volatility index of the first n points, the third point is automatically identified as the true ridgeline location. This mechanism avoids local interference at the starting point and, because the value of n is small (only covering the area near the slope crest), ensures the spatial comparability of the ridgeline during multiple landslide analyses.

[0030] Furthermore, S2's dynamic segmentation design balances noise resistance and statistical consistency requirements through a two-layer constraint: First, by using fluctuation index screening, it avoids the problem in existing methods where fixed endpoints are limited by local terrain noise (such as vegetation root disturbance or animal nests), leading to the mislabeling of atypical locations as ridgelines. For example, when there are two candidate points at the front of the observation line—point A is located in a fractured rock zone with strong wind erosion (the diurnal temperature variation causes the rock mass to expand and contract, resulting in periodic slope fluctuations), and point B is located in a stable soil layer consolidated by tree roots (the slope remains stable over a long period)—by comparing the fluctuation indices of the two, point B will be preferentially selected as the ridgeline.

[0031] Secondly, limiting n to a small range (e.g., 3 points) ensures that the ridgeline location of each landslide event does not deviate too far from the actual main ridgeline at the top of the slope. If n is too large (e.g., the first 10 points), a secondary platform hundreds of meters away from the top of the mountain may be misjudged as a ridgeline, resulting in a huge spatial span of "ridgelines" in different landslide events (e.g., landslide A is on the mountaintop platform, landslide B is on the shoulder plateau), which would destroy the comparability of subsequent S5 spatial relative indicators. This constraint anchors the ridgelines of all landslides within a limited elevation difference zone in the upper part of the slope (usually within 50 meters of the top of the slope), laying the foundation for horizontal comparison of different landslide locations (e.g., concentrated in the middle and upper parts of the mountain) in the S5 statistical chart. Meanwhile, the accurate positioning of the ridgeline directly affects the interpretation of the landslide trailing edge point in S3: the trailing edge point must be located below the ridgeline (to avoid incorrectly associating natural slope top deformation with landslide events), and the spatial relative index in S5 (the proportion of the distance from the ridgeline to the trailing edge point) truly reflects the cutting depth of the sliding surface in the slope profile (such as the distribution pattern of shallow and deep landslides), ultimately supporting the statistical chart to reveal the regional landslide activity pattern.

[0032] In S3, by analyzing the time-series changes in the slope surface, the trailing edge point of the landslide, representing the failure boundary at the top edge of the landslide, is accurately identified. Its design directly addresses the problem of slope fluctuation interference caused by non-landslide factors in natural terrain: since natural weathering, rainfall erosion, or biological activities (such as animal burrows) may cause local slope changes, relying solely on a single measurement or a fixed threshold is prone to misjudgment.

[0033] Therefore, S3 first calculates the absolute value of the slope change within adjacent observation periods (called the absolute variable) for each point based on the surface slope sequence obtained by S1 (slope dataset of the same collection point at different time points). This variable essentially reflects the stability of the slope morphology in the short term. Slope changes caused by natural fluctuations usually exhibit limited and continuous fluctuation characteristics (such as fluctuations of 0.5° to 2° caused by daily temperature differences). However, when the rear edge of a landslide forms, the fracturing or shearing of the soil and rock mass can trigger sudden, non-gradual steep changes (such as a sudden increase of 15° within a single monitoring period). Therefore, S3 performs initial screening using a preset natural slope fluctuation threshold: only when the absolute variable in the sequence of a collection point exceeds this threshold at least once is it marked as an initial rear edge point. When determining the natural slope fluctuation threshold, firstly, historical monitoring data of the surface slope of the target slope in a limited number of stable states (without landslide events) are selected, covering multiple complete natural cycles (such as the alternation of dry and rainy seasons, freeze-thaw cycles, etc.) to ensure that fluctuations caused by various natural disturbance factors (rainfall, weathering, vegetation growth, etc.) are included. For each collection point, the absolute variable of the slope change at adjacent time points (i.e., the absolute value of the adjacent difference in S3) is calculated to form an absolute variable dataset for all collection points. Then, the high percentile value of this absolute variable dataset is taken as the natural slope fluctuation threshold (e.g., 95% or 99% quantile), that is, only a maximum of 5% (or 1%) of natural fluctuation samples are allowed to exceed this natural slope fluctuation threshold. For example, if the maximum value of the absolute variable in a non-landslide period in a certain area is 10°, but 95% of the samples are below 5°, then the natural slope fluctuation threshold can be set to 5°.

[0034] For example, in a clay slope landslide, the surface soil undergoes slow creep due to rainfall (with all absolute variables less than the threshold), while the expansion of the main crack at the top causes a sharp increase in the slope at a certain point (exceeding the threshold). This point is then identified as the initial trailing edge. This mechanism separates physical damage from background noise, avoiding misjudging local depressions such as animal nests or shallow erosion as the landslide top boundary.

[0035] Furthermore, the initial trailing edge points may exhibit spatial dispersion (e.g., multiple locations on the slope simultaneously experiencing local deformation), necessitating the precise location of the dominant landslide trailing edge using spatial topological rules. The initial trailing edge point with the highest elevation is forcibly selected as the final result. This approach is based on the physical laws of landslide movement: the top boundary of the main landslide body inevitably occupies a relatively high position on the slope under the influence of gravity, while the points exceeding the threshold in the middle and lower parts often originate from secondary slippage or local collapse (not the main landslide boundary).

[0036] For example, the cracks at the top of the bedrock slope (at 120 meters above sea level) and the slippage in the middle rock debris accumulation area (at 80 meters above sea level) both meet the initial trailing edge conditions, but only the high-level cracks represent the main trailing edge of the landslide. This rule implies two core logics: First, by coupling the temporal abrupt change characteristics (exceeding threshold verification) with spatial elevation attributes, atypical signals in the low-level area (such as micro-slippage of soil caused by seepage at the toe) are excluded, ensuring that the trailing edge point represents overall failure; Second, maintaining spatial logical consistency between steps—the selected trailing edge point must be located below the ridgeline (stable slope top) determined in S2 and above the valleyline (stable slope toe) determined in S4. This positional characteristic gives the spatial relative index calculated in S5 (the ratio of the distance from the trailing edge point to the valleyline to the total distance from the slope top to the slope toe) physical meaning. For example, when the trailing edge point is close to the ridgeline, the index value is close to 1.0 (the landslide occurred at the slope top), and conversely, when it is close to the valleyline, the index approaches 0 (the landslide occurred at the slope toe). This indicator directly supports the horizontal comparison of different landslide locations in the statistical chart (e.g., landslides in the analysis area are concentrated in the upper or lower part of the mountain), while the elevation screening of S3 ensures the spatial comparability of the indicator. The entire process improves the robustness of trailing point identification in complex terrain through a hierarchical strategy of "temporal initial screening - spatial fine screening" and avoids the sensitivity of the single threshold method to local noise.

[0037] In S4, the valley line representing the stable boundary at the toe of the slope is determined. Its design addresses the shortcomings of the traditional method of directly observing the end of the observation line. The toe area is susceptible to water erosion, human activities, or local deposits. If the end point is located in such a sensitive area (such as a seasonal stream valley), its slope sequence will fluctuate drastically, leading to inaccurate positioning of the valley line.

[0038] Therefore, S4 adopts the dynamic segmentation logic of S2 but focuses on the tail end of the observation line: it is limited to the last n collection points (usually the last 2-3 points), and the anti-interference ability of each point is evaluated through the fluctuation index. The calculation method of this index is the same as that of S2: based on the surface slope sequence generated by S1, the absolute variable of the slope change between adjacent times of each point is first calculated (reflecting short-term instability), then all absolute variables are compared with the average slope of that point to obtain the relative fluctuation ratio (eliminating the influence of slope base difference), and finally the cumulative ratio value is used to obtain the fluctuation index (the smaller the value, the more stable the long-term shape of the point).

[0039] For example, on an observation line extending along a valley, the terminal point is located on a floodplain susceptible to erosion, where the slope sequence oscillates between 5° and 25° during the rainy season (high volatility index); while the penultimate point is located at a stable valley floor where bedrock is exposed, with the slope consistently maintaining a slight variation of 10° to 12° (low volatility index). By comparing the volatility indices of the last n points, the penultimate point is automatically identified as the true valley line. This strategy avoids local interference at the terminal point and, due to the small value of n (covering only the area near the toe of the slope), ensures the spatial consistency of the valley line in multiple landslide analyses.

[0040] Furthermore, S4 balances the requirements of terrain adaptability and statistical comparability through the collaborative design of dynamic segmentation and stability quantification: First, the fluctuation index screening can eliminate interference points caused by abnormal slope fluctuations due to non-slope factors (such as slope foot farming disturbance and temporary loading).

[0041] For example, if there are two candidate points at the end—point C is located in the fill engineering area (where the slope frequently changes due to construction machinery compaction), and point D is located at the bottom of the original bedrock valley (where the slope changes only slightly due to natural variations)—point D will be selected as the valley line based on the fluctuation index. Secondly, limiting n to a small range (such as 2 points) prevents the valley line from deviating too far from the actual main valley line at the toe of the slope.

[0042] If n is too large (as in the last four points), secondary depressions (such as alluvial fans at the foot of mountains) hundreds of meters away from the slope may be misjudged as valley lines, leading to significant differences in the spatial span of the "valley lines" for different landslide events (e.g., landslide C is in the main river channel, while landslide D is at the tributary mouth), thus undermining the basis for lateral comparison of the S5 spatial relative index (R value). This constraint anchors the valley lines of all landslides within a limited range at the lower part of the slope (usually extending 30 meters from the slope toe), providing a unified spatial reference system for statistical analysis.

[0043] The accurate positioning of the valley line is crucial for S5: 1) As the lower endpoint of the landslide boundary, it needs to form an effective slope segment with the trailing edge point identified by S3 to ensure that the physical meaning of the R value (distance from the trailing edge point to the valley line / total slope length) is clear; 2) Its location directly affects the landslide scale classification (e.g., R=0.2 indicates that the landslide occurred in the lower part of the mountain). If the temporary accumulation area is mistakenly identified as the valley line, it will lead to an underestimation of the R value and incorrect classification as a toe landslide. The collaboration between S4, S2, and S3 ensures the robustness of the entire landslide spatial location data chain.

[0044] In S5, a standardized index (spatial relative index R) is generated by quantifying the relative position of the landslide trailing edge point on the slope profile. Its calculation process closely depends on the key feature points determined in the preceding steps: 1) the ridgeline (slope top stability boundary) determined in S2; 2) the landslide trailing edge point (sliding top boundary position) identified in S3; and 3) the valley line (slope toe stability boundary) determined in S4.

[0045] The specific implementation includes three stages: First, based on the spatial analysis function of the GIS module, the first distance from the landslide trailing edge to the ridgeline (reflecting the spatial offset between the trailing edge and the top of the slope) and the second distance to the valley line (reflecting the proximity between the trailing edge and the foot of the slope) are obtained respectively; Second, the first distance and the second distance are added together to obtain the total length of the slope, representing the overall span from the top of the slope to the foot of the slope; Finally, by dividing the second distance by the total length of the slope through standardized calculation, a spatial relative index R with a value between 0 and 1 is generated.

[0046] The spatial relative index R has a clear physical meaning: the smaller the R value, the closer the landslide's trailing edge is to the slope's toe (e.g., R=0.2 indicates a landslide in the lower part of the mountain); the larger the R value, the closer it is to the slope's crest (e.g., R=0.8 indicates a landslide in the upper part of the mountain). For example, if the trailing edge of a landslide is close to the valley line (the second distance is extremely short), its R value approaches 0, indicating that the landslide occurred in the slope's toe area; conversely, if the trailing edge is close to the ridge line (the second distance is close to the total slope length), the R value approaches 1.0, indicating that the landslide originated at the slope's crest. This design solves the problem of incomparable absolute locations—the differences in absolute length between different slopes, after being converted by relative proportions, allow for the evaluation of landslide location patterns under a unified standard.

[0047] Furthermore, the statistical analysis phase of S5 transforms the spatial relative index R into an intuitive statistical map of landslide spatial locations: First, the range of values ​​from 0 to 1 is evenly divided into several intervals (typically divided into 7 levels: R<0.1 is the bottom of the slope, 0.8≤R<1.0 is the upper part of the mountain, etc.), with each interval corresponding to a specific slope landform location; then, the R values ​​of multiple landslide events are classified, the frequency of their falling into each interval is counted, and a distribution map is generated.

[0048] This statistical chart offers three key technical benefits: First, it allows for horizontal comparison of concentrated areas of different landslide events. For example, if 80% of landslides in a region have R-values ​​concentrated in the 0.6-0.8 range (upper and middle parts of the mountain), it reveals that landslides in this area are prone to occur in the middle sections of steep slopes. Second, it verifies the accuracy of previous location steps. If an R-value for a landslide is abnormal (e.g., R-value exceeds 1.0 at the trailing edge), it may indicate a misjudgment of characteristic points in S3 or S4 (e.g., the valley line is located too high), prompting data review. Third, it supports disaster prevention and mitigation decisions. The chart can visually display high-incidence landslide locations in a region (e.g., concentrated in the lower part of the mountain), guiding the targeted deployment of monitoring equipment or reinforcement projects.

[0049] The entire process relies on the dynamic segmentation of the preceding steps—the small-range n values ​​of S2 / S4 ensure that the ridgeline / valleyline of all landslides is located in the vicinity of the slope top / toe (avoiding absolute positional discrepancies), and the high-altitude screening of S3 ensures that the trailing edge is located at the top of the main sliding surface. The three work together to make the R value statistically comparable across events. For example, comparing two landslides: landslide X (total slope length 200 meters, trailing edge 40 meters from the toe, R=0.2) and landslide Y (total slope length 500 meters, trailing edge 100 meters from the toe, R=0.2), although their scales are vastly different, they can be classified as lower-level landslides of the same type for pattern analysis because their R values ​​are the same.

[0050] In summary, the entire landslide spatial location data processing method based on dynamic segmentation firstly overcomes the sensitivity of the fixed endpoint method to local terrain noise (such as the starting depression and the ending alluvial area) by limiting the number of small collection points before / after and introducing a fluctuation index to quantify stability (calculating the relative fluctuation ratio). This ensures that the ridgeline / valleyline is always anchored in the stable area near the top / toe of the slope (approximately 50 meters / 30 meters), eliminating statistical bias caused by differences in the spatial span of feature points from the source. Furthermore, a two-level processing method using both temporal and spatial dimensions is employed. The filtering strategy identifies landslide trailing edges by first filtering initial points of slope abrupt changes using a natural slope fluctuation threshold calibrated based on historical data, and then eliminating secondary deformation interference based on the physical laws of the highest elevation, significantly improving the identification accuracy of the top boundary of the main sliding surface in complex terrain. Furthermore, it normalizes cross-slope data using spatial relative indicators (distance from trailing edge to valley line / total slope length), transforming absolute positional differences into standard values ​​of 0-1. Combined with zonal statistical maps (e.g., R=0.6-0.8 for the middle and upper parts of the mountain), it visually reveals the regional landslide distribution patterns. In summary, the entire method enables lateral comparability of ridgelines, trailing edges, and valley lines of multiple landslide events under a unified benchmark, supporting the reliability of disaster pattern analysis. Simultaneously, the entire process requires no manual intervention in threshold setting or segmentation rule adjustment, adapting to complex geological environments.

[0051] like Figure 6 As shown, in one embodiment, S5 involves obtaining the spatial relative indices corresponding to this landslide based on the landslide trailing edge, ridgeline, and valleyline, including: S51. Obtain the first distance from the landslide trailing edge to the ridgeline, and obtain the second distance from the landslide trailing edge to the valleyline; S52. Add the first distance and the second distance to obtain the total distance of the target mountain slope area; S53. Divide the second distance by the total distance of the target mountain slope area to obtain the spatial relative index.

[0052] In this embodiment, it should be noted that in S51, the first distance—the spatial offset from the landslide's rear edge point to the ridgeline—is calculated using the GIS module. This distance reflects the degree of separation between the sliding top boundary and the stable slope top. The calculation requires precise measurement along the observation line's surface trajectory (to avoid straight-line projection errors). For example, the distance is shorter when the rear edge point is close to the ridgeline (e.g., the main crack is adjacent to the slope top), and longer when it is far away (e.g., the top boundary of a deep landslide shifts downwards). This parameter provides a positioning reference for the landslide in the slope top direction for the subsequent R value. If the distance is too long, it may indicate a deviation in ridgeline identification (e.g., S2 mistakenly identifies the second highest point as the slope top).

[0053] In S52, the total slope length is obtained by adding the first distance and the second distance (distance from the trailing edge to the valley line), representing the complete cross-sectional span from the stable slope crest (ridgeline) to the stable slope toe (valley line). This design eliminates local topographic interference: for example, the valley line is located at the bottom of the original bedrock valley due to S4 avoiding temporary soil dumping areas, and the total length includes the actual slope range. This value standardizes different slope sizes—landslides with total lengths of 200 meters and 500 meters—and the locational patterns can only be compared when combined with the R value.

[0054] In S53, the spatial relative index R (0~1) is generated by dividing the second distance by the total slope length. Its physical meaning is clear: R→0 indicates that the trailing edge point is closer to the slope toe (e.g., a soil landslide originating at the valley floor), while R→1 indicates it is closer to the slope crest (e.g., a rockfall originating at the mountain peak). For example, if the trailing edge point of a landslide is 70 meters from the valley line and 30 meters from the ridge line, then R=70 / (70+30)=0.7, corresponding to the "upper middle part of the mountain" classification. This normalization transforms absolute distance differences into a unified positional scale, supporting lateral comparisons of multiple events.

[0055] like Figure 6 As shown, in one embodiment, S5, generating a landslide spatial location statistical map based on spatial relative indices corresponding to multiple landslides includes: S54. Divide the range from 0 to 1 into multiple uniform value ranges, and mark the value ranges into which the spatial relative indicators corresponding to multiple landslides fall, so as to form a landslide spatial location statistical map.

[0056] In this embodiment, it should be noted that in S54, the R value is divided into regions according to the landform: 0~1 is evenly divided into 7 intervals (e.g., 0.6≤R<0.8 is the middle and upper part of the mountain), and the distribution of R values ​​of multiple landslides is statistically analyzed.

[0057] For example, in a certain area, 7 out of 10 landslides had R values ​​in the range of 0.6 to 0.8. The statistical chart directly shows that "the middle and upper part is a high-incidence area", revealing the control law of geological structure.

[0058] For example, 0.9 ≤ R < 1.0 is defined as a landslide at the top of a mountain. This threshold is associated with gravity-sensitive areas—top displacement often triggers deep instability. If a landslide has R = 0.95 but actually occurs on the mountainside (such as a misjudgment of the trailing edge of S3), the statistical chart can quickly expose the anomaly (an abnormally high frequency of landslides at the mountain top). This mechanism constitutes a closed-loop verification to ensure the reliability of the data chain.

[0059] Its scientific basis stems from the characteristics of slope distribution—the middle part of the mountain is often a stress concentration zone, and the R-value zoning strictly corresponds to this type of landform pattern.

[0060] In one implementation, the absolute variables between adjacent surface slope data in the surface slope sequence of each acquisition point obtained in S3 include: Obtain the difference between adjacent surface slope data in the surface slope sequence of the i-th sampling point, calculate the absolute value of the difference, and obtain the absolute variable.

[0061] In this embodiment, it should be noted that the absolute variable is defined as the absolute difference in surface slope between adjacent time points. Its calculation relies on time-series data from the same acquisition point: extracting the absolute values ​​of changes between all consecutive monitoring points. For example, if the slope sequence at a certain point is [10°, 8°, 15°], then the adjacent absolute variables are |10-8|=2°, |8-15|=7°. This variable strips away directional features and focuses on fluctuation intensity, providing abrupt change evidence without directional interference for the S3 initial screening.

[0062] like Figure 4 As shown, in one embodiment, S3, obtaining the fluctuation index corresponding to each of the first n collection points on the observation line based on the surface slope sequence corresponding to each of the first n collection points on the observation line includes: S21. Define the first n collection points on the observation line as front-end collection points; S22. Divide each absolute variable in the surface slope sequence of the j-th front-end acquisition point by the average value of the surface slope data in the surface slope sequence of the j-th front-end acquisition point, and obtain each fluctuation ratio in the surface slope sequence of the j-th front-end acquisition point. S23. Sum all the fluctuation proportions in the surface slope sequence of the j-th front-end acquisition point and obtain the fluctuation index corresponding to the j-th front-end acquisition point.

[0063] In this embodiment, it should be noted that in S21, the n consecutive sampling points at the top of the observation line (usually 2-3) are uniformly defined as the front sampling points, representing the area adjacent to the top of the slope. For example, if an observation line extending from the top of the mountain has a total of 50 points, only the first 3 points (the highest point) participate in the ridgeline determination. The core of this design is to limit the physical range—the topographic changes at the top of the slope are usually concentrated within 100 meters of the top. Limiting the value of n to a small value can avoid including secondary platforms far from the top of the mountain (such as mountain shoulder terraces) as candidates, ensuring that the ridgeline is always located within the actual top zone (approximately 50 meters), providing a unified benchmark for ridgeline positioning in cases of multiple landslides.

[0064] In S22, the relative fluctuation ratio is calculated for each front-end acquisition point: first, the absolute variables (absolute values ​​of slope changes) of adjacent time points in the slope sequence of that point are extracted; then, these absolute variables are divided one by one by the average value of the slope sequence of that point. The core of this operation is to eliminate the influence of differences in the base slope—for example, a 5° change in bedrock slope (average slope 35°) has a different physical meaning than a 5° change in soil slope (average slope 10°): the former has a fluctuation ratio of only 5 / 35≈14%, while the latter is as high as 5 / 10=50%, thus quantifying the relative instability. Through normalization processing, the fluctuation of acquisition points with different geological conditions is made comparable.

[0065] In S23, the relative fluctuation ratios of a single front-end acquisition point are summed to generate the fluctuation index corresponding to that point. This index is essentially an overall evaluation of the long-term stability of that point: if the slope fluctuation of a point is weak throughout the monitoring period (e.g., granite bare rock is only affected by temperature differences, with each fluctuation ratio less than 2%), the accumulated value is low; conversely, if it is frequently disturbed (e.g., the vegetation root system area undergoes repeated deformation during the rainy season, with a single fluctuation ratio reaching 20%), the accumulated value increases significantly. For example, if the fluctuation ratios of a certain point in 10 monitoring sessions are 0.5%, 0.3%, 18%, 0.4%, etc., the accumulated fluctuation index is 25.2%, far exceeding the value below 5% for stable points, thus automatically excluding unstable points.

[0066] like Figure 5 As shown, in one embodiment, S4, obtaining the fluctuation index corresponding to each of the last n collection points on the observation line based on the surface slope sequence corresponding to each of the last n collection points on the observation line includes: S41. Define the last n collection points on the observation line as backend collection points; S42. Divide each absolute variable in the surface slope sequence of the j-th back-end acquisition point by the average value of the surface slope data in the surface slope sequence of the j-th back-end acquisition point, and obtain each fluctuation ratio in the surface slope sequence of the j-th back-end acquisition point. S43. Sum all the fluctuation proportions in the surface slope sequence of the j-th back-end acquisition point and obtain the fluctuation index corresponding to the j-th back-end acquisition point.

[0067] In this embodiment, it should be noted that in S41, the last n consecutive sampling points (usually the last 2-3) of the observation line are defined as the rear sampling points, representing the area adjacent to the slope toe. For example, the slope toe may contain various landforms such as the main river channel and alluvial fans, but only the last two points are used for determination. This design avoids secondary depressions (such as piedmont deposits) more than 100 meters away from the slope toe from interfering with the valley line positioning, ensuring that the valley line is always located within the actual slope toe zone (approximately 30 meters). Physically, the main drainage channel at the slope toe is usually adjacent to the end, and a small range of n values ​​ensures spatial comparability of valley lines for different landslide events.

[0068] In S42, the relative fluctuation ratio of the back-end data collection points is calculated. The calculation method is consistent with S22 (absolute variable / mean slope), but it is optimized for the characteristics of the slope toe area. Slope toes are susceptible to water erosion, resulting in a low base slope (e.g., an average slope of 8° in floodplains), while artificial fill areas have a high base slope (e.g., an average slope of 25° in roadbeds). By dividing by the mean, a 2° variation is transformed into a high fluctuation ratio of 25% in floodplains and a low ratio of 8% in roadbed areas, thus exposing the instability of the former. This normalization unifies the stability assessment scale for different slope toe landforms.

[0069] In S43, the fluctuation ratio of all backend data collection points is accumulated to generate a fluctuation index, which serves as the final basis for valley line selection. This index incorporates temporal stability information: for example, the proportion of bedrock valley bottom points is approximately 1% each time (annual index approximately 10%), while the temporary soil dumping area fluctuates repeatedly due to settlement (single proportion reaches 15%, annual index exceeds 80%). The point with the smallest index (most stable) is automatically selected as the valley line.

[0070] A landslide spatial location data processing system based on dynamic segmentation is also provided, the system comprising: The acquisition module is used to acquire the target mountain slope and the observation line located within the target mountain slope and extending from top to bottom, and to acquire the GIS module used to collect the slope surface slope of each collection point on the observation line. It also acquires the data acquisition period when the landslide occurred on the target mountain slope and records it as the processing period, and acquires the surface slope sequence corresponding to each collection point on the observation line within the processing period based on the GIS module. The first data processing module is used to obtain the fluctuation index corresponding to each of the first n collection points on the observation line according to the surface slope sequence corresponding to each of the first n collection points, and to record the location of the collection point with the smallest fluctuation index corresponding to each of the first n collection points as the ridge line. The second data processing module is used to obtain the absolute variables between adjacent surface slope data in the surface slope sequence of each collection point, obtain the collection point corresponding to the surface slope sequence with absolute variables exceeding the natural slope fluctuation threshold and take the collection point as the initial trailing edge point, and select the initial trailing edge point with the highest altitude from multiple initial trailing edge points and take it as the landslide trailing edge point. The third data processing module is used to obtain the fluctuation index corresponding to each of the last n collection points on the observation line based on the surface slope sequence corresponding to each of the last n collection points, and to record the location of the collection point with the smallest fluctuation index corresponding to each of the last n collection points as the valley line. The spatial statistics module is used to obtain spatial relative indicators based on the landslide trailing edge, ridgeline, and valleyline, and to generate a landslide spatial location statistics map based on the spatial relative indicators corresponding to multiple landslides.

[0071] In one implementation, the spatial statistics module is further configured to: obtain a first distance from the landslide trailing edge to the ridgeline, and obtain a second distance from the landslide trailing edge to the valley line; add the first distance and the second distance to obtain the total distance of the target mountain slope area; divide the second distance by the total distance of the target mountain slope area to obtain a spatial relative index.

[0072] In one implementation, the spatial statistics module is also used to: divide the range from 0 to 1 into multiple uniform value ranges, and mark the value ranges into which the spatial relative indicators corresponding to multiple landslides fall, so as to form a landslide spatial location statistics map.

[0073] In one embodiment, the second data processing module is further configured to: obtain the difference between adjacent surface slope data in the surface slope sequence of the i-th sampling point, calculate the absolute value of the difference, and obtain an absolute variable.

[0074] In this embodiment, it should be noted that the specific method of performing the above-mentioned landslide spatial location data processing system based on dynamic segmentation has been described in detail in the embodiments of the landslide spatial location data processing method based on dynamic segmentation, and will not be elaborated here.

[0075] The preferred embodiments of this disclosure have been described in detail above with reference to the accompanying drawings. However, this disclosure is not limited to the specific details of the above embodiments. Within the scope of the technical concept of this disclosure, various simple modifications can be made to the technical solutions of this disclosure, and these simple modifications all fall within the protection scope of this disclosure.

[0076] It should also be noted that the various specific technical features described in the above embodiments can be combined in any suitable manner without contradiction. To avoid unnecessary repetition, this disclosure will not describe the various possible combinations separately.

[0077] Furthermore, various different embodiments of this disclosure can be combined in any way, as long as they do not violate the spirit of this disclosure, they should also be regarded as the content disclosed in this disclosure.

[0078] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them. 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 or all of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention, and they should all be covered within the scope of the claims and specification of the present invention.

Claims

1. A method for processing landslide spatial location data based on dynamic segmentation, characterized in that, include: The target mountain slope and the observation line located within the target mountain slope and extending from top to bottom are obtained. The GIS module used to collect the slope surface slope of each collection point on the observation line is obtained. The data collection period when the landslide occurred on the target mountain slope is obtained and recorded as the processing period. Based on the GIS module, the surface slope sequence corresponding to each collection point on the observation line within the processing period is obtained. Based on the surface slope sequence corresponding to the first n collection points on the observation line, obtain the fluctuation index corresponding to each of the first n collection points on the observation line, and record the location of the collection point with the smallest fluctuation index corresponding to each of the first n collection points as the ridge line. Obtain the absolute variables between adjacent surface slope data in the surface slope sequence of each collection point, obtain the collection point corresponding to the surface slope sequence with absolute variables exceeding the natural slope fluctuation threshold, and take the collection point as the initial trailing edge point. Select the initial trailing edge point with the highest altitude from multiple initial trailing edge points and take it as the landslide trailing edge point. Based on the surface slope sequence corresponding to the last n collection points on the observation line, obtain the fluctuation index corresponding to each of the last n collection points on the observation line, and record the location of the collection point with the smallest fluctuation index corresponding to each of the last n collection points as the valley line. Spatial relative indicators are obtained from the landslide trailing edge, ridgeline, and valleyline, and a landslide spatial location statistical map is generated based on the spatial relative indicators corresponding to multiple landslides.

2. The landslide spatial location data processing method based on dynamic segmentation according to claim 1, characterized in that, The spatial relative indicators for this landslide obtained based on the landslide trailing edge, ridgeline, and valleyline include: Obtain the first distance from the landslide trailing edge to the ridgeline, and the second distance from the landslide trailing edge to the valley line; Add the first distance and the second distance to get the total distance of the target mountain slope area; Divide the second distance by the total distance of the target mountain slope area to obtain the spatial relative index.

3. The landslide spatial location data processing method based on dynamic segmentation according to claim 2, characterized in that, The process of generating a landslide spatial location statistical map based on spatial relative indices corresponding to multiple landslides includes: The range from 0 to 1 is divided into multiple uniform value ranges. The value ranges into which the spatial relative indicators corresponding to multiple landslides fall are marked to form a statistical map of the spatial location of landslides.

4. The landslide spatial location data processing method based on dynamic segmentation according to claim 1, characterized in that, The absolute variables between adjacent surface slope data in the surface slope sequence of each acquisition point include: Obtain the difference between adjacent surface slope data in the surface slope sequence of the i-th sampling point, calculate the absolute value of the difference, and obtain the absolute variable.

5. The landslide spatial location data processing method based on dynamic segmentation according to any one of claims 1 to 4, characterized in that, The process of obtaining the fluctuation index corresponding to each of the first n collection points on the observation line based on the surface slope sequence corresponding to each of the first n collection points includes: The first n data collection points on the observation line are defined as front-end data collection points; Divide each absolute variable in the surface slope sequence of the j-th front-end acquisition point by the average value of the surface slope data in the surface slope sequence of the j-th front-end acquisition point, and obtain each fluctuation ratio in the surface slope sequence of the j-th front-end acquisition point. The fluctuation ratios of all fluctuations in the surface slope sequence of the j-th front-end acquisition point are summed to obtain the fluctuation index corresponding to the j-th front-end acquisition point.

6. The landslide spatial location data processing method based on dynamic segmentation according to any one of claims 1 to 4, characterized in that, The process of obtaining the fluctuation index corresponding to each of the last n collection points on the observation line based on the surface slope sequence corresponding to each of the last n collection points includes: Define the last n data collection points on the observation line as backend data collection points; Divide each absolute variable in the surface slope sequence of the j-th back-end acquisition point by the average value of the surface slope data in the surface slope sequence of the j-th back-end acquisition point, and obtain each fluctuation ratio in the surface slope sequence of the j-th back-end acquisition point. The fluctuation ratios of all fluctuations in the surface slope sequence of the j-th back-end acquisition point are summed to obtain the fluctuation index corresponding to the j-th back-end acquisition point.

7. A landslide spatial location data processing system based on dynamic segmentation, characterized in that, The system is used to implement the landslide spatial location data processing method based on dynamic segmentation as described in any one of claims 1 to 6, the system comprising: The acquisition module is used to acquire the target mountain slope and the observation line located within the target mountain slope and extending from top to bottom, and to acquire the GIS module used to collect the slope surface slope of each collection point on the observation line. It also acquires the data acquisition period when the landslide occurred on the target mountain slope and records it as the processing period, and acquires the surface slope sequence corresponding to each collection point on the observation line within the processing period based on the GIS module. The first data processing module is used to obtain the fluctuation index corresponding to each of the first n collection points on the observation line according to the surface slope sequence corresponding to each of the first n collection points, and to record the location of the collection point with the smallest fluctuation index corresponding to each of the first n collection points as the ridge line. The second data processing module is used to obtain the absolute variables between adjacent surface slope data in the surface slope sequence of each collection point, obtain the collection point corresponding to the surface slope sequence with absolute variables exceeding the natural slope fluctuation threshold and take the collection point as the initial trailing edge point, and select the initial trailing edge point with the highest altitude from multiple initial trailing edge points and take it as the landslide trailing edge point. The third data processing module is used to obtain the fluctuation index corresponding to each of the last n collection points on the observation line based on the surface slope sequence corresponding to each of the last n collection points, and to record the location of the collection point with the smallest fluctuation index corresponding to each of the last n collection points as the valley line. The spatial statistics module is used to obtain spatial relative indicators based on the landslide trailing edge, ridgeline, and valleyline, and to generate a landslide spatial location statistics map based on the spatial relative indicators corresponding to multiple landslides.

8. The landslide spatial location data processing system based on dynamic segmentation according to claim 7, characterized in that, The spatial statistics module is also used for: Obtain the first distance from the landslide trailing edge to the ridgeline, and the second distance from the landslide trailing edge to the valley line; Add the first distance and the second distance to get the total distance of the target mountain slope area; Divide the second distance by the total distance of the target mountain slope area to obtain the spatial relative index.

9. The landslide spatial location data processing system based on dynamic segmentation according to claim 7, characterized in that, The spatial statistics module is also used for: The range from 0 to 1 is divided into multiple uniform value ranges. The value ranges into which the spatial relative indicators corresponding to multiple landslides fall are marked to form a statistical map of the spatial location of landslides.

10. The landslide spatial location data processing system based on dynamic segmentation according to claim 7, characterized in that, The second data processing module is also used for: Obtain the difference between adjacent surface slope data in the surface slope sequence of the i-th sampling point, calculate the absolute value of the difference, and obtain the absolute variable.

Images