A geological survey data management system and method for nearshore zones

Abstract

The application discloses a kind of offshore zone Geological Survey Data Management System and method, it is related to shoreline monitoring technical field, including the following steps: collecting offshore zone remote sensing image, and remote sensing image pre-processing is carried out, obtains first coast remote sensing image;Based on the first coast remote sensing image, the remote sensing image of water-land contact area is extracted, obtains second coast remote sensing image, and carries out first shoreline extraction processing, obtains first shoreline information;Based on the second coast remote sensing image, carries out second shoreline extraction processing, obtains second shoreline information;Based on the first shoreline information and second shoreline information, carries out shoreline correction processing, obtains real-time shoreline information;The present application is used to solve the problem that the existing shoreline monitoring technology cannot accurately distinguish seawater and tidal flat in the process of ebb and flow when distinguishing water and land in image based on remote sensing image and spectral index, accurately identifies the problem of shoreline.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of coastline monitoring, in particular to a nearshore zone geological survey data management system and method. BACKGROUND

[0002] Coastline monitoring technology refers to a comprehensive technical system that uses various technical means to continuously observe, collect, analyze and evaluate the spatial position, morphological characteristics and dynamic changes of the coastline. Its core goal is to master the evolution law of the coastline and provide scientific basis for coastline resource management, ecological protection, disaster prevention and reduction, and marine economic development.

[0003] When the existing coastline monitoring technology distinguishes water and land in the image based on remote sensing images and spectral indices, it cannot accurately distinguish seawater and tidal flats. Because the tidal flats are rich in water, fine-grained silt and organic matter, their spectral reflectance in the visible and near-infrared bands overlaps significantly with seawater. Traditional water indices, such as NDWI, rely on the characteristic that "water has low reflectance in the near-infrared band", but the reflectance of wet silt in the near-infrared band may be lower than that of dry land, leading to misjudgment as water by threshold method. Moreover, during the rising and falling tide process, the tidal flat silt undergoes continuous changes of dry, wet and submerged, and its spectral characteristics fluctuate dynamically with water content, which will cause errors in the identification of the coastline. For example, at the beginning of the falling tide, wet sand is misjudged as water, and the coastline lags behind the actual position. In the later stage of the falling tide, dry sand is correctly identified as land, but the intermediate transition zone is identified ambiguously. Moreover, using spectral indices to distinguish water and land in the image relies on a globally fixed threshold, which cannot adapt to the rapid changes in silt spectrum. For example, the patent application with publication number CN113221813A discloses a coastline remote sensing extraction method, which distinguishes water and land in the image by spectral indices, and cannot accurately distinguish seawater and tidal flats. Therefore, when the existing coastline monitoring technology distinguishes water and land in the image based on remote sensing images and spectral indices, it cannot accurately distinguish seawater and tidal flats during the rising and falling tide process, and accurately identify the coastline. SUMMARY

[0004] The present application aims to at least solve one of the technical problems in the prior art by collecting remote sensing images of the near coastal zone, and performing remote sensing image preprocessing to obtain a first coastal remote sensing image; extracting remote sensing images of the water-land contact area based on the first coastal remote sensing image to obtain a second coastal remote sensing image, and performing first coastline extraction processing to obtain first coastline information; performing second coastline extraction processing based on the second coastal remote sensing image to obtain second coastline information; and performing coastline correction processing to obtain real-time coastline information; to solve the problem that the existing coastline monitoring technology cannot accurately distinguish seawater and tidal flat during the rising and falling tide when distinguishing water and land in the image based on remote sensing images and spectral index, and accurately identifying the coastline.

[0005] To achieve the above-mentioned purpose, in a first aspect, the present application provides a geological survey data management system for the near coastal zone, comprising a remote sensing acquisition module, a first extraction module, a second extraction module, and a coastline monitoring module.

[0006] The remote sensing acquisition module comprises an acquisition unit and a preprocessing unit, the acquisition unit is configured to acquire remote sensing images of the near coastal zone, and the preprocessing unit is configured to perform remote sensing image preprocessing to obtain a first coastal remote sensing image.

[0007] The first extraction module extracts remote sensing images of the water-land contact area based on the first coastal remote sensing image to obtain a second coastal remote sensing image, and performs first coastline extraction processing to obtain first coastline information.

[0008] The second extraction module performs second coastline extraction processing based on the second coastal remote sensing image to obtain second coastline information.

[0009] The coastline monitoring module performs coastline correction processing based on the first coastline information and the second coastline information to obtain real-time coastline information.

[0010] Further, the acquisition unit is configured with an acquisition strategy, which comprises:

[0011] The near coastal zone to be monitored is denoted as a monitoring coastal area, when the monitoring coastal area is rising or falling, the remote sensing images of the monitoring coastal area are collected by a UAV at a first time interval, the collection position, collection height, and setting parameters of the collection equipment of the remote sensing images of the monitoring coastal area are ensured to be consistent each time, the collection time is recorded, and the collected remote sensing images are sorted according to the collection time and marked as original remote sensing images, wherein the first time interval is t1.

[0012] Further, the preprocessing unit is configured with a preprocessing strategy, which comprises:

[0013] For any one of the original remote sensing image is recorded as the first remote sensing image; Set the first division grid size a1*a1; The first remote sensing image is divided into a plurality of grid regions by using the first division grid, and any one grid region is recorded as the first grid region;

[0014] The average value and the standard deviation of the brightness of all pixel points in the first grid region are calculated, and are recorded as PE0 and PB0 in order respectively;

[0015] The first neighborhood size is set as b1*b1; The region with the first grid region as the center and the first neighborhood size around is recorded as the similar grid neighborhood of the first grid region, and any one grid region in the similar grid neighborhood of the first grid region is recorded as the neighborhood grid region;

[0016] The average value and the standard deviation of the brightness of all pixel points in the neighborhood grid region are obtained, and are recorded as PE1 and PB1 in order respectively; and the feature similarity value of the first grid region and the neighborhood grid region is calculated according to the first formula, and the first formula is as follows: Wherein, TX represents the feature similarity value;

[0017] The feature similarity values of all grid regions in the similar grid neighborhood of the first grid region and the first grid region are repeatedly obtained, and are sorted according to the size, and the grid regions corresponding to the smallest k1 feature similarity values are marked as the similar grid regions of the first grid region.

[0018] Further, the preprocessing strategy further comprises:

[0019] The total number of pixel points in the first grid region and the corresponding similar grid region is obtained and recorded as ZN0, and the number of pixel points of each brightness level is obtained and recorded as H[m], wherein m represents the brightness level;

[0020] The first cutoff ratio is set as e1, and the second cutoff ratio is set as e2, e2>e1; ZN1 and ZN2 are calculated, wherein ZN1=e1*ZN0, ZN2=e2*ZN0;

[0021] From the smallest brightness level, the corresponding pixel point number is accumulated in the direction of increasing brightness level until the accumulated pixel point number is greater than or equal to ZN1, and the maximum brightness level accumulated at this time is obtained and recorded as the underexposure threshold QB;

[0022] From the largest brightness level, the corresponding pixel point number is accumulated in the direction of decreasing brightness level until the accumulated pixel point number is greater than or equal to ZN2, and the minimum brightness level accumulated at this time is obtained and recorded as the overexposure threshold GB;

[0023] arranging the brightness of all the pixel points in the first grid region in ascending order to form a first brightness sequence, calculating the average value of the brightness of the k2th percentile to the k3th percentile in the first brightness sequence as a low-brightness average value DL, and calculating the average value of the brightness of the k4th percentile to the k5th percentile in the first brightness sequence as a high-brightness average value GL, wherein k2, k3, k4 and k5 are set percentile numbers, and k2 < k3 < k4 < k5;

[0024] If DL is greater than or equal to QB and GL is greater than GB, the first grid region is marked as an overexposure region; if DL is less than QB and GL is less than GB, the first grid region is marked as an underexposure region; if DL is less than QB and GL is greater than GB, the first grid region is marked as an over-underexposure region, otherwise the first grid region is marked as a normal region; and the marking is repeated for all the grid regions in the first remote sensing image;

[0025] Gaussian kernels with corresponding standard deviations σ1, σ2 and σ3 are set, and are denoted as a small kernel, a medium kernel and a large kernel respectively, wherein σ1 < σ2 < σ3;

[0026] Gaussian filtering is performed on all the underexposure regions based on the small kernel, Gaussian filtering is performed on all the overexposure regions based on the large kernel, and Gaussian filtering is performed on all the over-underexposure regions and the normal regions based on the medium kernel; after completion, a preprocessed image corresponding to the first remote sensing image is obtained; and the processing is repeated for all the remote sensing images in the original remote sensing image, and after completion, the first coastal remote sensing image is obtained.

[0027] Further, the first extraction module is configured with a first extraction strategy, and the first extraction strategy comprises:

[0028] For any one image in the first coastal remote sensing image, denoted as a second remote sensing image; and the collection time corresponding to the second remote sensing image is denoted as a second collection time, the last collection time of the second collection time is denoted as a third collection time, and the preprocessed image corresponding to the third collection time is denoted as a third remote sensing image;

[0029] A second division grid size a2*a2 is set; and the second remote sensing image and the third remote sensing image are divided into a plurality of grid regions by using the second division grid;

[0030] The gray scale range 0-255 is uniformly divided into k6 gray scale intervals; and the intervals corresponding to all the pixel points in the second remote sensing image and the third remote sensing image are counted.

[0031] Further, the first extraction strategy further comprises:

[0032] For any one pixel point in the third remote sensing image, denoted as a first pixel point, if the gray scale level of the first pixel point in the third remote sensing image is different from the gray scale level in the second remote sensing image, the first pixel point is marked as a boundary pixel point, and the marking is repeated for all pixel points in the third remote sensing image;

[0033] For any one grid region in the third remote sensing image, if there is a boundary pixel point, the corresponding grid region is marked as a water-land contact region, and the water-land contact regions in the third remote sensing image are repeatedly obtained, and after completion, the water-land contact image corresponding to the third remote sensing image is obtained; the water-land contact images corresponding to all images in the first coastal remote sensing image are repeatedly obtained, and after completion, the second coastal remote sensing image is obtained;

[0034] For all boundary pixel points, the connection order of the boundary pixel points in the shortest connection path is calculated, and all boundary pixel points are connected in turn according to the connection order, and the coastline corresponding to the third acquisition time and the third remote sensing image is obtained, and the coastlines corresponding to all images in the first coastal remote sensing image are repeatedly obtained, and after completion, the first coastline information is obtained.

[0035] Further, the second extraction module is configured with a second extraction strategy, and the second extraction strategy includes:

[0036] For any one image in the second coastal remote sensing image, denoted as a fourth remote sensing image, for any one water-land contact region in the fourth remote sensing image, denoted as a water-land grid region;

[0037] The second neighborhood size is set to b2*b2; the region centered on the water-land grid region and adding the region of the second neighborhood size is denoted as the neighborhood grid region of the water-land grid region; the green band reflectivity GR, the short-wave infrared band reflectivity SW, the red band reflectivity RE, and the blue band reflectivity BL of all pixel points in the neighborhood grid region are obtained, and the comprehensive index of all pixel points in the neighborhood grid region is calculated according to a second formula, and the second formula is as follows Wherein CI is the comprehensive index, q1 and q2 are weight coefficients;

[0038] According to the comprehensive index of all pixel points in the neighborhood grid region, the OTSU algorithm is used to calculate the best comprehensive index threshold, denoted as YZ; for any one pixel point in the water-land grid region, denoted as a water-land pixel point, if the CI of the water-land pixel point is greater than YZ, the water-land pixel point is marked as a water-like pixel point, otherwise the water-land pixel point is marked as a land pixel point;

[0039] All water-like pixel points and land pixel points in all water-land contact regions in the fourth remote sensing image are repeatedly obtained.

[0040] Further, the second extraction strategy further includes:

[0041] For any one water-like pixel, if the number of land-like pixel points in the 8-neighborhood is greater than k6, it is marked as a land pixel point, otherwise it is marked as a water body pixel point;

[0042] For any one land-like pixel point, if the number of water-like pixel points in the 8-neighborhood is greater than k6, it is marked as a water body pixel point, otherwise it is marked as a land pixel point;

[0043] Repeat the process of obtaining all water body pixel points and land pixel points in all water-land contact areas in the fourth remote sensing image;

[0044] Use the connected component labeling algorithm to obtain all connected regions composed of water body pixel points or land pixel points in the fourth remote sensing image. Mark them as water body regions and land regions in order, obtain the water body pixel points at the boundary between the water body regions and the land regions, and mark them as edge pixel points;

[0045] For all edge pixel points, calculate the connection order of each edge pixel point under the condition of the shortest connection path, and connect all edge pixel points in order according to the connection order to obtain the coastline corresponding to the fourth remote sensing image. Repeat the process of obtaining the coastline corresponding to the second coastal remote sensing image, and complete the process to obtain the second coastline information.

[0046] Further, the coastal monitoring module is configured with a coastal monitoring strategy, which includes:

[0047] For any one of the first coastline information and the second coastline information corresponding to the remote sensing image collected at the collection time, mark them as the first coastline and the second coastline in order;

[0048] Select k7 corresponding nodes from the first coastline and the second coastline, obtain the coordinates of each group of corresponding nodes in the corresponding remote sensing image, and calculate the midpoint coordinates of each group of corresponding nodes. Connect all the midpoint coordinates obtained in the corresponding order to obtain the real-time coastline information at the corresponding collection time.

[0049] In a second aspect, the application provides a method for managing geological survey data of a nearshore zone, which includes the following steps:

[0050] Collect remote sensing images of the nearshore zone and perform remote sensing image preprocessing to obtain a first coastal remote sensing image;

[0051] Extract the remote sensing image of the water-land contact area based on the first coastal remote sensing image to obtain a second coastal remote sensing image, and perform first coastline extraction processing to obtain first coastline information;

[0052] Perform second coastline extraction processing based on the second coastal remote sensing image to obtain second coastline information

[0053] Based on the first and second coastline information, coastline correction processing is performed to obtain real-time coastline information.

[0054] The beneficial effects of this invention are as follows: This invention acquires remote sensing images of the near-shore zone and performs remote sensing image preprocessing to obtain a first coastal remote sensing image; based on the first coastal remote sensing image, it extracts remote sensing images of the land-water contact area to obtain a second coastal remote sensing image, and performs first coastline extraction processing to obtain first coastline information; based on the second coastal remote sensing image, it performs second coastline extraction processing to obtain second coastline information; based on the first coastline information and the second coastline information, it performs coastline correction processing to obtain real-time coastline information; when distinguishing water bodies and land in the image based on remote sensing images and spectral indices, it can accurately distinguish seawater and tidal flats during high and low tides, and accurately identify the coastline.

[0055] This invention divides images into underexposed, overexposed, overexposed / underexposed, and normal regions by acquiring similar grid areas and their corresponding feature values. Differentiated Gaussian filtering is applied to different regions to effectively improve the local overexposure and underexposure problems caused by uneven illumination and sensor noise in remote sensing images, thereby enhancing the clarity of land-water boundaries. The first coastline is detected based on temporal changes in boundary pixels, while the second coastline is separated based on spectral indices and OTSU thresholds for block-based mapping. This combines temporal change features with spectral features, avoiding the limitations of relying solely on spectral characteristics in a single threshold method, reducing noise interference from a single method, and improving the robustness of coastline positioning. Secondary correction of pixels is performed through mutual inspection of 8 neighboring pixels, effectively suppressing salt-and-pepper noise and isolated pixel interference, making waterway boundaries more accurate. Attached Figure Description

[0056] Figure 1 This is a schematic diagram of the system of the present invention;

[0057] Figure 2 This is a flowchart illustrating the steps of the method of the present invention;

[0058] Figure 3 This is a schematic diagram of the similar grid neighborhood of the present invention;

[0059] Figure 4 This is a schematic diagram of the neighborhood grid region of the present invention;

[0060] Figure 5 This is a schematic diagram of the structure of the electronic device of the present invention; Detailed Implementation

[0061] 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.

[0062] Example 1, please refer to Figure 1 As shown, this application provides a geological survey data management system for near-coastal zones, including a remote sensing acquisition module, a first extraction module, a second extraction module, and a coastal monitoring module;

[0063] The remote sensing acquisition module includes an acquisition unit and a preprocessing unit. The acquisition unit is used to acquire remote sensing images of the near-coastal zone, and the preprocessing unit is used to preprocess the remote sensing images to obtain the first coastal remote sensing image.

[0064] The acquisition unit is configured with an acquisition strategy, which includes: designating the near-coastal zone to be monitored as the monitoring coastal area; during high or low tide in the monitoring coastal area, using a drone to acquire remote sensing images of the monitoring coastal area at a first time interval, ensuring that the acquisition position, acquisition height, and acquisition device settings are consistent for each acquisition of remote sensing images of the monitoring coastal area, and ensuring that the coordinate position of the same geographical location in images acquired at different times is consistent; and recording the acquisition time, and sorting the acquired remote sensing images according to the acquisition time, marking them as original remote sensing images, wherein the first time interval is t1; the first time interval can be set according to the actual application scenario, in this embodiment, the first time interval t1 is 2 minutes;

[0065] The preprocessing unit is configured with a preprocessing strategy, which includes: denoting any one of the original remote sensing images as the first remote sensing image; setting the size of the first dividing grid to a1*a1; dividing the first remote sensing image into multiple grid regions using the first dividing grid, and denoting any one of the grid regions as the first grid region; in this embodiment, a1 is 16 pixels.

[0066] Calculate the average brightness and standard deviation of all pixels in the first grid area, and denote them as PE0 and PB0 respectively in order;

[0067] Please see Figure 3 As shown, the size of the first neighborhood is set to b1*b1; the area centered on the first grid region and surrounding the first neighborhood is denoted as the similar grid neighborhood of the first grid region, and any grid region in the similar grid neighborhood of the first grid region is denoted as the neighboring grid region; in this embodiment, b1 is 5;

[0068] The average value and the standard deviation of the brightness of all the pixel points in the neighborhood grid area are obtained, and are denoted as PE1 and PB1 in sequence respectively; and the characteristic similarity value of the first grid area and the neighborhood grid area is calculated according to the first formula, and the first formula is as follows: Wherein, TX represents the characteristic similarity value;

[0069] The characteristic similarity values of all the grid areas in the similar grid neighborhood of the first grid area and the first grid area are repeatedly obtained, and are sorted according to the size, and the grid areas corresponding to the smallest k1 characteristic similarity values are marked as the similar grid areas of the first grid area; the smaller the characteristic similarity value is, the smaller the difference between the average value and the standard deviation of the brightness of the two areas is, that is, the smaller the characteristic difference of the two areas is; in this embodiment, k1 = 5;

[0070] The total number of the pixel points in the first grid area and the corresponding similar grid area is obtained, and is denoted as ZN0, and the number of the pixel points of each brightness level is obtained, and is denoted as H[m], wherein m represents the brightness level; for example, H[0] represents the number of the pixel points with the brightness level of 0;

[0071] The first truncation ratio is set as e1, and the second truncation ratio is set as e2, and e2>e1; ZN1 and ZN2 are calculated, wherein ZN1 = e1*ZN0, and ZN2 = e2*ZN0; in this embodiment, e1 = e2 = 2%;

[0072] From the smallest brightness level, the corresponding number of the pixel points is accumulated in the direction of increasing the brightness level, until the accumulated number of the pixel points is greater than or equal to ZN1, and the largest brightness level accumulated at this time is obtained, and is denoted as the underexposure threshold QB; for example, ZN0 = 1280, ZN1 = 25.6; H[0] + H[1] + H[2] + H[3] + H[4] = 26, and then QB = 4;

[0073] From the largest brightness level, the corresponding number of the pixel points is accumulated in the direction of decreasing the brightness level, until the accumulated number of the pixel points is greater than or equal to ZN2, and the smallest brightness level accumulated at this time is obtained, and is denoted as the overexposure threshold GB; for example, ZN0 = 1280, ZN2 = 25.6; H

[255] + H

[254] + H

[254] + H

[253] = 26, and then QB = 253;

[0074] Arranging the brightness of all pixel points in the first grid area in ascending order to form a first brightness sequence, calculating the average value of the brightness of the k2th percentile to the k3th percentile in the first brightness sequence as a low-brightness average value DL, and calculating the average value of the brightness of the k4th percentile to the k5th percentile in the first brightness sequence as a high-brightness average value GL, wherein k2, k3, k4 and k5 are set percentile numbers, and k2 < k3 < k4 < k5; in this embodiment, k2 = 10, k3 = 40, k4 = 50 and k5 = 90; by calculating the average values of the brightness of the k2th to k3th percentiles and the k4th to k5th percentiles, the interference of extreme pixels in the image can be excluded, and only the brightness distribution of the main pixels is analyzed;

[0075] If DL ≥ QB and GL > GB, it indicates that there are few low-brightness pixels in the region, the overall brightness is high, and the proportion of high-brightness pixels is high, and the first grid area is marked as an overexposure region;

[0076] If DL < QB and GL < GB, it indicates that the overall region is dark, and the first grid area is marked as an underexposure region;

[0077] If DL < QB and GL > GB, it indicates that the brightness distribution in the region is uneven, and the first grid area is marked as an over-underexposure region, otherwise, the first grid area is marked as a normal region; the marking of all grid areas in the first remote sensing image is repeated;

[0078] A Gaussian kernel with corresponding standard deviations σ1, σ2 and σ3 is set, and is denoted as a small kernel, a medium kernel and a large kernel respectively, wherein σ1 < σ2 < σ3; in this embodiment, σ3 = 5.0, σ2 = 2.5 and σ1 = 1.0;

[0079] Gaussian filtering is performed on all underexposure regions based on the small kernel, on all overexposure regions based on the large kernel, and on all over-underexposure regions and normal regions based on the medium kernel; after completion, a preprocessed image corresponding to the first remote sensing image is obtained; the processing of all remote sensing images in the original remote sensing image is repeated, and after completion, the first coastal remote sensing image is obtained;

[0080] The underexposure region usually has a problem of mixed noise and details, the small kernel can avoid excessive blurring of details and prevent loss of features of ground objects due to filtering, the Gaussian filtering window of the small kernel is small, the weighted average range of pixel values is limited, and therefore the smoothing strength is weak, which can maximize the retention of image details while improving the brightness of the underexposure region; the Gaussian kernel window of the medium kernel is moderate, and noise suppression and detail retention are considered, which is suitable for regions with relatively uniform or mixed exposure;

[0081] In the specific implementation process, the similar regions are screened based on the neighborhood grid-based brightness mean value and standard deviation, the dynamic association of homogeneous regions is realized, the limitations of single threshold division are avoided, the reliability of subsequent judgment is improved, the image is divided into underexposed, overexposed, over-underexposed and normal regions, and differential Gaussian filtering is adopted for different regions; the local overexposure and underexposure problems of remote sensing images caused by uneven illumination and sensor noise are effectively improved, the image quality is improved, the clarity of the water-land boundary is improved, and the noise interference is suppressed, thereby providing a high-precision data basis for subsequent coastline extraction and other remote sensing analysis.

[0082] The first extraction module extracts a remote sensing image of a water-land contact area based on the first coastal remote sensing image, obtains a second coastal remote sensing image, and performs a first coastline extraction process to obtain first coastline information.

[0083] The first extraction module is configured with a first extraction strategy, and the first extraction strategy includes: for any one image in the first coastal remote sensing image, denoted as a second remote sensing image; and the collection time corresponding to the second remote sensing image is denoted as a second collection time, the collection time one time earlier than the second collection time is denoted as a third collection time, and the preprocessed image corresponding to the third collection time is denoted as a third remote sensing image; for example, the first time interval t1 is 2 minutes, and the image collected and preprocessed after 11:30 is denoted as the second remote sensing image, and then the image collected and preprocessed after 11:28 is denoted as the third remote sensing image.

[0084] The second division grid size is set to a2*a2; the second remote sensing image and the third remote sensing image are divided into a plurality of grid regions by using the second division grid; in this embodiment, a2 is 128 pixel points.

[0085] The gray scale range 0-255 is evenly divided into k6 gray scale intervals; the intervals corresponding to all pixel points in the second remote sensing image and the third remote sensing image are counted; in this embodiment, k6 = 128, that is, every two gray scale levels are an interval; by mapping the continuous gray scale values to discrete intervals, the slight gray scale fluctuation can be classified into the same level, and the result deviation caused by the slight change of individual pixels or features in the subsequent process is avoided.

[0086] For any one pixel point in the third remote sensing image, denoted as a first pixel point, if the gray scale interval of the first pixel point in the third remote sensing image is different from the gray scale interval in the second remote sensing image, the first pixel point is marked as a boundary pixel point, and the marking of all pixel points in the third remote sensing image is repeated; that is, the pixel point has changed in the images collected continuously two times.

[0087] For any one grid region in the third remote sensing image, if there is a boundary pixel point, mark the corresponding grid region as a water-land contact region, repeat to obtain all water-land contact regions in the third remote sensing image, and after completion, obtain the water-land contact image corresponding to the third remote sensing image; repeat to obtain the water-land contact image corresponding to all images in the first coastal remote sensing image, and after completion, obtain the second coastal remote sensing image; the water-land contact region is the junction region of seawater and land, and subsequent processing can be mainly performed on this part of the region, so that the calculation amount can be reduced and the analysis efficiency can be improved;

[0088] For all boundary pixel points, the connection order of each boundary pixel point in the shortest connection path is calculated, and all boundary pixel points are sequentially connected according to the connection order, to obtain the coastline corresponding to the third acquisition moment and the third remote sensing image, repeat to obtain the coastline corresponding to all images in the first coastal remote sensing image, and after completion, obtain the first coastline information; other suitable methods for fitting the coastline can also be selected according to actual application scenarios;

[0089] In the specific implementation process, during the rising tide and the falling tide, the water-land boundary will change significantly, the ground object type at the same position will change from land to water or from water to land, and the water and the land are embodied as different spectral reflectances on the remote sensing image, corresponding to different gray scales; by comparing the gray scales of the same pixel point in adjacent moments of remote sensing images, if there is a difference in the gray scale, it means that the ground object type at the position of the pixel point changes, that is, it is in the water-land junction region, and therefore the pixel point is marked as a boundary pixel point.

[0090] The second extraction module performs second coastline extraction processing based on the second coastal remote sensing image to obtain second coastline information;

[0091] The second extraction module is configured with a second extraction strategy, and the second extraction strategy includes: for any one image in the second coastal remote sensing image, denoted as a fourth remote sensing image, for any one water-land contact region in the fourth remote sensing image, denoted as a water-land grid region;

[0092] Please refer to Figure 4 As shown in the figure, the second neighborhood size is set as b2*b2; the region with the waterway grid region as the center and the surrounding second neighborhood size is denoted as the neighborhood grid region of the waterway grid region; the green light band reflectivity GR, the short wave infrared band reflectivity SW, the red light band reflectivity RE, and the blue light band reflectivity BL of all pixel points in the neighborhood grid region are obtained, and the comprehensive index of all pixel points in the neighborhood grid region is calculated according to the second formula, and the second formula is as follows Wherein CI is the comprehensive index, q1 and q2 are weight coefficients; in this embodiment, b2=3, q1=0.6, and q2=0.5;

[0093] (GR-SW) / (GR+SW) in the second formula represents a normalized water index, water has a high reflectivity in the green light band and a very low reflectivity (strong absorption) in the near-infrared band, so the NDWI value is close to 1, while land has a high reflectivity in the near-infrared band, and the NDWI value is close to -1-0; and (RE-BL) / (RE+BL) represents a normalized sediment index, water has a high reflectivity in the blue light band and a very high reflectivity in the red light band, so the NSI is negative; the scattering of sediment to red light is stronger than that to blue light, so the NSI is positive; and because the typical value of water in the normalized water index is positive and the typical value of water in the normalized sediment index is negative, direct weighted summation will weaken the distinguishing ability due to the difference in the numerical signs of the two, so the opposite of one of the indexes is taken to enhance the response of the comprehensive index to seawater, so the weighted summation of the two uses a negative sign;

[0094] According to the comprehensive index of all pixel points in the neighborhood grid area, the OTSU algorithm is used to calculate the best comprehensive index threshold, denoted as YZ; for any one pixel point in the waterway grid area, denoted as a water-land pixel point, if the CI of the water-land pixel point is greater than YZ, the water-land pixel point is marked as a water-like pixel point, otherwise the water-land pixel point is marked as a land pixel point; other threshold determination methods can also be selected according to actual application scenarios;

[0095] All water-like pixel points and land pixel points in all water-land contact areas in the fourth remote sensing image are repeatedly obtained;

[0096] For any one water-like pixel point, if the number of land-like pixel points in the surrounding 8-neighborhood is greater than k6, it is marked as a land pixel point, otherwise it is marked as a water pixel point;

[0097] For any one land-like pixel point, if the number of water-like pixel points in the surrounding 8-neighborhood is greater than k6, it is marked as a water pixel point, otherwise it is marked as a land pixel point; in this embodiment, k6 is 6, that is, if the number of land-like pixel points in the surrounding 8-neighborhood of the water-like pixel point is 7 or 8, it is marked as a land pixel point; if the number of land-like pixel points in the surrounding 8-neighborhood of the land-like pixel point is 7 or 8, it is marked as a water pixel point; through mutual inspection of 8-neighborhood pixel points, secondary correction is performed according to the proportion of neighborhood pixels of the same type, which effectively suppresses pepper and salt noise and isolated pixel interference, makes the water-land edge smoother and more continuous, and is more accurate;

[0098] All water pixel points and land pixel points in all water-land contact areas in the fourth remote sensing image are repeatedly obtained;

[0099] Connected component labeling algorithms are used to obtain all connected regions composed of water pixel points or land pixel points in the fourth remote sensing image; in order, they are denoted as water regions and land regions, water pixel points at the junction of the water regions and the land regions are obtained, denoted as edge pixel points;

[0100] For all edge pixels, the connection order of each edge pixel in the shortest connection path is calculated, and all edge pixels are sequentially connected in the connection order to obtain the coastline corresponding to the fourth remote sensing image. The coastlines corresponding to all images in the second coastal remote sensing image are repeatedly obtained, and after completion, the second coastline information is obtained. Alternatively, other suitable methods for fitting the coastline can be selected according to the actual application scenario.

[0101] In the specific implementation process, the traditional method uses a globally fixed spectral threshold to separate water bodies and tidal flats, which cannot adapt to different lighting conditions and tidal dynamic change scenes, and is prone to cause misclassification. By obtaining a neighborhood grid region and separating water bodies and land based on the spectral features of the neighborhood grid region, the local spectral features of the water-land contact area can be focused on, the spectral fluctuations caused by tides can be dynamically adapted, the failure problem of the globally fixed threshold in different tidal stages can be avoided, and the analysis of subtle features by global statistics can be avoided, thereby more accurately distinguishing water bodies and tidal flats.

[0102] The coastline monitoring module performs coastline correction processing based on the first coastline information and the second coastline information to obtain real-time coastline information.

[0103] The coastline monitoring module is configured with a coastline monitoring strategy, which includes: for the first coastline information and the second coastline information corresponding to the remote sensing image collected at any collection time, the first coastline and the second coastline are sequentially recorded respectively; that is, at the same collection time and at the same position, two coastlines extracted by the first coastline extraction processing and the second coastline extraction processing;

[0104] From the first coastline and the second coastline, k7 corresponding nodes are selected, the coordinates of each group of corresponding nodes in the corresponding remote sensing image are obtained, and the midpoint coordinates of each group of corresponding nodes are calculated. All midpoint coordinates obtained are sequentially connected in the corresponding order to obtain the real-time coastline information at the corresponding collection time. Alternatively, other suitable methods for merging the two coastlines can be selected according to the actual application scenario;

[0105] In the specific implementation process, by extracting two coastlines, the first coastline detects boundary pixels based on time sequence change, and the second coastline is based on spectral index, which combines time sequence change characteristics and spectral characteristics, avoids the limitation of single threshold method relying only on spectral characteristics, and uses midpoint coordinate average method to fuse two types of coastlines, reduces the noise interference of single method, and improves the robustness of coastline positioning.

[0106] Embodiment 2, please refer to Figure 2 As shown in the figure, the application provides a nearshore zone geological survey data management method, including the following steps:

[0107] Step S1, collecting remote sensing images of the nearshore zone and performing remote sensing image preprocessing to obtain a first coastal remote sensing image; step S1 includes the following substeps:

[0108] Step S101, recording the nearshore zone to be monitored as a monitoring coastal area, using a UAV to collect remote sensing images of the monitoring coastal area at a first time interval when the monitoring coastal area is experiencing a rising tide or a falling tide, ensuring that the collection position, collection height, and setting parameters of the collection equipment for each collection of remote sensing images of the monitoring coastal area remain consistent, recording the collection time, and sorting the collected remote sensing images according to the collection time and marking them as original remote sensing images, wherein the first time interval is t10;

[0109] Step S102, recording any one of the original remote sensing images as a first remote sensing image; setting the first division grid size as a1*a1; dividing the first remote sensing image into multiple grid areas using the first division grid, and recording any one of the grid areas as a first grid area;

[0110] Step S103, calculating the average value and the standard deviation of the brightness of all pixel points in the first grid area, and recording them in order as PE0 and PB0, respectively;

[0111] Step S104, setting the first neighborhood size as b1*b1; recording the area with the first neighborhood size around the center of the first grid area as the similar grid neighborhood of the first grid area, and recording any one of the grid areas in the similar grid neighborhood of the first grid area as a neighborhood grid area;

[0112] Step S105, obtaining the average value and the standard deviation of the brightness of all pixel points in the neighborhood grid area, and recording them in order as PE1 and PB1, respectively; and calculating the feature similarity value of the first grid area and the neighborhood grid area according to the first formula, which is as follows: wherein TX represents the feature similarity value;

[0113] Step S106, repeatedly obtaining the feature similarity values of all grid areas in the similar grid neighborhood of the first grid area and the first grid area, sorting them according to the size, and marking the grid areas corresponding to the smallest k1 feature similarity values as the similar grid areas of the first grid area;

[0114] Step S107, obtaining the total number of pixel points in the first grid area and the corresponding similar grid area, denoted as ZN0, and obtaining the number of pixel points of each brightness level, denoted as H[m], wherein m represents the brightness level;

[0115] Step S108, set the first truncation ratio as e1 and the second truncation ratio as e2, e2>e1; calculate ZN1 and ZN2, wherein ZN1=e1*ZN0, ZN2=e2*ZN0;

[0116] Step S109, start from the smallest brightness level, accumulate the corresponding pixel point number in the direction of increasing brightness level until the accumulated pixel point number is greater than or equal to ZN1, and obtain the maximum brightness level at this time, denoted as the underexposure threshold QB;

[0117] Step S110, start from the maximum brightness level, accumulate the corresponding pixel point number in the direction of decreasing brightness level until the accumulated pixel point number is greater than or equal to ZN2, and obtain the minimum brightness level at this time, denoted as the overexposure threshold GB;

[0118] Step S111, arrange the brightness of all pixel points in the first grid area in ascending order, denoted as the first brightness sequence, calculate the average value of the brightness of the k2th percentile to the k3th percentile in the first brightness sequence, denoted as the low brightness average value DL, and calculate the average value of the brightness of the k4th percentile to the k5th percentile in the first brightness sequence, denoted as the high brightness average value GL, wherein k2, k3, k4 and k5 are set percentage digits, k2<k3<k4<k5;

[0119] Step S112, if DL≥QB and GL>GB, mark the first grid area as an overexposure area; if DL<QB and GL<GB, mark the first grid area as an underexposure area; if DL<QB and GL>GB, mark the first grid area as an over-underexposure area, otherwise mark the first grid area as a normal area; repeat the marking for all grid areas in the first remote sensing image;

[0120] Step S113, distribute the Gaussian kernel with corresponding standard deviations σ1, σ2 and σ3, denoted as small kernel, medium kernel and large kernel, respectively, wherein σ1<σ2<σ3;

[0121] Step S114, for all underexposure areas, perform Gaussian filtering based on the small kernel, for all overexposure areas, perform Gaussian filtering based on the large kernel, and for all over-underexposure areas and normal areas, perform Gaussian filtering based on the medium kernel; after completion, obtain the preprocessed image corresponding to the first remote sensing image; repeat the processing for all remote sensing images in the original remote sensing image, and after completion, obtain the first coastal remote sensing image.

[0122] Step S2, extract the remote sensing image of the water-land contact area based on the first coastal remote sensing image to obtain a second coastal remote sensing image, and perform first coastline extraction processing to obtain first coastline information; step S2 includes the following substeps:

[0123] Step S201, for any one image in the first coastal remote sensing image, record it as a second remote sensing image; and record the collection time corresponding to the second remote sensing image as a second collection time, record the last collection time of the second collection time as a third collection time, and record the preprocessed image corresponding to the third collection time as a third remote sensing image;

[0124] Step S202, set the second division grid size as a2*a2; divide the second remote sensing image and the third remote sensing image into a plurality of grid regions by using the second division grid;

[0125] Step S203, evenly divide the gray scale range 0-255 into k6 gray scale intervals; count the intervals corresponding to all pixel points in the second remote sensing image and the third remote sensing image;

[0126] Step S204, for any one pixel point in the third remote sensing image, record it as a first pixel point, if the gray scale interval of the first pixel point in the third remote sensing image is different from the gray scale interval in the second remote sensing image, mark the first pixel point as a boundary pixel point, and repeat the marking for all pixel points in the third remote sensing image;

[0127] Step S205, for any one grid region in the third remote sensing image, if there is a boundary pixel point, mark the corresponding grid region as a water-land contact region, repeat to obtain all water-land contact regions in the third remote sensing image, and complete to obtain the water-land contact image corresponding to the third remote sensing image; repeat to obtain the water-land contact image corresponding to all images in the first coastal remote sensing image, and complete to obtain the second coastal remote sensing image;

[0128] Step S206, for all boundary pixel points, calculate the connection order of each boundary pixel point under the condition that the connection path is the shortest, and connect all boundary pixel points in turn according to the connection order, to obtain the coastline corresponding to the third collection time and the third remote sensing image, repeat to obtain the coastline corresponding to all images in the first coastal remote sensing image, and complete to obtain the first coastline information.

[0129] Step S3, perform second coastline extraction processing based on the second coastal remote sensing image to obtain second coastline information; step S3 includes the following substeps:

[0130] Step S301, for any one image in the second coastal remote sensing image, record it as a fourth remote sensing image, and for any one water-land contact region in the fourth remote sensing image, record it as a water-land grid region;

[0131] Step S302, set the second neighborhood size as b2*b2; record the area with the waterway grid area as the center and the area with the second neighborhood size as the neighborhood grid area of the waterway grid area; obtain the green band reflectivity GR, the short-wave infrared band reflectivity SW, the red band reflectivity RE and the blue band reflectivity BL of all the pixel points in the neighborhood grid area, and calculate the comprehensive index of all the pixel points in the neighborhood grid area according to the second formula, the second formula is as follows wherein CI is the comprehensive index, q1 and q2 are weight coefficients;

[0132] Step S303, calculate the best comprehensive index threshold YZ by using the OTSU algorithm according to the comprehensive index of all the pixel points in the neighborhood grid area; for any one pixel point in the waterway grid area, record as a water-land pixel point, if the CI of the water-land pixel point is greater than YZ, mark the water-land pixel point as a water-like pixel point, otherwise mark the water-land pixel point as a land pixel point;

[0133] Step S304, repeatedly obtain all the water-like pixel points and land pixel points in all the water-land contact areas in the fourth remote sensing image;

[0134] Step S305, for any one water-like pixel point, if the number of the land-like pixel points in the surrounding 8-neighborhood is greater than k6, mark it as a land pixel point, otherwise mark it as a water body pixel point;

[0135] Step S306, for any one land-like pixel point, if the number of the water-like pixel points in the surrounding 8-neighborhood is greater than k6, mark it as a water body pixel point, otherwise mark it as a land pixel point;

[0136] Step S307, repeatedly obtain all the water body pixel points and land pixel points in all the water-land contact areas in the fourth remote sensing image;

[0137] Step S308, obtain all the connected regions composed of water body pixel points or land pixel points in the fourth remote sensing image by using the connected component labeling algorithm; record as water body regions and land regions in order respectively, obtain the water body pixel points at the boundary of the water body regions and the land regions, record as edge pixel points;

[0138] Step S309, for all the edge pixel points, calculate the connection order of each edge pixel point under the condition that the connection path is the shortest, and connect all the edge pixel points in turn according to the connection order, obtain the coastline corresponding to the fourth remote sensing image, repeatedly obtain the coastline corresponding to all the images in the second coastal remote sensing image, and complete the second coastline information after completion.

[0139] Step S4, perform coastline correction processing based on the first coastline information and the second coastline information to obtain real-time coastline information; step S4 includes the following sub-steps:

[0140] In step S401, the first coastline information and the second coastline information corresponding to the remote sensing image collected at any one collection time are sequentially recorded as the first coastline and the second coastline, respectively.

[0141] In step S402, k7 nodes corresponding to each other are selected from the first coastline and the second coastline, the coordinates of each set of nodes corresponding to each other in the corresponding remote sensing image are obtained, the midpoint coordinates of each set of nodes corresponding to each other are calculated, and all the midpoint coordinates obtained are sequentially connected in the corresponding order to obtain the real-time coastline information at the corresponding collection time.

[0142] Embodiment 3, please refer to Figure 5 as shown, Figure 5 An electronic device is shown in the structure diagram, which can include a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory complete mutual communication through the communication bus. The memory stores computer readable instructions, and the processor can call the instructions in the memory, and when the computer readable instructions are executed by the processor, the steps in the nearshore zone geological survey data management method are run to realize the following functions: collecting remote sensing images of the nearshore zone, and performing remote sensing image preprocessing to obtain first coast remote sensing images; extracting remote sensing images of water-land contact areas based on the first coast remote sensing images to obtain second coast remote sensing images, and performing first coastline extraction processing to obtain first coastline information; performing second coastline extraction processing based on the second coast remote sensing images to obtain second coastline information; and performing coastline correction processing based on the first coastline information and the second coastline information to obtain real-time coastline information.

[0143] In addition, the logical instructions in the memory described above can be implemented in the form of a software function unit and sold or used as an independent product, which can be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the present application essentially or the part that contributes to the prior art or part of the technical solutions can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a plurality of instructions to make a computer device (which can be a personal computer, a server, or a network device, etc.) execute all or part of the steps of the method described in the embodiments of the present application. The foregoing storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk or an optical disk, and various storage program codes.

[0144] In the embodiment 4, the application further provides a computer readable storage medium, and the application provides a storage medium, which stores a computer program. When the computer program is executed by a processor, the steps in the method for managing geological survey data of a near-shore zone are executed to realize the following functions: collecting a remote sensing image of the near-shore zone, and performing remote sensing image preprocessing to obtain a first coastal remote sensing image; extracting a remote sensing image of a water-land contact area based on the first coastal remote sensing image to obtain a second coastal remote sensing image, and performing first coastline extraction processing to obtain first coastline information; performing second coastline extraction processing based on the second coastal remote sensing image to obtain second coastline information; and performing coastline correction processing based on the first coastline information and the second coastline information to obtain real-time coastline information.

[0145] Through the description of the above embodiments, the embodiments of the present application can be provided as a method, a system or a computer program product. Based on such understanding, the above technical solutions can be embodied in the form of a software product, which can be stored in a computer readable storage medium, such as a ROM / RAM, a magnetic disk, an optical disk, etc., and includes a plurality of instructions to make a computer device (which can be a personal computer, a server, or a network device, etc.) execute the method described in each embodiment or some parts of the embodiment.

[0146] In the embodiments provided by the present application, it should be understood that the disclosed system or method can be implemented in other manners. The above described embodiments are merely exemplary, for example, the division of the modules or units is only a logical function division, and there can be another division manner in actual implementation, for example, a plurality of modules or units can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the displayed or discussed mutual couplings or direct couplings or communication connections between different modules can be indirect couplings or communication connections through some interfaces, and there can be electric, mechanical or other forms.

[0147] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, rather than limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: it can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement to some technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A geological survey data management system for near-coastal zones, characterized in that, It includes a remote sensing acquisition module, a first extraction module, a second extraction module, and a coastal monitoring module; The remote sensing acquisition module includes an acquisition unit and a preprocessing unit. The acquisition unit is used to acquire remote sensing images of the near-coastal zone, and the preprocessing unit is used to perform remote sensing image preprocessing to obtain a first coastal remote sensing image. The first extraction module extracts remote sensing images of the land-water contact area based on the first coastal remote sensing image to obtain the second coastal remote sensing image, and performs the first coastline extraction processing to obtain the first coastline information. The second extraction module performs second coastline extraction processing based on the second coastline remote sensing image to obtain second coastline information; The coastal monitoring module performs coastal correction processing based on the first and second coastal information to obtain real-time coastal information. The preprocessing unit is configured with a preprocessing strategy, which includes: Any one of the original remote sensing images is denoted as the first remote sensing image; the size of the first dividing grid is set to a1*a1; the first remote sensing image is divided into multiple grid regions using the first dividing grid, and any one of the grid regions is denoted as the first grid region; Calculate the average brightness and standard deviation of all pixels in the first grid area, and denote them as PE0 and PB0 respectively in order; Set the size of the first neighborhood to b1*b1; denote the area of ​​the first neighborhood size surrounding the first grid region as the similar grid neighborhood of the first grid region, and denote any grid region in the similar grid neighborhood of the first grid region as the neighborhood grid region; Obtain the average brightness and standard deviation of all pixels within the neighboring grid region, and denote them as PE1 and PB1 respectively; then calculate the feature similarity value between the first grid region and the neighboring grid region according to the first formula, as follows: Where TX represents the feature similarity value; Repeatedly obtain the feature similarity values ​​of all grid regions in the similar grid neighborhood of the first grid region and the first grid region, sort them by size, and mark the grid regions corresponding to the smallest k1 feature similarity values ​​as the similar grid regions of the first grid region; Get the total number of pixels in the first grid region and the corresponding similar grid regions, denoted as ZN0, and get the number of pixels appearing at each brightness level, denoted as H[m], where m represents the brightness level; Set the first cutoff ratio to e1 and the second cutoff ratio to e2, where e2 > e1; calculate ZN1 and ZN2, where ZN1 = e1 * ZN0 and ZN2 = e2 * ZN0; Starting from the lowest brightness level, the corresponding number of pixels is accumulated in the direction of increasing brightness level until the accumulated number of pixels is greater than or equal to ZN1, and the highest brightness level accumulated at this time is obtained, which is recorded as the underexposure threshold QB. Starting from the highest brightness level, the number of corresponding pixels is accumulated in the direction of decreasing brightness level until the accumulated number of pixels is greater than or equal to ZN2, and the lowest brightness level accumulated at this time is obtained, which is recorded as the overexposure threshold GB. Arrange the brightness of all pixel points in the first grid area in ascending order, denoted as the first brightness sequence. Calculate the average brightness from the k2-th percentile to the k3-th percentile in the first brightness sequence, denoted as the low brightness average DL, and calculate the average brightness from the k4-th percentile to the k5-th percentile in the first brightness sequence, denoted as the high brightness average GL, where k2, k3, k4, and k5 are set percentage positions, and k2 < k3 < k4 < k5; If DL ≥ QB and GL > GB, mark the first grid area as an overexposed area; if DL < QB and GL < GB, mark the first grid area as an underexposed area; if DL < QB and GL > GB, mark the first grid area as an over- and under-exposed area, otherwise mark the first grid area as a normal area; repeat marking all grid areas in the first remote sensing image; Set Gaussian kernels with corresponding standard deviations of σ1, σ2, and σ3, denoted as the small kernel, the medium kernel, and the large kernel respectively, where σ1 < σ2 < σ3; Perform Gaussian filtering on all underexposed areas based on the small kernel, perform Gaussian filtering on all overexposed areas based on the large kernel, and perform Gaussian filtering on all over- and under-exposed areas and normal areas based on the medium kernel; after completion, obtain the preprocessed image corresponding to the first remote sensing image; repeat the processing of all remote sensing images in the original remote sensing image, and after completion, obtain the first coastal remote sensing image.

2. The geological survey data management system for near-coastal zones according to claim 1, characterized in that, The acquisition unit is configured with an acquisition strategy, and the acquisition strategy includes: Denote the near coastal zone to be monitored as the monitored coastal area. When the tide is rising or ebbing in the monitored coastal area, use a drone to acquire remote sensing images of the monitored coastal area at the first time interval, ensure that the acquisition position, acquisition height, and setting parameters of the acquisition device for each acquisition of remote sensing images of the monitored coastal area are kept consistent, record the acquisition time, and sort the acquired remote sensing images according to the acquisition time, and mark them as the original remote sensing images, where the first time interval is t1.

3. A geological survey data management system for near-coastal zones according to claim 2, characterized in that, The first extraction module is configured with a first extraction strategy, and the first extraction strategy includes: For any image in the first coastal remote sensing image, denote it as the second remote sensing image; denote the acquisition moment corresponding to the second remote sensing image as the second acquisition moment, denote the previous acquisition moment of the second acquisition moment as the third acquisition moment, and denote the preprocessed image corresponding to the third acquisition moment as the third remote sensing image; Set the second divided grid size to a2 * a2; use the second divided grid to divide the second remote sensing image and the third remote sensing image into multiple grid areas; Evenly divide the gray level range of 0 - 255 into k6 gray levels; count the levels corresponding to all pixel points in the second remote sensing image and the third remote sensing image.

4. A geological survey data management system for near-coastal zones according to claim 3, characterized in that, The first extraction strategy further includes: For any pixel point in the third remote sensing image, denote it as the first pixel point. If the gray level of the first pixel point in the third remote sensing image is different from that in the second remote sensing image, mark the first pixel point as a boundary pixel point, and repeat marking all pixel points in the third remote sensing image; For any grid region in the third remote sensing image, if there are boundary pixels, the corresponding grid region is marked as a land-water contact region. All land-water contact regions in the third remote sensing image are repeatedly acquired. After completion, the land-water contact image corresponding to the third remote sensing image is obtained. The land-water contact images corresponding to all images in the first coastal remote sensing image are repeatedly acquired. After completion, the second coastal remote sensing image is obtained. For all boundary pixels, calculate the connection order of each boundary pixel in the case of the shortest connection path, and connect all boundary pixels in sequence according to the connection order to obtain the coastline corresponding to the third acquisition time and the third remote sensing image. Repeat the acquisition of the coastline corresponding to all images in the first coastline remote sensing image, and obtain the first coastline information.

5. A geological survey data management system for near-coastal zones according to claim 4, characterized in that, The second extraction module is configured with a second extraction strategy, which includes: Any image in the second coastal remote sensing image is recorded as the fourth remote sensing image, and any water-land contact area in the fourth remote sensing image is recorded as a water-land grid area. Set the size of the second neighborhood to b2*b2; denote the area centered on the waterway grid region and including the surrounding second neighborhood as the neighboring grid region of the waterway grid region; obtain the green band reflectance GR, short-wave infrared band reflectance SW, red band reflectance RE, and blue band reflectance BL of all pixels in the neighboring grid region, and calculate the comprehensive index of all pixels in the neighboring grid region according to the second formula, as follows: CI is the composite index, and q1 and q2 are the weighting coefficients; Based on the comprehensive index of all pixels in the neighborhood grid region, the optimal comprehensive index threshold is calculated using the OTSU algorithm and denoted as YZ. For any pixel in the waterway grid region, it is denoted as a water-land pixel. If the CI of the water-land pixel is greater than YZ, then the water-land pixel is marked as a water-like pixel; otherwise, it is marked as a land pixel. Repeatedly acquire all water-like pixels and land pixels in all water-land contact areas of the fourth remote sensing image.

6. A geological survey data management system for near-coastal zones according to claim 5, characterized in that, The second extraction strategy also includes: For any water-like pixel, if the number of land-like pixels in its 8-neighborhood is greater than k6, it is marked as a land pixel; otherwise, it is marked as a water pixel. For any land-like pixel, if the number of water-like pixels in its 8-neighborhood is greater than k6, it is marked as a water pixel; otherwise, it is marked as a land pixel. Repeatedly acquire all water body pixels and land pixels in all water-land contact areas of the fourth remote sensing image; The connected component labeling algorithm is used to obtain all connected regions in the fourth remote sensing image composed of water pixels or land pixels respectively; they are recorded as water regions and land regions in order, and the water pixels at the boundary between the water region and the land region are recorded as edge pixels. For all edge pixels, calculate the connection order of each edge pixel in the case of the shortest connection path, and connect all edge pixels in sequence according to the connection order to obtain the coastline corresponding to the fourth remote sensing image. Repeat the acquisition of the coastline corresponding to all images in the second coastline remote sensing image, and obtain the second coastline information after completion.

7. A geological survey data management system for near-coastal zones according to claim 6, characterized in that, The coastal monitoring module is configured with coastal monitoring strategies including: For any remote sensing image acquired at any acquisition time, the information of the first coastline and the information of the second coastline are recorded as the first coastline and the second coastline respectively in sequence. Select k7 corresponding nodes on the first and second coastlines, obtain the coordinates of each group of corresponding nodes in the corresponding remote sensing image, calculate the midpoint coordinates of each group of corresponding nodes, and connect all the obtained midpoint coordinates in the corresponding order to obtain the real-time coastline information at the corresponding acquisition time.

8. A method for managing geological survey data in near-coastal zones, used to implement any one of the geological survey data management systems for near-coastal zones as described in claims 1-7, characterized in that, Includes the following steps: Remote sensing images of the near-coastal zone were acquired and preprocessed to obtain the first coastal remote sensing image. Based on the remote sensing image of the first coastline, remote sensing images of the water-land contact area are extracted to obtain the remote sensing image of the second coastline, and the first coastline is extracted and processed to obtain the information of the first coastline. The second coastline information was obtained by extracting and processing the second coastline based on the remote sensing image of the second coastline. Based on the first and second coastline information, coastline correction processing is performed to obtain real-time coastline information.

Images