Traffic network ecological gradient influence mode recognition method based on template matching

Abstract

The invention provides a template matching-based traffic network ecological gradient influence pattern recognition method, which comprises the following steps of: dividing road network data of a road section to be evaluated into a plurality of heterogeneous road sections according to a fixed interval length, constructing a multi-ring buffer area for the heterogeneous road sections, and generating a spatial data set containing a plurality of gradient units; calculating and synthesizing an ecological fragmentation index of a gradient unit through an objective weighting method in combination with the spatial data set and land utilization data; then, the ecological fragmentation indexes are grouped according to heterogeneous road sections to which the ecological fragmentation indexes belong, and gradient characteristic curves of the heterogeneous road sections are generated according to distance descending sorting; constructing a plurality of traffic ecological influence gradient template curves representing different change modes according to the common attributes of the curves; and finally, calculating the shape similarity between the characteristic curve and the template curve, determining a corresponding template type, and outputting a traffic ecological influence gradient mode classification result of the to-be-evaluated road section so as to solve the defect that the traffic ecological gradient influence mode cannot be systematically and standardly identified and classified.

Description

Technical Field

[0001] This invention relates to the field of geographic information technology, and in particular to a method for identifying the influence patterns of ecological gradients in transportation networks based on template matching. Background Technology

[0002] Transportation networks are vital carriers of human activities, playing a significant role in promoting economic development and regional interaction. However, the spatial structure and density of transportation networks also have a profound impact on the ecological environment, potentially leading to ecological fragmentation. The impacts of transportation networks on the ecological environment exhibit multi-scale and multi-gradient characteristics, and identifying and extracting these complex impact patterns has become a major technical bottleneck in the field of transportation ecology.

[0003] Currently, most studies on the ecological impact of roads use buffer zone analysis, which involves establishing buffer zones at different distances on both sides of the road and calculating ecological indicators within each buffer zone to assess the gradient changes in impact.

[0004] However, existing technologies often focus on the analysis of single or a few ecological indicators, lacking systematic integration and comprehensive quantification of multi-dimensional ecological characteristics, making it difficult to fully reflect the overall response status of the ecosystem. Secondly, existing technologies typically only perform independent calculations and simple comparisons of ecological indicators at different distance gradients, failing to effectively extract and quantify the overall pattern characteristics of ecological impacts changing with distance, and thus unable to form a standardized impact type classification system. Furthermore, existing methods rely heavily on researchers' experience and judgment, lacking automated pattern recognition and classification mechanisms, resulting in highly subjective and low reproducibility of analysis results, making it difficult to effectively compare and integrate research findings from different regions. Summary of the Invention

[0005] This invention provides a template matching-based method for identifying traffic network ecological gradient impact patterns, which addresses the shortcomings of existing technologies in systematically and standardizedly identifying and classifying traffic ecological gradient impact patterns.

[0006] On the one hand, this invention provides a method for identifying the influence patterns of ecological gradients in transportation networks based on template matching, including: Obtain road network data and land use data for the road section to be evaluated; The road network data is divided into several heterogeneous road segments according to the segment interval length; and a multi-ring buffer is constructed for the heterogeneous road segments to generate a spatial dataset composed of multiple gradient units.

[0007] Based on the spatial dataset and the land use data, the ecological fragmentation index of the gradient unit is calculated and synthesized using an objective weighting method. The ecological fragmentation index is grouped according to the heterogeneous road segments to which it belongs, and sorted in descending order of distance to generate gradient characteristic curves for the heterogeneous road segments. Based on the common properties of the gradient characteristic curves, various gradient template curves representing different change patterns of traffic ecological impact are constructed. Calculate the shape similarity between the gradient feature curve and the traffic ecological impact gradient template curve, and based on the similarity, determine the template type corresponding to the road segment to be evaluated, and output the traffic ecological impact gradient pattern classification result of the road segment to be evaluated.

[0008] Optionally, the road network data is divided into several heterogeneous road segments according to the segment interval length; and a multi-ring buffer is constructed for the heterogeneous road segments to generate a spatial dataset composed of multiple gradient units, including: The road network data is segmented according to the segment interval length to obtain several heterogeneous road segments of equal length. For any heterogeneous road segment, a multi-ring buffer is constructed to obtain multiple gradient units corresponding to the heterogeneous road segment; the number of rings in the multi-ring buffer is set to k, and the single-ring distance within a single ring is set to n kilometers, where n is a positive number greater than 0. By summarizing the spatial surface features of the multi-ring buffer corresponding to heterogeneous road segments, a spatial dataset consisting of multiple gradient units is obtained.

[0009] Optionally, the road network data is segmented according to the segmentation interval length to obtain several heterogeneous road segments of equal length, further comprising: The road network data is processed to extract road level attributes, resulting in a road level classification. Establish a mapping table between road grade classification and segment interval length, number of rings, and single ring distance to obtain the mapping relationship of grade parameters; For road network data of different levels, the data is segmented according to the mapping parameters corresponding to the level parameters and a multi-ring buffer is constructed to obtain several heterogeneous road segments.

[0010] Optionally, the step of calculating and synthesizing the ecological fragmentation index of the gradient unit based on the spatial dataset and the land use data using an objective weighting method includes: Based on the land use data, obtain the ecological land data within the gradient unit; Based on the ecological land use data, calculate the ecological landscape pattern index of the gradient unit; The landscape pattern index is standardized to obtain a standardized landscape index. Based on the standardized landscape index set, the weights of the landscape pattern index are calculated using an objective weighting method. For any gradient unit, the standardized landscape index is weighted and summed with the corresponding weights to obtain the ecological fragmentation index of the gradient unit.

[0011] Optionally, the ecological fragmentation index is grouped according to the heterogeneous road segments to which it belongs, and sorted in descending order of distance to generate gradient characteristic curves for the heterogeneous road segments, including: The gradient units are grouped according to the heterogeneity of the road segments, so that the heterogeneous road segments correspond to a group of gradient units; The gradient units within the same heterogeneous road segment group are sorted according to their distance from the road, forming a gradient unit sequence arranged in distance order; Using the distance order of the gradient unit sequence as the x-axis and the ecological fragmentation index value corresponding to each gradient unit in the gradient unit sequence as the y-axis, an ecological impact gradient characteristic curve is generated for the heterogeneous road segment.

[0012] Optionally, based on the common properties of the gradient characteristic curves, various gradient template curves representing different change patterns of traffic ecological impact are constructed, including: Data analysis is performed on the gradient characteristic curve to obtain common attributes, including the standard length L of the gradient characteristic curve and the numerical range of the ecological fragmentation index. Based on the aforementioned common attributes, a standard horizontal axis coordinate sequence of length L is generated, and L equally spaced vertical axis candidate values ​​are generated within the specified numerical range. The candidate values ​​of the vertical axis are arranged according to a preset variation pattern to form different sequences of vertical coordinates; By combining the ordinate sequence with the standard abscissa sequence, a gradient template curve representing the traffic ecological impact of different change patterns is constructed.

[0013] Optionally, calculating the shape similarity between the gradient feature curve and the traffic ecological impact gradient template curve includes: The gradient feature curve and the template curve are used as sequence data to be compared; The dynamic time warping algorithm is used to calculate the dynamic time warping distance between the gradient feature curve and the traffic ecological impact gradient template curve. Based on the dynamic time warping, the similarity between the gradient feature curve and the template curve is determined, and the shape similarity is obtained.

[0014] Optionally, a dynamic time warping algorithm is used to calculate the dynamic time warped distance between the gradient feature curve and the traffic ecological impact gradient template curve, including: Constructing gradient feature curves Template curve of traffic ecological impact gradient An n×m dimensional distance matrix D; where the matrix element d(i,j) in the distance matrix D is the Euclidean distance between two points; Construct the cumulative distance matrix C; wherein the formula for element c(i,j) in the cumulative matrix C is: ; Find the element using backtracking. To element The minimum cost path; the cumulative distance of the minimum cost path is the dynamic time warped distance between the gradient feature curve and the traffic ecological impact gradient template curve.

[0015] Optionally, it also includes: Based on the shape similarity of the gradient feature curves of the road segments, abnormal road segments with a shape similarity matching degree lower than a preset threshold are identified; The gradient feature curves of the abnormal road segments are extracted, and the curve shapes are identified through cluster analysis. Curve shapes that appear more frequently than a set threshold are added to the traffic ecological impact gradient template curve.

[0016] Optionally, it also includes: Calculate the shape similarity metric between the gradient characteristic curve of each road segment and the traffic ecological impact gradient template curve; When the similarity metric value is lower than the similarity threshold, the road segment is marked as a road segment to be checked.

[0017] On the other hand, the present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the template matching-based traffic network ecological gradient influence pattern recognition method as described above.

[0018] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the template matching-based traffic network ecological gradient influence pattern recognition method as described above.

[0019] On the other hand, the present invention also provides a computer program product, including a computer program that, when executed by a processor, implements the template matching-based traffic network ecological gradient influence pattern recognition method as described above.

[0020] This invention provides a method for identifying traffic ecological gradient impact patterns based on template matching. The method acquires road network data and land use data for the road segment to be evaluated. The road network data is then divided into several heterogeneous segments at fixed intervals, creating multi-ring buffer zones to generate a spatial dataset containing multiple gradient units. Next, the spatial dataset and land use data are combined, and the ecological fragmentation index of the gradient units is calculated and synthesized using an objective weighting method. Subsequently, the ecological fragmentation index is grouped according to its heterogeneous road segment and sorted in descending order of distance to generate gradient feature curves for the heterogeneous road segments. Then, based on the common attributes of the curves, various traffic ecological impact gradient template curves representing different change patterns are constructed. Finally, the shape similarity between the feature curves and the template curves is calculated to determine the corresponding template type, and the classification results of the traffic ecological impact gradient patterns of the road segment to be evaluated are output. This method addresses the deficiency of systematically and standardizedly identifying and classifying traffic ecological gradient impact patterns. Attached Figure Description

[0021] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0022] Figure 1 This is a schematic diagram of the process for identifying the gradient impact of transportation networks on the ecological environment, provided in an embodiment of the present invention. Figure 2 This is a schematic diagram of land use data and road network data provided in an embodiment of the present invention; Figure 3 This is a schematic diagram of the ecological fragmentation index corresponding to the gradient unit provided in this embodiment of the invention; Figure 4 This is a schematic diagram of the basic template for the traffic ecological impact gradient provided in this embodiment of the invention; Figure 5 This is a schematic diagram of the basic model of the impact of traffic ecological gradient on different road sections provided in the embodiments of the present invention; Figure 6 This is a schematic diagram illustrating the traffic ecological gradient impact types corresponding to different road sections provided in this embodiment of the invention; Figure 7 This is a schematic diagram of the structure of the electronic device provided in an embodiment of the present invention. Detailed Implementation

[0023] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0024] Figure 1 This is a schematic diagram of the process for identifying the influence pattern of traffic network ecological gradient based on template matching, provided in an embodiment of the present invention.

[0025] like Figure 1 As shown in the embodiments of the present invention, the method for identifying the influence pattern of ecological gradient in traffic networks based on template matching mainly includes: 101. Obtain road network data and land use data for the road section to be evaluated.

[0026] Among them, road network data is road network distribution data based on vector line elements. The road network data must include road grade, direction, and actual length. The data accuracy must be adapted to the research scale, such as using a 1:10,000 scale for the city area and a 1:50,000 scale for the county area.

[0027] Land use data should be selected from data collected during the same period or in recent years as road network data. This data can be obtained through remote sensing image inversion products and land spatial planning databases. It is necessary to clearly distinguish between forest land, grassland, ecological land, construction land, cultivated land, and other types. At the same time, data preprocessing should be completed to ensure that the projection coordinate system of road network data and land use data is consistent, and the data should be cropped to the target study area where the road section to be evaluated is located.

[0028] 102. Divide the road network data into several heterogeneous road segments according to the segment interval length; and construct a multi-ring buffer for the heterogeneous road segments to generate a spatial dataset composed of multiple gradient units.

[0029] The gradient unit is a ring-shaped region with varying widths, centered on a heterogeneous road segment and extending outwards. Each ring-shaped region constitutes a gradient unit, and these gradient units collectively form the spatial dataset used to analyze the impact of ecological and environmental gradients. When dividing heterogeneous road segments, the segment interval length must comprehensively consider factors such as road characteristics, research objectives, and data accuracy to ensure that each heterogeneous road segment accurately reflects the local characteristics of the road network. Simultaneously, the construction of multi-ring buffer zones requires a reasonable determination of the number and width of buffer zones based on the sensitivity of ecological and environmental gradient changes and research needs, in order to comprehensively capture the multi-scale impacts of roads on the ecological environment.

[0030] Specifically, the road network data is divided into several heterogeneous road segments according to the segment interval length; and a multi-ring buffer is constructed for the heterogeneous road segments to generate a spatial dataset composed of multiple gradient units, including: The road network data is segmented according to the segment interval length to obtain several heterogeneous road segments of equal length.

[0031] Specifically, for the road network data to be evaluated, the segmentation interval is first determined based on road traffic volume and surrounding ecosystem type. For example, a segmentation interval of m kilometers is set, dividing the road line elements into multiple segments of equal length as heterogeneous road segments.

[0032] For any heterogeneous road segment, construct a multi-ring buffer to obtain multiple gradient units corresponding to the heterogeneous road segment.

[0033] The number of rings in the multi-ring buffer zone is set to k, and the distance between single rings within a single ring zone is set to n kilometers, where n is a positive number greater than 0.

[0034] By summarizing the spatial surface features of the multi-ring buffer corresponding to heterogeneous road segments, a spatial dataset consisting of multiple gradient units is obtained.

[0035] For each heterogeneous road segment, a multi-ring buffer zone is established based on the radius of influence of the segment on the surrounding ecosystem. For example, the number of rings (k) in the multi-ring buffer zone is set to 5, and the distance of a single ring is n = 1.5 km. Using buffer zone analysis tools, five ring surface features are generated extending from the road to both sides: 0-1.5 km, 1.5-3 km, 3-4.5 km, 4.5-6 km, and 6-7.5 km. Each ring is an independent gradient unit, accurately capturing the gradient differences in the impact on the ecosystem at different distances. Finally, the buffer surface features of all heterogeneous road segments are aggregated, assigned road segment numbers, ring sequence numbers, and distances from the road, forming a spatial dataset.

[0036] In addition, the road network data is segmented according to the segment interval length to obtain several heterogeneous road segments of equal length, which also includes: For road network data, perform the operation of extracting road level attributes to obtain road level classification; Establish a mapping table between road grade classification and segment interval length, number of rings, and single ring distance to obtain the mapping relationship of grade parameters; For road network data of different levels, the data is segmented according to the mapping parameters corresponding to the level parameters and a multi-ring buffer is constructed to obtain several heterogeneous road segments.

[0037] Specifically, in application scenarios that require differentiating the intensity of road impact, such as ecological impact assessment of urban agglomeration transportation networks and protection of cross-regional ecological corridors, it is necessary to assess the differentiated impacts of roads of different grades on farmland and forest land along the route.

[0038] First, road classification attributes are extracted from road network data to identify expressways, national highways, provincial highways, and county roads. Then, based on measured data of traffic flow, speed, and ecological impact for different road classifications, a mapping relationship of classification parameters is established. For example, expressways correspond to a segment interval of 10km, 7 ring roads, and a single ring distance of 2km; national highways correspond to a segment interval of 8km, 5 ring roads, and a single ring distance of 1.5km; and county roads correspond to a segment interval of 5km, 3 ring roads, and a single ring distance of 1km, thus forming a mapping relationship of classification parameters.

[0039] Finally, for road segments of different levels, the corresponding segment intervals in the level parameter mapping relationship are called to divide the lines, and multi-ring buffer zones are constructed according to the corresponding number of rings and single ring distance. In the end, heterogeneous road segments that are adapted to the ecological impact range of each road level are obtained, avoiding the problems of incomplete highway impact assessment and over-classification of county roads caused by uniform parameters.

[0040] 103. Based on spatial datasets and land use data, the ecological fragmentation index of gradient units is calculated and synthesized using an objective weighting method.

[0041] The ecological fragmentation index is used to quantitatively assess the degree of ecosystem fragmentation within each gradient unit. The calculation of the ecological fragmentation index comprehensively considers multiple ecological indicators such as land use type, patch density, and edge density. The objective weighting method automatically assigns weights based on the actual contribution of each indicator to ecological fragmentation, avoiding biases that may arise from subjective weighting. By weighted summing of the indicators, the ecological fragmentation index value for each gradient unit is obtained. A higher ecological fragmentation index indicates a higher degree of ecosystem fragmentation within the gradient unit and a more significant impact from human activities such as transportation networks.

[0042] Specifically, based on spatial datasets and land use data, an ecological fragmentation index for gradient units is calculated and synthesized using an objective weighting method, including: Based on land use data, obtain ecological land use data within the gradient unit.

[0043] Ecological land data includes various types such as forest land, grassland, and water bodies, and is a crucial foundation for assessing the state of ecosystems. When acquiring ecological land data, the distribution range and area information of various ecological land types can be extracted from land use data.

[0044] Based on ecological land use data, calculate the ecological landscape pattern index of the gradient unit; Based on land use data within the study area, the ecological landscape pattern index of ecological land within each heterogeneous gradient unit is calculated. The ecological landscape pattern index mainly includes patch density (PD), edge length (TE), landscape segmentation index (DIVISION), separation index (SPLIT), average nearest neighbor distance (ENN_MN), and average patch area (AREA_MN).

[0045] For example, patch density can be calculated by dividing the number of all ecological land patches within a gradient unit by the total area of ​​the unit. Edge length is calculated by measuring the total length of the edges of all ecological land patches. The landscape segmentation index measures the degree of ecological land segmentation; a higher value indicates more severe segmentation. The separation index reflects the degree of separation by calculating the average distance between ecological land patches. The average nearest neighbor distance is the average distance from each ecological land patch to its nearest neighbor, used to describe the spatial distribution relationship between patches. The average patch area is the average area of ​​all ecological land patches.

[0046] The landscape pattern index is standardized to obtain the standardized landscape index. The six landscape indices calculated from all heterogeneous gradient units were standardized using the following formula: ; ; ; in, Indicating the landscape index The result after standardization of the i-th value. This represents the total number of gradient units, which ultimately forms a standardized landscape index dataset of gradient units.

[0047] Based on a standardized set of landscape indices, the weights of the landscape pattern indices are calculated using an objective weighting method. Based on the standardized landscape indices corresponding to all gradient units, the weight of each index is calculated using the CRITIC (CRiteria ImportanceThrough Intercriteria Correlation) weighting method. Finally, based on the standardized values ​​and weights of different indices, the ecological fragmentation index within each heterogeneous gradient unit is calculated. The CRITIC weight calculation method is as follows: ; ; ; ; in, This represents the data in the i-th row and j-th column of the standardized landscape index dataset representing all gradient units. The final calculated weight is... This represents the weight of the index in the j-th column of the standardized landscape index dataset; For any gradient unit, the standardized landscape index is weighted and summed with its corresponding weight to obtain the ecological fragmentation index of the gradient unit.

[0048] Specifically, the weights of each column of the standardized landscape index dataset are determined according to the above calculation method. Then, for each gradient unit, the standardized landscape index data corresponding to the gradient unit is obtained, each standardized landscape index of the gradient unit is multiplied by its corresponding weight, and finally all the product results are added together. The sum is the ecological fragmentation index of the gradient unit.

[0049] 104. Group the ecological fragmentation index according to the heterogeneous road segments to which it belongs, and sort them in descending order of distance to generate gradient characteristic curves for heterogeneous road segments.

[0050] The process involves grouping heterogeneous road segments according to their ecological fragmentation indices, sorting the gradient units within each group in descending order of distance from the road, and using distance as the horizontal axis and the corresponding ecological fragmentation index as the vertical axis to generate a unique ecological impact gradient characteristic curve for each heterogeneous road segment.

[0051] Specifically, the ecological fragmentation index is grouped according to its corresponding heterogeneous road segments and sorted in descending order of distance to generate gradient characteristic curves for heterogeneous road segments, including: The gradient units are grouped according to the heterogeneity of the road segments, so that each heterogeneous road segment corresponds to a group of gradient units; Gradient units within the same heterogeneous road segment group are sorted according to their distance from the road, forming a gradient unit sequence arranged in distance order; Using the distance order of the gradient unit sequence as the x-axis and the ecological fragmentation index value corresponding to each gradient unit in the gradient unit sequence as the y-axis, an ecological impact gradient characteristic curve is generated for heterogeneous road sections.

[0052] In generating gradient feature curves, it is necessary to select all gradient units belonging to a single road segment for the ecological fragmentation index of all heterogeneous road segment units, and calculate the average value of the ecological fragmentation index of the gradient units. For each heterogeneous road segment, the averaged ecological fragmentation index of a single road segment is selected, and the ecological fragmentation index of all gradient units of the same road segment is sorted according to the order of the roads from near to far. For each heterogeneous road segment, based on the length of the sorted ecological fragmentation index sequence on a single road segment, the result is equal to k, and an index sequence starting from 1 to k with an interval of 1 is generated. Using the index sequence as the horizontal axis and the sorted ecological fragmentation index sequence as the vertical axis, the ecological fragmentation index data of each road segment unit is reorganized to generate characteristic curves of ecological impact gradients for different road segments.

[0053] 105. Based on the common properties of gradient characteristic curves, construct multiple gradient template curves representing different change patterns of traffic ecological impact.

[0054] Among them, such as Figure 4 As shown, the traffic ecological impact gradient template curves include four types: gradually increasing, gradually decreasing, increasing then decreasing, and decreasing then increasing. Based on the common attributes of gradient characteristic curves, various traffic ecological impact gradient template curves representing different change patterns are constructed, specifically including: Data analysis of the gradient characteristic curve yields common attributes, including the standard length L of the gradient characteristic curve and the numerical range of the ecological fragmentation index.

[0055] When performing data analysis on gradient feature curves, the standard length L is extracted first, which is the uniform length of all feature curves, to ensure that the horizontal axis dimension of the traffic ecological impact gradient template curve is consistent with that of the feature curve. Secondly, the numerical range of the ecological fragmentation index is determined by statistically analyzing the maximum and minimum values ​​of the ecological fragmentation index of all gradient units, thus clarifying the value boundaries of the vertical axis data and avoiding insufficient template adaptability due to data differences of individual curves.

[0056] Based on common attributes, a standard horizontal axis coordinate sequence of length L is generated, and L equally spaced vertical axis candidate values ​​are generated within the numerical range. Specifically, for the horizontal axis, a standard index sequence of length L is generated, starting from 1 and ending at intervals of 1. For example, when L=5, the horizontal axis sequence is 1, 2, 3, 4, 5. The horizontal axis sequence corresponds one-to-one with the distance sorting index of the feature curve, ensuring the consistency of spatial gradient logic. On the vertical axis, within the determined range of the ecological fragmentation index, L candidate values ​​are generated using an equal-interval interpolation method. For example, when the value range is 0-0.6 and L=5, the candidate values ​​are 0, 0.15, 0.3, 0.45, and 0.6, ensuring that the vertical axis data covers the entire index range and matches the length of the horizontal axis.

[0057] The candidate values ​​of the vertical axis are arranged according to a preset variation pattern to form different sequences of vertical coordinates.

[0058] Among them, based on four preset traffic ecological impact change models, the candidate values ​​of the vertical axis are arranged differently. The gradually increasing model arranges the candidate values ​​in ascending order, presenting a gradient logic that the farther the distance, the higher the fragmentation index. The gradually decreasing model arranges the candidate values ​​in descending order, corresponding to the characteristic that the farther the distance, the lower the fragmentation index. The first increasing and then decreasing model combines the first two arrangement sequences and takes the value combination below the intersection of the two to form a sequence with high in the middle and low at both ends. The first decreasing and then increasing model takes the value combination above the intersection of the superimposed sequence to form a sequence with low in the middle and high at both ends. Each arrangement accurately matches the core change law of a certain type of ecological impact gradient.

[0059] By combining the ordinate sequence with the standard abscissa sequence, a gradient template curve representing the traffic ecological impact of different change patterns is constructed.

[0060] This process involves combining different ordinate sequences with a standard abscissa sequence, connecting the coordinate points to form a complete curve. Each combination of the ordinate and abscissa sequences corresponds to a traffic ecological impact gradient template curve with a pre-defined change pattern, ultimately constructing four basic template types: gradually increasing, gradually decreasing, increasing then decreasing, and decreasing then increasing. The traffic ecological impact gradient template curve serves as a standardized reference, directly comparing its shape with the gradient characteristic curves of various road segments, providing a unified standard for pattern classification.

[0061] 106. Calculate the shape similarity between the gradient feature curve and the traffic ecological impact gradient template curve, and based on the similarity, determine the template type corresponding to the road segment to be evaluated, and output the traffic ecological impact gradient pattern classification result of the road segment to be evaluated.

[0062] The classification results for the traffic ecological impact gradient patterns are as follows: gradually increasing, gradually decreasing, increasing then decreasing, and decreasing then increasing. In the specific calculation, a similarity measurement algorithm is used to accurately quantify the degree of shape difference between the gradient feature curve and each traffic ecological impact gradient template curve. Based on the calculated similarity values, the type corresponding to the template curve with the highest similarity is determined as the traffic ecological impact gradient pattern type of the road segment to be evaluated, and the classification result of the traffic ecological impact gradient pattern for that road segment is then output.

[0063] Specifically, the shape similarity between the gradient characteristic curve and the traffic ecological impact gradient template curve is calculated, including: The gradient feature curve and the template curve are used as the sequence data to be compared.

[0064] The gradient feature curve sequence data originates from specific heterogeneous road segments. The gradient units are indexed by their distance from the road from nearest to farthest, and the corresponding ecological fragmentation index is used as the sequence value, forming a feature sequence. The sequence data of the traffic-ecological impact gradient template curve is constructed based on a preset pattern, such as a gradually decreasing template, ordered by the standard horizontal axis index, with the vertical axis values ​​arranged in descending order of interval, forming a template sequence. By converting curves into sequence data, the curve shape problem is transformed into a sequence similarity problem, adapting to the input requirements of the dynamic time warping algorithm.

[0065] The dynamic time warping algorithm is used to calculate the dynamic time warping distance between the gradient feature curve and the gradient template curve of traffic ecological impact.

[0066] When calculating the dynamic time warping distance based on the transformed feature sequence and the template sequence, it is first necessary to construct a distance matrix D with consistent sequence dimensions; then construct a cumulative distance matrix C; finally, by backtracking from the lower right corner of the cumulative matrix to the upper left corner, the minimum cost path is found. The total cumulative distance of the minimum cost path is the dynamic time warping distance between the two curves, thus completing the quantitative representation of the shape difference.

[0067] The dynamic time warping algorithm is used to calculate the dynamic time warped distance between the gradient feature curve and the gradient template curve of traffic ecological impact. Specifically, this includes: Constructing gradient feature curves Template curve of traffic ecological impact gradient An n×m dimensional distance matrix D; where the matrix element d(i,j) in the distance matrix D is the Euclidean distance between two points; Construct the cumulative distance matrix C; where the elements of the cumulative matrix C are... The formula is: ; Find the element using backtracking. To element The minimum cost path; the cumulative distance of the minimum cost path is the dynamic time warped distance between the gradient feature curve and the traffic ecological impact gradient template curve.

[0068] In calculating the dynamic time-warped distance between the gradient characteristic curve and the traffic ecological impact gradient template curve, the input data is first defined as: gradient characteristic curve. of to The ecological fragmentation index corresponding to n gradient units from near to far for a heterogeneous road segment, and the traffic-ecological impact gradient template curve. of to The m standard values ​​correspond to the preset influence pattern; based on this, an n×m dimensional distance matrix D is first constructed, where each element in the distance matrix D... By calculating the gradient characteristic curve Ecological fragmentation index of the i-th gradient unit Template curve of traffic ecological impact gradient The j-th standard value The Euclidean distance is obtained.

[0069] This is used to quantify the local differences between corresponding data points on the two curves.

[0070] Next, we construct the cumulative distance matrix C, which has the same dimensions as the distance matrix D. The calculation of the element c(i,j) in the cumulative distance matrix C follows the formula c(i,j)=d(i,j)+min{c(i,j)}. 1,j),c(i,j 1),c(i 1,j The rule 1) is that the cumulative distance at the current position is equal to the current local distance d(i,j) plus the distance from above in the previous step (c(i)). 1,j), left side (c(i,j)) 1)) or the top left corner (c(i 1,j 1) The minimum cumulative distance is used to achieve flexible alignment of the two curves, avoiding misjudgment of similarity caused by local small fluctuations in the ecological fragmentation index.

[0071] Finally, a backtracking method is used, starting from c(n,m) in the lower right corner of the cumulative distance matrix C, and tracing back to c(1,1) in the upper left corner. Each time, based on c(i,j), min{c(i 1,j),c(i,j 1),c(i 1,j The rules for the position derivation of 1)} determine the previous node, and the final path is the minimum cost path that aligns the two curves. The total cumulative distance of the minimum cost path is the dynamic time warping distance between the gradient feature curve and the template curve. The smaller the dynamic time warping distance, the higher the shape similarity between the two, which provides a quantitative basis for the subsequent classification of ecological impact patterns.

[0072] The similarity between the gradient feature curve and the template curve is determined by dynamic time warping, thus obtaining the shape similarity.

[0073] Based on the physical meaning of dynamic time warping distance, the smaller the dynamic time warping distance, the closer the sequence shapes are. A correspondence between distance and similarity is established, and shape similarity results are output. For example, for a feature sequence of a heterogeneous road segment, the dynamic time warping distance is calculated between it and four types of template sequences: gradually increasing, gradually decreasing, increasing then decreasing, and decreasing then increasing. Four distance values ​​are obtained: d1=0.8, d2=0.2, d3=1.5, and d4=1.1. By comparison, d2 is the smallest, therefore, the feature curve is determined to have the highest shape similarity to the gradually decreasing template. This indicates that the current heterogeneous road segment is of the gradually decreasing type.

[0074] In some embodiments, the traffic network ecological gradient impact pattern recognition method based on template matching provided by the present invention further includes: Based on the shape similarity of the gradient feature curves of road segments, abnormal road segments with a shape similarity matching degree lower than a preset threshold are identified; Extract gradient feature curves of abnormal road sections, identify curve shapes through cluster analysis, and supplement the traffic ecological impact gradient template curves with curve shapes that appear more frequently than a set threshold.

[0075] The preset threshold can be flexibly set according to actual application scenarios and needs. For example, it can be set to classify road segments with a similarity matching degree of less than 70% as abnormal road segments. After identifying abnormal road segments, gradient feature curves of the abnormal road segments are extracted using data extraction tools. For the extracted gradient feature curves, clustering analysis algorithms, such as K-means clustering, are used to classify them based on the shape characteristics of the curves. During the clustering analysis process, a frequency threshold is set. When the frequency of a certain type of curve shape appearing in all abnormal road segment curves exceeds the set threshold, this type of curve shape is added to the traffic ecological impact gradient template curve, thereby continuously improving the template curve library and enhancing the accuracy and comprehensiveness of subsequent pattern recognition.

[0076] In some embodiments, the traffic network ecological gradient impact pattern recognition method based on template matching provided by the present invention further includes: Calculate the shape similarity measure between the gradient characteristic curve of each road segment and the gradient template curve of traffic ecological impact; When the similarity metric value is lower than the similarity threshold, the road segment is marked as a road segment to be checked.

[0077] Specifically, algorithms such as cosine similarity or dynamic time warping can be used to calculate the shape similarity measure between the gradient feature curve of each road segment and the gradient template curve of traffic ecological impact. Cosine similarity or dynamic time warping can effectively measure the similarity between two curves in shape, providing a basis for subsequent judgment.

[0078] Then, a similarity threshold is set, which can be flexibly adjusted according to the actual application scenario and needs. When the calculated similarity metric value is lower than this threshold, it indicates that the gradient characteristic curve of the road segment differs significantly in shape from the gradient template curve of the traffic ecological impact. In this case, the road segment is marked as a road segment to be verified. For road segments to be verified, further data analysis can be performed to determine whether there are indeed any anomalies in traffic ecological impact.

[0079] In some embodiments, the present invention selects land use data and major road network data of a certain region in a certain year as experimental objects. The spatial resolution of the land use data of the experimental objects is 30m×30m, and the road data comes from OSM (OpenStreetMap), and the data distribution is as follows. Figure 2 As shown in the figure. The distance interval m between road segments was set to 10 kilometers, the number of buffer zones k was set to 5, and the distance between individual buffer zones was 1 kilometer. The spatial distribution of the ecological fragmentation index obtained from the experiment is shown in the figure. Figure 3 As shown, the four basic templates generated are as follows: Figure 4 As shown, the characteristic curves of the ecological impact gradient for all heterogeneous road segments are as follows: Figure 5 As shown, the traffic ecological gradient impact types corresponding to different road segments are ultimately generated as follows: Figure 6 As shown in the figure. This invention, through a comprehensive analysis of the transportation network and the ecological environment, demonstrates the differentiated impacts of different road sections on the ecosystem using actual data.

[0080] Figure 7 This is a schematic diagram of the structure of the electronic device provided in an embodiment of the present invention.

[0081] like Figure 7 As shown, the electronic device may include a processor 710, a communications interface 720, a memory 730, and a communication bus 740. The processor 710, communications interface 720, and memory 730 communicate with each other via the communication bus 740. The processor 710 can call logical instructions from the memory 730 to execute a template-matching-based traffic network ecological gradient impact pattern recognition method.

[0082] Furthermore, the logical instructions in the aforementioned memory 730 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, essentially, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0083] On the other hand, the present invention also provides a computer program product, which includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer is able to execute the template matching-based traffic network ecological gradient impact pattern recognition method provided by the above methods.

[0084] In another aspect, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to perform the template matching-based traffic network ecological gradient impact pattern recognition method provided by the above methods.

[0085] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0086] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0087] 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 of the technical features; and these 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.

Claims

1. A method for recognizing traffic network eco-gradient influence patterns based on template matching, characterized in that, The method comprises the following steps: Obtain road network data and land use data of a road section to be evaluated; Divide the road network data into a plurality of heterogeneous road sections according to the length of the section interval; Construct a multi-ring buffer zone for each heterogeneous road section to generate a spatial data set composed of a plurality of gradient units; Based on the spatial data set and the land use data, calculate the ecological fragmentation index of the gradient unit by an objective weighting method; Group the ecological fragmentation indexes according to the corresponding heterogeneous road sections, and sort them in descending order of distance to generate a gradient characteristic curve of the heterogeneous road section; According to the common attributes of the gradient characteristic curve, construct a plurality of traffic ecological impact gradient template curves representing different change modes; Calculate the shape similarity of the gradient characteristic curve and the traffic ecological impact gradient template curve, and determine the template type corresponding to the road section to be evaluated based on the similarity, and output the traffic ecological impact gradient mode classification result of the road section to be evaluated.

2. The template matching based traffic network eco-gradient influence pattern recognition method according to claim 1, characterized in that, The road network data is divided into a plurality of heterogeneous road sections according to the length of the section interval; and a multi-ring buffer zone is constructed for each heterogeneous road section to generate a spatial data set composed of a plurality of gradient units, which comprises: Segmenting the road network data according to the length of the section interval to obtain a plurality of heterogeneous road sections with equal length; For any one of the heterogeneous road sections, a multi-ring buffer zone is constructed to obtain a plurality of gradient units corresponding to the heterogeneous road section; the number of ring bands of the multi-ring buffer zone is set to k, and the single ring distance inside a single ring band is set to n kilometers, where n is a positive number greater than 0; Collecting the spatial surface elements of the multi-ring buffer zone corresponding to the heterogeneous road sections to obtain a spatial data set composed of a plurality of gradient units.

3. The template matching based traffic network eco-gradient influence pattern recognition method according to claim 2, characterized in that, Segmenting the road network data according to the length of the section interval to obtain a plurality of heterogeneous road sections with equal length, further comprises: Performing a road grade attribute extraction operation on the road network data to obtain a road grade classification; Establishing a mapping table of road grade classification, section interval length, ring band number and single ring distance to obtain a grade parameter mapping relationship; Segmenting and constructing a multi-ring buffer zone for road network data of different grades according to the corresponding mapping parameters of the grade parameter mapping relationship to obtain a plurality of heterogeneous road sections.

4. The template matching based traffic network eco-gradient influence pattern recognition method according to claim 1, characterized in that, The ecological fragmentation index of the gradient unit is calculated and synthesized based on the spatial data set and the land use data by an objective weighting method, which comprises: According to the land use data, obtain the ecological land data within the gradient unit; According to the ecological land data, calculate the ecological landscape pattern index of the gradient unit; Standardizing the landscape pattern index to obtain a standardized landscape index; Based on the standardized landscape index set, the weight of the landscape pattern index is calculated by an objective weighting method; For any one of the gradient units, the standardized landscape index and the corresponding weight are weighted and summed to obtain the ecological fragmentation index of the gradient unit.

5. The template matching based traffic network eco-gradient influence pattern recognition method according to claim 1, characterized in that, The ecological fragmentation indexes are grouped according to the corresponding heterogeneous road sections, and sorted in descending order of distance to generate a gradient characteristic curve of the heterogeneous road section, which comprises: grouping the gradient units according to heterogeneous road segments, so that the heterogeneous road segments correspond to a group of gradient units; sorting the gradient units in the same heterogeneous road segment group according to the distance from the road to form a gradient unit sequence arranged in distance order; generating an ecological impact gradient characteristic curve for the heterogeneous road segment with the distance order of the gradient unit sequence as the horizontal coordinate and the ecological fragmentation index value corresponding to each gradient unit in the gradient unit sequence as the vertical coordinate.

6. The template matching based traffic network eco-gradient influence pattern recognition method according to claim 1, characterized in that, According to the common attributes of the gradient characteristic curve, a plurality of traffic ecological impact gradient template curves representing different change modes are constructed, including: performing data analysis on the gradient characteristic curve to obtain common attributes, including the standard length L of the gradient characteristic curve and the numerical range of the ecological fragmentation index; based on the common attributes, generate a standard horizontal axis coordinate sequence with a length of L and L equally spaced vertical axis candidate values within the numerical range; arrange the vertical axis candidate values according to the preset change mode to form different vertical coordinate sequences; combine the vertical coordinate sequence with the standard horizontal axis coordinate sequence to construct traffic ecological impact gradient template curves representing different change modes.

7. The template matching based traffic network eco-gradient influence pattern recognition method according to claim 1, characterized in that, Calculate the shape similarity of the gradient characteristic curve and the traffic ecological impact gradient template curve, including: the gradient characteristic curve and the template curve are taken as sequence data to be compared; adopting dynamic time warping algorithm to calculate the dynamic time warping distance between the gradient characteristic curve and the traffic ecological impact gradient template curve; based on the dynamic time warping, determine the similarity degree of the gradient characteristic curve and the template curve to obtain the shape similarity.

8. The template matching based traffic network eco-gradient influence pattern recognition method according to claim 7, characterized in that, Adopting dynamic time warping algorithm to calculate the dynamic time warping distance between the gradient characteristic curve and the traffic ecological impact gradient template curve, including: Constructing a gradient characteristic curve with a traffic ecological impact gradient template curve a n x m dimensional distance matrix D; wherein a matrix element d(i,j) in the distance matrix D is the Euclidean distance between two points; constructing a cumulative distance matrix C; wherein the formula of the element c(i,j) in the cumulative matrix C is: ； finding a minimum cost path from an element to an element by backtracking; the cumulative distance of the minimum cost path is a dynamic time warping distance of the gradient characteristic curve and the traffic ecological impact gradient template curve.

9. The template matching based traffic network eco-gradient influence pattern recognition method according to claim 1, characterized in that, Further comprising: based on the shape similarity of the road segment gradient characteristic curve, identifying abnormal road segments with a shape similarity matching degree below a preset threshold; extracting the gradient characteristic curve of the abnormal road segment, identifying the curve shape through clustering analysis, and supplementing the curve shape with a frequency exceeding a set threshold to the traffic ecological impact gradient template curve.

10. The template matching based traffic network eco-gradient influence pattern recognition method according to claim 9, characterized in that, Further comprising: calculating the shape similarity measure value between the gradient characteristic curve of each road segment and the traffic ecological impact gradient template curve; when the similarity measure value is lower than the similarity threshold, mark the road segment as a road segment to be checked.

Images