Coastal ecological region identification method based on remote sensing image

Abstract

The application discloses a coastal ecological region identification method based on remote sensing images, relates to the technical field of coastal ecological region identification, and comprises the following steps: collecting remote sensing images of a coastal region which are not blocked by clouds in a whole year, performing image preprocessing on the remote sensing images to obtain preprocessed images; identifying the supratidal zone, the intertidal zone and the subtidal zone of the coastal region through the preprocessed images; analyzing the change characteristic lines of different pixel points based on the time sequence of the remote sensing images, grouping the pixel points based on the change characteristic lines, and obtaining different initial regions; performing similarity analysis on the initial regions, merging similar initial regions, and finally obtaining the coastal ecological region; and the application is used to solve the problem that the existing coastal ecological region identification technology cannot accurately identify the intertidal zone and the coastal ecological region in the intertidal zone, thereby failing to accurately divide the ecological region of the coastline.

Description

Technical Field

[0001] This invention relates to the field of coastal ecological area identification technology, specifically a method for coastal ecological area identification based on remote sensing imagery. Background Technology

[0002] Coastal ecoregions refer to the transitional zones between land and sea. They are complex ecosystems formed by the combined forces of the atmosphere, land, and ocean, with boundaries extending from the coastal land influenced by the ocean to the subtidal shallow waters at the ocean's edge. Coastal ecoregion identification technology refers to a technical system that comprehensively utilizes remote sensing, geographic information systems, ground surveys, and artificial intelligence to accurately define, classify, and dynamically monitor the types, spatial distribution, boundaries, and ecological status of ecosystems within the coastal zone.

[0003] Existing coastal ecological zone identification technologies often struggle to accurately delineate coastal ecological zones, only providing a rough outline. Furthermore, the presence of the intertidal zone due to tidal forces, influenced by both land and sea, further complicates identification. Current technologies cannot accurately distinguish between the supratidal zone, intertidal zone, and other areas, hindering independent analysis of the intertidal zone and leading to significant errors in the identified coastal ecological zones. Moreover, existing technologies fail to precisely identify the intertidal zone and its sub-zones, resulting in an inability to accurately delineate ecological zones along the coastline. Summary of the Invention

[0004] This invention aims to at least partially address one of the technical problems in the prior art. It collects year-round remote sensing images of coastal areas that are not obscured by clouds, preprocesses these images to obtain preprocessed images, identifies the relative positions of the ocean and land in the preprocessed images, analyzes the supratidal, intertidal, and subtidal zones of the coastal area based on these relative positions, and then analyzes the changing feature lines of different pixels in the preprocessed images based on the temporal sequence of the remote sensing images. Based on these changing feature lines, the pixels in the preprocessed images are divided into different initial regions. Finally, a similarity analysis is performed on the initial regions, and similar initial regions are merged to obtain the final coastal ecological region. This addresses the problem that existing coastal ecological region identification technologies cannot accurately identify the intertidal zone and the coastal ecological regions within the intertidal zone, resulting in the inability to accurately delineate ecological regions along the coastline.

[0005] To achieve the above objectives, this application provides a method for identifying coastal ecological areas based on remote sensing imagery, comprising the following steps: Collect remote sensing images of the coastal area throughout the year that are not obscured by clouds, and perform image preprocessing on the remote sensing images to obtain preprocessed images; Preprocessed images are used to identify the supratidal, intertidal, and subtidal zones of coastal areas. Based on the temporal analysis of remote sensing images, the characteristic lines of change of different pixels are analyzed, and the pixels are grouped based on the characteristic lines of change to obtain different initial regions; A similarity analysis is performed on the initial regions, and similar initial regions are merged to finally obtain the coastal ecological regions.

[0006] Furthermore, remote sensing images of the coastal area that are not obscured by clouds throughout the year are collected. The remote sensing images are then preprocessed to obtain preprocessed images, which includes the following sub-steps: Collect all remote sensing images of the coastal area that are not obscured by clouds from January 1 to December 31. Image enhancement is performed on remote sensing images to obtain preprocessed images.

[0007] Furthermore, identifying the supratidal, intertidal, and subtidal zones of a coastal area through preprocessed images includes the following sub-steps: Identify the relative positions of ocean and land in preprocessed images; The analysis of the supratidal, intertidal, and subtidal zones of the coastal area is based on the relative positions of the ocean and land.

[0008] Furthermore, identifying the relative positions of ocean and land in the preprocessed image includes the following sub-steps: The ocean and land in the preprocessed image are identified by image recognition technology. The midpoints of two parallel sides of the four sides of the preprocessed image are connected to obtain dividing lines, which include vertical dividing lines and horizontal dividing lines. The areas above and below the upper and lower dividing lines are named the upper partition and lower partition, respectively; the areas to the left and right of the left and right dividing lines are named the left partition and right partition, respectively. The upper, lower, left, and right partitions are collectively referred to as image partitions. The proportion of ocean to land in the image partitions is calculated and named the ocean-to-land ratio. If the ocean-to-land ratio is greater than or equal to one, the image partition is marked as an ocean area; otherwise, the image partition is marked as a land area. The ocean area and the land area are collectively referred to as ocean-to-land partitions. If the land and sea partitions of the upper and lower partitions are different, the preprocessed image is marked as vertically opposite; if the land and sea partitions of the left and right partitions are different, the preprocessed image is marked as horizontally opposite. If the preprocessed image is both vertically and horizontally relative, then the land-sea ratios of the upper, lower, left, and right partitions are labeled as Q1, Q2, Q3, and Q4, respectively. Calculate (Q1+Q2) / (Q3+Q4). If the calculation result is greater than or equal to one, then the relative positions of the preprocessed image are set to vertically relative; otherwise, the relative positions of the preprocessed image are set to horizontally relative.

[0009] Furthermore, analyzing the supratidal, intertidal, and subtidal zones of a coastal area based on the relative positions of the ocean and land includes the following sub-steps: If the preprocessed image is vertically opposite, then each column of pixels in the preprocessed image is used as a group of tidal zone analysis points; if the preprocessed image is horizontally opposite, then each row of pixels in the preprocessed image is used as a group of tidal zone analysis points. The grayscale values ​​of the pixels in the tidal zone analysis group are numbered according to the direction from ocean to land, using the symbol G. h This indicates that h is a non-zero natural number and h is the index of G; With h as the X-axis, G h Establish a two-dimensional coordinate system for the Y-axis, named the tidal zone analysis diagram, and set G... h Enter the tidal zone analysis diagram according to h, and enter G in the tidal zone analysis diagram. h The corresponding coordinate point is marked as P. h ; P h With P h+1 Connect them, and mark the resulting straight line as PL. h , obtain PL h The slope, denoted as PK h ; PK h Perform cluster analysis to classify PK h Merge into different partitioned clusters, and statistically analyze the PK values ​​in the partitioned clusters. h The number of similarities is named the similarity number. The cluster with the highest similarity number is named the similarity cluster. The PK (primary similarity) in the similarity cluster is then counted. h The maximum value of is represented by the symbol KU; Starting from h=1, determine PK. h Is it less than or equal to KU? If so, then set G. h The corresponding pixel is named an ocean point, and h is incremented by one and analyzed again. If not, the analysis is stopped. After stopping the analysis, the remaining pixels are named land points. Each tidal zone analysis group is analyzed separately to obtain all ocean points and land points. The pixel in the nth row and mth column of the preprocessed image is labeled as PI(n,m). Analyze the PI(n,m) with the same n and m in different preprocessed images. If all PI(n,m) are ocean points, then PI(n,m) is labeled as a subtidal point. If PI(n,m) contains both ocean points and land points, then PI(n,m) is labeled as an intertidal point. If all PI(n,m) are land points, then PI(n,m) is labeled as a supratidal point. The area composed of supratidal points is the supratidal zone, the area composed of intertidal points is the intertidal zone, and the area composed of subtidal points is the subtidal zone.

[0010] Furthermore, based on the temporal analysis of remote sensing images, the characteristic lines of change for different pixels are analyzed, and the pixels are grouped based on these characteristic lines to obtain different initial regions, including the following sub-steps: Temporal analysis of remote sensing images: characteristic lines of change in different pixels in preprocessed images; The pixels in the preprocessed image are divided into different initial regions based on the changing feature lines.

[0011] Furthermore, the temporal analysis of the preprocessed image based on remote sensing imagery, which includes the following sub-steps, involves the following: Assuming the remote sensing image was captured on day d of a year, the preprocessed image corresponding to the remote sensing image will be labeled as PC. d Where d is a non-zero natural number; PC d PI(n,m) in the figure is labeled PC d (n,m), PC d The grayscale value corresponding to (n,m) is labeled as PG. d (n,m); Analyze each value of n and m independently, with d as the horizontal axis, PG d Establish a two-dimensional coordinate system with (n,m) as the vertical axis, and name it the time series variation graph. d (n,m) Enter the time series change graph according to d, perform linear regression on the time series change graph, obtain the slope of the regression function, and name it as the time series change feature. Each PI(n,m) has a time series change feature.

[0012] Furthermore, dividing the pixels in the preprocessed image into different initial regions based on the changing feature lines includes the following sub-steps: Cluster analysis was performed on the temporal variation characteristics to classify them into different temporal feature clusters; The initial region is formed by merging pixels that are in the same temporal feature cluster and are consecutively adjacent in the preprocessed image.

[0013] Furthermore, a similarity analysis is performed on the initial regions, and similar initial regions are merged to finally obtain the coastal ecoregion, which includes the following sub-steps: The average value of the temporal variation characteristics within an initial region is named the regional variation characteristic. When performing similarity analysis on any initial region, it is named the main analysis region, and the initial regions adjacent to the main analysis region are named the secondary analysis regions. PK in similarity clustering h The maximum value is named the same-region threshold; Obtain the regional change characteristics of the main analysis region and the sub-analysis region, and label them as R1 and R2 respectively. Calculate |R1-R2|. If the calculation result is less than or equal to the same region threshold, output the region merged signal; otherwise, output the region unmerged signal. The main and sub-analytical regions are processed based on regional merging and regional dismerging signals. Similar initial regions are merged to obtain the final coastal ecological region.

[0014] Furthermore, based on the regional merging signal and the regional dismerging signal, the main analysis region and the sub-analysis region are processed to merge similar initial regions, ultimately obtaining the coastal ecological region, including the following sub-steps: If the output region merge signal is used, the corresponding sub-analysis region is merged with the main analysis region. The merged region is a coastal ecological region. Each initial region is analyzed, and finally different coastal ecological regions are obtained by merging. The coastal ecological area is outlined and displayed to technical personnel using visualization technology.

[0015] The beneficial effects of this invention are as follows: This invention collects remote sensing images of coastal areas throughout the year that are not obscured by clouds, performs image preprocessing on the remote sensing images to obtain preprocessed images, and then identifies the relative positions of the ocean and land in the preprocessed images. Based on the relative positions of the ocean and land, it analyzes the supratidal, intertidal, and subtidal zones of the coastal area. The advantage is that it can accurately delineate the supratidal, intertidal, and subtidal zones around the coastline, and then independently analyze the ecological areas in the intertidal zone in subsequent analysis, thereby improving the accuracy and rationality of coastal ecological area identification. This invention preprocesses images by analyzing the temporal variation of characteristic lines of different pixels in the preprocessed image based on remote sensing imagery. Then, based on these characteristic lines, the pixels in the preprocessed image are divided into different initial regions. Finally, similar initial regions are merged through similarity analysis to obtain coastal ecological regions. The advantage is that the vegetation within the same ecological region is the same, and their growth cycles are the same. Therefore, based on the similarity of the temporal variation characteristic lines, different coastal ecological regions can be divided, thus improving the accuracy and effectiveness of coastal ecological region identification. Attached Figure Description

[0016] Figure 1 This is a flowchart of the steps of the method of the present invention; Figure 2 This is a preprocessed image for the present invention; Figure 3 This is a schematic diagram of the upper partition, lower partition, left partition, and right partition of the present invention; Figure 4 This is a schematic diagram of the tidal zone analysis diagram of the present invention; Figure 5 This is a schematic diagram of the timing variation of the present invention; Figure 6 This is a schematic diagram of the initial partitioning of the present invention. Detailed Implementation

[0017] 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. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0018] Example 1, please refer to Figure 1 As shown, this application provides a method for identifying coastal ecological areas based on remote sensing imagery, including the following steps: Step S1 involves collecting remote sensing images of the coastal area throughout the year that are not obscured by clouds, and preprocessing the remote sensing images to obtain preprocessed images. Step S1 includes the following sub-steps: Please see Figure 2 As shown, in step S101, all remote sensing images of the coastal area that are not obscured by clouds from January 1 to December 31 are collected. Step S102: Image enhancement is performed on the remote sensing image to obtain a preprocessed image; In practice, all unobstructed remote sensing images of the coastal area from January 1st to December 31st are collected. Each image is taken at noon, and the shooting position and angle must be identical. Since tides are a geographical phenomenon influenced by the moon's gravity, they follow a predictable pattern, with a 15-day cycle. Therefore, the high tide of the following day is delayed by 0.8 hours. Taking images at noon each day allows for analysis of the intertidal zone, eliminating concerns about identical tide levels at the same time. Furthermore, noon typically sees less cloud cover, facilitating the capture of unobstructed images. If an image is obstructed by clouds, it is discarded. Cloud obstruction can be identified manually or by AI; this is not specifically required in this embodiment. Image enhancement utilizes existing image preprocessing techniques, which are not described in detail here. The preprocessed image is shown below. Figure 2 As shown.

[0019] Step S2 involves identifying the supratidal, intertidal, and subtidal zones of the coastal area through preprocessed images. Step S2 includes the following sub-steps: Step S201: Identify the relative positions of the ocean and land in the preprocessed image; Step S201 includes the following sub-steps: Step S2011: Identify the ocean and land in the preprocessed image using image recognition technology, and connect the midpoints of two parallel sides among the four sides of the preprocessed image to obtain dividing lines, including upper and lower dividing lines and left and right dividing lines. Step S2012: Name the area above and below the upper and lower dividing lines as the upper partition and lower partition, respectively; name the area to the left and right of the left and right dividing lines as the left partition and right partition, respectively. Step S2013: The upper partition, lower partition, left partition and right partition are collectively referred to as image partitions. The proportion of ocean and land in the image partitions is calculated and named as the ocean-land ratio. If the ocean-land ratio is greater than or equal to one, the image partition is marked as an ocean area; otherwise, the image partition is marked as a land area. The ocean area and the land area are collectively referred to as ocean-land partitions. Step S2014: If the land and sea partitions of the upper partition and the lower partition are different, then mark the preprocessed image as vertically opposite; if the land and sea partitions of the left partition and the right partition are different, then mark the preprocessed image as horizontally opposite. Step S2015: If the preprocessed image is both vertically and horizontally relative, then the land-sea ratios of the upper, lower, left, and right partitions are labeled as Q1, Q2, Q3, and Q4, respectively. Calculate (Q1+Q2) / (Q3+Q4). If the calculation result is greater than or equal to one, then the relative positions of the preprocessed image are set to vertically relative; otherwise, the relative positions of the preprocessed image are set to horizontally relative. In practice, existing AI image recognition technologies can easily identify oceans and land in preprocessed images, thus enabling image preprocessing. Figure 2 For example, the partitioning results in an upper partition, a lower partition, a left partition, and a right partition, as follows: Figure 3 As shown, taking the right partition as an example, the ratio of pixels belonging to the ocean to pixels belonging to the land in the right partition is 9.26. The result is rounded to two decimal places, meaning the land-sea ratio is 9.26, which is greater than 1. Therefore, the right partition is considered a sea area. Similarly, the land-sea ratios for the other image partitions are calculated, resulting in ratios of 1.86, 0.61, and 0.18 for the upper, lower, and left partitions, respectively. This indicates the upper partition is a sea area, the lower partition is a land area, and the left partition is a land area. Therefore, the upper partition... The land and sea partitions of the left and right sub-sub ...

[0020] Step S202: Analyze the supratidal zone, intertidal zone, and subtidal zone of the coastal area based on the relative positions of the ocean and land; Step S202 includes the following sub-steps: Step S2021: If the preprocessed image is vertically opposite, then each column of pixels in the preprocessed image is used as a group of tidal zone analysis points; if the preprocessed image is horizontally opposite, then each row of pixels in the preprocessed image is used as a group of tidal zone analysis points. Step S2022: Number the grayscale values ​​of the pixels in the tidal zone analysis group according to the direction from ocean to land, using the symbol G. h This indicates that h is a non-zero natural number and h is the index of G; Step S2023, with h as the X-axis, G h Establish a two-dimensional coordinate system for the Y-axis, named the tidal zone analysis diagram, and set G... h Enter the tidal zone analysis diagram according to h, and enter G in the tidal zone analysis diagram. h The corresponding coordinate point is marked as P. h ; Step S2024, P h With P h+1 Connect them, and mark the resulting straight line as PL. h , obtain PL h The slope, denoted as PK h ; In practice, since the preprocessed image is left-right aligned, each row of pixels in the preprocessed image is used as a group for tidal zone analysis, for example... Figure 2 There are 533 rows of pixels, resulting in 533 tidal zone analysis groups. Since the left partition has the largest land-sea ratio, the direction from ocean to land is actually from right to left, and G is obtained by numbering. h Given that 1 ≤ h ≤ 1031, the tidal zone analysis diagram is constructed as follows: Figure 4 As shown, due to the large number of coordinate points, it is not conducive to data observation, therefore... Figure 4 Only a small amount of data is listed as an example. Figure 4 China has already made provisions for PL h Establish connection and obtain PK. h ,For example Figure 4 In the example, PL1 to PL9 correspond to PK1 to PK9 as 0, 0, 0, 1, 0, 0, 1, 0 and 1 respectively.

[0021] Step S2025, for PK h Perform cluster analysis to classify PK h Merge into different partitioned clusters, and statistically analyze the PK values ​​in the partitioned clusters. h The number of similarities is named the similarity number. The cluster with the highest similarity number is named the similarity cluster. The PK (primary similarity) in the similarity cluster is then counted. h The maximum value of is represented by the symbol KU; Step S2026, starting from h=1, determine PK. h Is it less than or equal to KU? If so, then set G. h The corresponding pixel is named an ocean point, and h is incremented by one and analyzed again. If not, the analysis is stopped. Step S2027: After stopping the analysis, name the remaining pixels as land points, and analyze each tidal zone analysis group separately to obtain all ocean points and land points. Step S2028: Mark the pixel in the nth row and mth column of the preprocessed image as PI(n,m). Analyze the PI(n,m) with the same n and m in different preprocessed images. If all PI(n,m) are ocean points, mark PI(n,m) as subtidal points. If PI(n,m) contains both ocean points and land points, mark PI(n,m) as intertidal points. If all PI(n,m) are land points, mark PI(n,m) as supratidal points. The area composed of supratidal points is the supratidal zone, the area composed of intertidal points is the intertidal zone, and the area composed of subtidal points is the subtidal zone. In practice, for all PK h Cluster analysis yielded 8 sub-clusters. Among them, due to the similar colors of the ocean surface, the grayscale values ​​varied little, thus forming continuous pk clusters. hTypically smaller, and PK also exists on some land areas. h Smaller values ​​are usually the most common, so similar clusters can be found by counting similarity numbers. Additionally, the key-priority (PK) values ​​in the clusters can also be calculated. h The average value is used to find the minimum value among the average values ​​to obtain similar clusters. PK h A smaller value indicates that the grayscale values ​​of two pixels are similar, meaning the two pixels belong to the same type of object. Similarity clustering is obtained through statistical analysis. In this embodiment, the PK value in the similarity clustering... h The maximum value is 21, meaning KU is 21. Taking PK1 as an example, PK1 is 0, which is less than KU. Therefore, the pixel corresponding to G1 is named an ocean point. Similarly, analyzing all ocean points, when the ocean is in contact with land, the grayscale value here usually changes significantly, i.e., PK... h If the value is greater than 21, it indicates that the specific boundary between the ocean and land has been found. This allows for a detailed distinction between the ocean and land in the preprocessed image. Since different preprocessed images represent different stages of high tide, analyzing preprocessed images throughout the year reveals the division of the same pixel at different times. If a pixel is a land point in one period but an ocean point in another, it indicates that this location falls within the high and low tide range, i.e., the intertidal zone. This allows for the division of the supratidal zone, intertidal zone, and subtidal zone in the preprocessed image. The supratidal zone is the land portion that is not submerged by seawater, the intertidal zone is the land portion that is submerged by seawater, and the subtidal zone is the portion that is always ocean.

[0022] Step S3 involves analyzing the temporal characteristics of different pixels in the remote sensing image and grouping the pixels based on these characteristic lines to obtain different initial regions. Step S3 includes the following sub-steps: Step S301: Based on the temporal analysis of remote sensing images, the change feature lines of different pixels in the preprocessed image are analyzed. Step S301 includes the following sub-steps: Step S3011: Assuming the remote sensing image was captured on day d within a year, the preprocessed image corresponding to the remote sensing image is labeled as PC. d Where d is a non-zero natural number; Step S3012, PC d PI(n,m) in the figure is labeled PC d (n,m), PC d The grayscale value corresponding to (n,m) is labeled as PG. d (n,m); Please see Figure 5 As shown, in step S3013, each value of n and m is analyzed independently, with d as the horizontal axis and PG... dEstablish a two-dimensional coordinate system with (n,m) as the vertical axis, and name it the time series variation graph. d (n,m) Enter the time series change graph according to d, perform linear regression on the time series change graph, obtain the slope of the regression function, and name it as the time series change feature. Each PI(n,m) has a time series change feature. In practice, for example, if a remote sensing image is captured on the first day of a year, it is labeled as PC1. If the remote sensing image captured on the second day is obscured by clouds, it is discarded, so there is no PC2. And so on. Taking PI(1,1) as an example, a total of 248 remote sensing images are obtained in a year, meaning there are 248 different values ​​of PC for each day. d (1,1), the time series variation diagram is constructed as follows: Figure 5 As shown, due to the large amount of data, this embodiment only lists 248 PCs. d Partial data of (1,1) are provided for reference. The temporal variation feature obtained by linear regression is -0.0101. The temporal variation feature reveals the change of the pixel from the beginning to the end of the year. If the pixels belong to the same ecological environment, their changes should be similar. The temporal variation feature of each pixel is analyzed.

[0023] Step S302: Divide the pixels in the preprocessed image into different initial regions based on the changing feature lines; Step S302 includes the following sub-steps: Step S3021: Perform cluster analysis on the time-series variation characteristics and divide the time-series variation characteristics into different time-series feature clusters; Please see Figure 6 As shown, in step S3022, pixels that are in the same temporal feature cluster and are consecutively adjacent in the preprocessed image are merged into one region, which is the initial region. In practice, cluster analysis yields different temporal feature clusters. Temporal variation features within the same cluster represent values ​​that are very similar, typically indicating the same ecological region. Merging consecutive adjacent pixels within the same cluster yields the initial region. Analyzing the supratidal, intertidal, and subtidal zones separately yields different initial regions, such as... Figure 6 As shown, Figure 6 A total of 6 initial regions were delineated.

[0024] Step S4 involves performing a similarity analysis on the initial regions, merging similar initial regions to ultimately obtain the coastal ecological region. Step S4 includes the following sub-steps: Step S401: Calculate the average value of the time series change characteristics in an initial region and name it the region change characteristics. When performing similarity analysis on any initial region, name it the main analysis region and name the initial region adjacent to the main analysis region the secondary analysis region. Step S402, in the similarity cluster, PK h The maximum value is named the same-region threshold; Step S403: Obtain the regional change characteristics of the main analysis region and the sub-analysis region, and label them as R1 and R2 respectively. Calculate |R1-R2|. If the calculation result is less than or equal to the same region threshold, output the region merging signal; otherwise, output the region non-merging signal. In practice, each pixel has a temporal variation feature, and there are multiple pixels within an initial region. Their average value is calculated to obtain the region variation feature. The threshold for the same region is 21, because the PK in similarity clustering... h This indicates that two pixels belong to the same ecosystem, and the minimum value must be 0, while the maximum value is equal to the difference between the two. This is used to determine whether two initial regions belong to the same ecosystem. For example, if the regional variation feature of an initial region is 14, and the regional variation feature of an adjacent initial region is 18, the difference between the two is 4, which is less than 21. Therefore, the region merging signal is output, indicating that the two initial regions belong to the same ecosystem.

[0025] Step S404: Based on the regional merging signal and the regional disintegration signal, the main analysis region and the sub-analysis region are processed to merge similar initial regions and finally obtain the coastal ecological region. Step S404 includes the following sub-steps: Step S4041: If the output region merge signal is a signal, then the corresponding sub-analysis region is merged with the main analysis region. The merged region is a coastal ecological region. Each initial region is analyzed, and finally different coastal ecological regions are obtained by merging. Step S4042: Outline the coastal ecological area and present it to technical personnel using visualization technology; In practice, if the output region merge signal is used, the corresponding sub-analysis region is merged with the main analysis region. The merged region is a coastal ecological region. Each initial region is analyzed. For example, A is adjacent to B but not adjacent to C, while B is adjacent to C. In this case, A and B are merged, and B and C are also merged. In this case, A, B and C are actually the same ecology, that is, A, B and C are merged into the same coastal ecological region. And so on, where A, B and C each represent an initial region.

[0026] Example 2: This application provides an electronic device, which may include a processor, a communication interface, a memory, and a communication bus. The processor, communication interface, and memory communicate with each other via the communication bus. The memory stores computer-readable instructions. The processor can call the instructions in the memory. When the computer-readable instructions are executed by the processor, steps such as those in the coastal ecological region identification method based on remote sensing imagery are performed to achieve the following functions: collecting remote sensing images of the coastal region throughout the year that are not obscured by clouds; performing image preprocessing on the remote sensing images to obtain preprocessed images; identifying the supratidal, intertidal, and subtidal zones of the coastal region using the preprocessed images; analyzing the temporal characteristics of different pixels based on the remote sensing images, and grouping the pixels based on the characteristic lines of change to obtain different initial regions; performing similarity analysis on the initial regions, merging similar initial regions, and finally obtaining the coastal ecological region.

[0027] Furthermore, when the logical instructions in the aforementioned memory can be implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion 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 this application. 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.

[0028] Example 3: This application also provides a computer program product, which includes a computer program stored on a computer-readable storage medium. The computer program includes program instructions. When the program instructions are executed by a computer, the computer can execute the coastal ecological region identification method based on remote sensing imagery provided by the above methods. The method includes: collecting remote sensing images of the coastal region that are not obscured by clouds throughout the year; performing image preprocessing on the remote sensing images to obtain preprocessed images; identifying the supratidal zone, intertidal zone, and subtidal zone of the coastal region through the preprocessed images; analyzing the change feature lines of different pixels based on the temporal sequence of the remote sensing images, and grouping the pixels based on the change feature lines to obtain different initial regions; performing similarity analysis on the initial regions, merging similar initial regions, and finally obtaining the coastal ecological region.

[0029] Example 4: This application also provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, it performs the steps of the above-described method for identifying coastal ecological regions based on remote sensing images to achieve the following functions: collecting remote sensing images of coastal regions throughout the year that are not obscured by clouds; performing image preprocessing on the remote sensing images to obtain preprocessed images; identifying the supratidal, intertidal, and subtidal zones of the coastal region through the preprocessed images; analyzing the temporal characteristics of different pixels based on the remote sensing images, and grouping the pixels based on the characteristic lines of change to obtain different initial regions; performing similarity analysis on the initial regions, merging similar initial regions, and finally obtaining the coastal ecological region.

[0030] Based on the above description of the embodiments, the embodiments of the present invention can be provided as methods, systems, or computer program products. Based on this understanding, the above technical solutions, in essence or in terms of their contribution 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 certain parts of the embodiments.

[0031] In the embodiments provided in this application, it should be understood that the disclosed system or method can be implemented in other ways. The embodiments described above are merely illustrative. For example, the division of modules or units is only a logical functional division, and there may be other division methods in actual implementation. Furthermore, multiple modules or units may be combined or integrated into another system, or some features may be ignored or not executed. Additionally, the coupling or direct coupling or communication connection shown or discussed may be through some communication interfaces. The indirect coupling or communication connection between systems, modules, and units may be electrical, mechanical, or other forms.

[0032] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A method for identifying coastal ecological areas based on remote sensing imagery, characterized in that, Includes the following steps: Collect remote sensing images of the coastal area throughout the year that are not obscured by clouds, and perform image preprocessing on the remote sensing images to obtain preprocessed images; Preprocessed images are used to identify the supratidal, intertidal, and subtidal zones of coastal areas. Based on the temporal analysis of remote sensing images, the characteristic lines of change of different pixels are analyzed, and the pixels are grouped based on the characteristic lines of change to obtain different initial regions; A similarity analysis is performed on the initial regions, and similar initial regions are merged to finally obtain the coastal ecological regions.

2. The coastal ecological area identification method based on remote sensing imagery according to claim 1, characterized in that, Collect remote sensing images of the coastal area throughout the year that are not obscured by clouds, and perform image preprocessing on the remote sensing images to obtain preprocessed images. The preprocessed images include the following sub-steps: Collect all remote sensing images of the coastal area that are not obscured by clouds from January 1 to December 31. Image enhancement is performed on remote sensing images to obtain preprocessed images.

3. The coastal ecological area identification method based on remote sensing imagery according to claim 2, characterized in that, Identifying the supratidal, intertidal, and subtidal zones of a coastal area through preprocessed images includes the following sub-steps: Identify the relative positions of ocean and land in preprocessed images; The analysis of the supratidal, intertidal, and subtidal zones of the coastal area is based on the relative positions of the ocean and land.

4. The coastal ecological area identification method based on remote sensing imagery according to claim 3, characterized in that, Identifying the relative positions of ocean and land in a preprocessed image includes the following sub-steps: The ocean and land in the preprocessed image are identified by image recognition technology. The midpoints of two parallel sides of the four sides of the preprocessed image are connected to obtain dividing lines, which include vertical dividing lines and horizontal dividing lines. The areas above and below the upper and lower dividing lines are named the upper partition and lower partition, respectively; the areas to the left and right of the left and right dividing lines are named the left partition and right partition, respectively. The upper, lower, left, and right partitions are collectively referred to as image partitions. The proportion of ocean to land in the image partitions is calculated and named the ocean-to-land ratio. If the ocean-to-land ratio is greater than or equal to one, the image partition is marked as an ocean area; otherwise, the image partition is marked as a land area. The ocean area and the land area are collectively referred to as ocean-to-land partitions. If the land and sea partitions of the upper and lower partitions are different, the preprocessed image is marked as vertically opposite; if the land and sea partitions of the left and right partitions are different, the preprocessed image is marked as horizontally opposite. If the preprocessed image is both vertically and horizontally relative, then the land-sea ratios of the upper, lower, left, and right partitions are labeled as Q1, Q2, Q3, and Q4, respectively. Calculate (Q1+Q2) / (Q3+Q4). If the calculation result is greater than or equal to one, then the relative positions of the preprocessed image are set to vertically relative; otherwise, the relative positions of the preprocessed image are set to horizontally relative.

5. The coastal ecological area identification method based on remote sensing imagery according to claim 4, characterized in that, Analyzing the supratidal, intertidal, and subtidal zones of a coastal area based on the relative positions of the ocean and land includes the following sub-steps: If the preprocessed image is vertically opposite, then each column of pixels in the preprocessed image is used as a group of tidal zone analysis points; if the preprocessed image is horizontally opposite, then each row of pixels in the preprocessed image is used as a group of tidal zone analysis points. The grayscale values ​​of the pixels in the tidal zone analysis group are numbered according to the direction from ocean to land, using the symbol G. h This indicates that h is a non-zero natural number and h is the index of G; With h as the X-axis, G h Establish a two-dimensional coordinate system for the Y-axis, named the tidal zone analysis diagram, and set G... h Enter the tidal zone analysis diagram according to h, and enter G in the tidal zone analysis diagram. h The corresponding coordinate point is marked as P. h ; P h With P h+1 Connect them, and mark the resulting straight line as PL. h , obtain PL h The slope, denoted as PK h ; PK h Perform cluster analysis to classify PK h Merge into different partitioned clusters, and statistically analyze the PK values ​​in the partitioned clusters. h The number of similarities is named the similarity number. The cluster with the highest similarity number is named the similarity cluster. The PK (primary similarity) in the similarity cluster is then counted. h The maximum value of is represented by the symbol KU; Starting from h=1, determine PK. h Is it less than or equal to KU? If so, then set G. h The corresponding pixel is named an ocean point, and h is incremented by one and analyzed again. If not, the analysis is stopped. After stopping the analysis, the remaining pixels are named land points. Each tidal zone analysis group is analyzed separately to obtain all ocean points and land points. The pixel in the nth row and mth column of the preprocessed image is labeled as PI(n,m). Analyze the PI(n,m) with the same n and m in different preprocessed images. If all PI(n,m) are ocean points, then PI(n,m) is labeled as a subtidal point. If PI(n,m) contains both ocean points and land points, then PI(n,m) is labeled as an intertidal point. If all PI(n,m) are land points, then PI(n,m) is labeled as a supratidal point. The area composed of supratidal points is the supratidal zone, the area composed of intertidal points is the intertidal zone, and the area composed of subtidal points is the subtidal zone.

6. The method for identifying coastal ecological areas based on remote sensing imagery according to claim 5, characterized in that, The temporal analysis of remote sensing images to identify the characteristic lines of change in different pixels, and the grouping of pixels based on these characteristic lines to obtain different initial regions, includes the following sub-steps: Temporal analysis of remote sensing images: characteristic lines of change in different pixels in preprocessed images; The pixels in the preprocessed image are divided into different initial regions based on the changing feature lines.

7. The method for identifying coastal ecological areas based on remote sensing imagery according to claim 6, characterized in that, Temporal analysis of remote sensing images involves preprocessing the image to identify characteristic lines representing changes in different pixels. This process includes the following sub-steps: Assuming the remote sensing image was captured on day d of a year, the preprocessed image corresponding to the remote sensing image will be labeled as PC. d Where d is a non-zero natural number; PC d PI(n,m) in the figure is labeled PC d (n,m), PC d The grayscale value corresponding to (n,m) is labeled as PG. d (n,m); Analyze each value of n and m independently, with d as the horizontal axis, PG d Establish a two-dimensional coordinate system with (n,m) as the vertical axis, and name it the time series variation graph. d (n,m) Enter the time series change graph according to d, perform linear regression on the time series change graph, obtain the slope of the regression function, and name it as the time series change feature. Each PI(n,m) has a time series change feature.

8. The method for identifying coastal ecological areas based on remote sensing imagery according to claim 7, characterized in that, Dividing pixels in the preprocessed image into different initial regions based on changing feature lines includes the following sub-steps: Cluster analysis was performed on the temporal variation characteristics to classify them into different temporal feature clusters; The initial region is formed by merging pixels that are in the same temporal feature cluster and are consecutively adjacent in the preprocessed image.

9. The method for identifying coastal ecological areas based on remote sensing imagery according to claim 8, characterized in that, The process of performing a similarity analysis on the initial regions, merging similar initial regions, and finally obtaining the coastal ecoregions includes the following sub-steps: The average value of the temporal variation characteristics within an initial region is named the regional variation characteristic. When performing similarity analysis on any initial region, it is named the main analysis region, and the initial regions adjacent to the main analysis region are named the secondary analysis regions. PK in similarity clustering h The maximum value is named the same-region threshold; Obtain the regional change characteristics of the main analysis region and the sub-analysis region, and label them as R1 and R2 respectively. Calculate |R1-R2|. If the calculation result is less than or equal to the same region threshold, output the region merged signal; otherwise, output the region unmerged signal. The main and sub-analytical regions are processed based on regional merging and regional dismerging signals. Similar initial regions are merged to obtain the final coastal ecological region.

10. The method for identifying coastal ecological areas based on remote sensing imagery according to claim 9, characterized in that, Based on the regional merging and regional dismerging signals, the main analysis region and the sub-analysis region are processed, and similar initial regions are merged to finally obtain the coastal ecological region, including the following sub-steps: If the output region merge signal is used, the corresponding sub-analysis region is merged with the main analysis region. The merged region is a coastal ecological region. Each initial region is analyzed, and finally different coastal ecological regions are obtained by merging. The coastal ecological area is outlined and displayed to technical personnel using visualization technology.

Images