CN111862117B - Sea ice block watershed segmentation method based on pixel optimization - Google Patents

Sea ice block watershed segmentation method based on pixel optimization Download PDF

Info

Publication number
CN111862117B
CN111862117B CN202010686569.3A CN202010686569A CN111862117B CN 111862117 B CN111862117 B CN 111862117B CN 202010686569 A CN202010686569 A CN 202010686569A CN 111862117 B CN111862117 B CN 111862117B
Authority
CN
China
Prior art keywords
image
sea
sea ice
ice
gray
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN202010686569.3A
Other languages
Chinese (zh)
Other versions
CN111862117A (en
Inventor
卫志军
王安良
季顺迎
陈晓东
杜祥璞
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Dalian University of Technology
Original Assignee
Dalian University of Technology
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Dalian University of Technology filed Critical Dalian University of Technology
Priority to CN202010686569.3A priority Critical patent/CN111862117B/en
Publication of CN111862117A publication Critical patent/CN111862117A/en
Application granted granted Critical
Publication of CN111862117B publication Critical patent/CN111862117B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/11—Region-based segmentation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/60—Analysis of geometric attributes
    • G06T7/62—Analysis of geometric attributes of area, perimeter, diameter or volume
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10032—Satellite or aerial image; Remote sensing
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/20—Special algorithmic details
    • G06T2207/20112—Image segmentation details
    • G06T2207/20152—Watershed segmentation

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Geometry (AREA)
  • Image Processing (AREA)

Abstract

A watershed algorithm based on pixel optimization can effectively identify sea ice blocks and calculate the size distribution of the sea ice, and belongs to the technical field of image processing and automatic detection. In the processing process, firstly, calibrating the actual size of the sea ice satellite image and converting a gray image by utilizing longitude and latitude; and (4) carrying out boundary segmentation on the sea ice blocks by a watershed method. Carrying out binarization processing on the sea ice and the sea water image after being divided, and distinguishing the sea water and the sea ice through pixel overall characteristics such as gray average value, maximum value, variance and the like; then, carrying out segmentation boundary optimization on the segmentation result by using a corrosion-expansion method, removing invalid segmentation, and overlapping the segmentation result of the copy image optimization with the original image; and finally, calculating the equivalent diameter and the distribution interval thereof, and fitting to obtain a sea ice size probability density distribution function. The invention effectively solves the technical problems that sea ice and sea water in a high-definition satellite image are difficult to distinguish and the size distribution of the sea ice is difficult to count through a pixel optimization watershed sea ice segmentation method.

Description

Sea ice block watershed segmentation method based on pixel optimization
Technical Field
The invention belongs to the technical field of image processing and automatic detection, and relates to a sea ice block watershed segmentation method based on pixel optimization, which can be used for the fields of marine transportation, marine resource exploitation and the like.
Background
Bohai sea and the northern part of the yellow sea are main icing sea areas in China. The Bohai sea is located in a northern hemisphere middle latitude area (37 degrees 07 '-41' 0'N; 117 degrees 35' -121 '10' E), and is embraced among Shandong, hebei, tianjin and Liaoning province I city, and is a semi-closed inner sea. The Bohai sea has a south-north length of 560km, an east-west width of 300km and an area of about 78000km in a sea area 2 Average water depth 18m and maximum water depth 78m. The north of the yellow sea refers to the offshore area of the Liaodong peninsula in the west of the channel of Laozi mountain, where the mouth of the channel is the mouth of the Dongda Duck and the green river.
Sea ice appears in Bohai sea every year in winter, the ice condition difference is large every year, the sea production activity is threatened, and serious disasters are caused when the ice condition is serious. In recent years, with the rapid economic development of the Bohai and the Bohai, a plurality of petroleum platforms and ports are successively established in an icing sea area, the production of marine oil and gas development, shipping, marine fishery and the like is increasingly active, and the influence of sea ice on marine economic activities cannot be ignored. When a sea ice disaster happens, a navigation channel and a port can be blocked, and facilities of the port are damaged; the cutting, collision and clamping of the flowing ice seriously threatens the safety of ship navigation.
The size distribution of the sea ice is an important state parameter which can influence the thermodynamic and kinetic processes of the sea ice, and the size distribution of the sea ice can span from meters to kilometers after the sea ice is broken under the action of external force to form broken ice. Factors that affect the size distribution of sea ice include primarily the thickness, concentration of the sea ice, geographical conditions (such as shoreline), wind, and the intensity of the ocean currents. The response of ice floes to dynamic and thermal effects changes due to changes in shape and size. Therefore, the analysis of the size distribution characteristics of the floating ice has important significance for researching the momentum and heat balance of the Bohai sea area.
Disclosure of Invention
Aiming at the requirements of the problems, the invention provides a watershed algorithm based on pixel optimization, which can effectively identify the broken ice blocks of sea ice and calculate the size distribution.
In order to achieve the purpose, the technical scheme provided by the invention comprises the following steps:
a sea ice block watershed segmentation method based on pixel optimization comprises the following steps:
firstly, a high-definition satellite image with a representative carrier resolution reaching the meter level is added, and the actual sizes of the sea ice and sea water satellite images are calibrated by utilizing the longitude and latitude to obtain the actual area represented by each pixel. And converts the original image into a grayscale image.
And secondly, identifying and segmenting sea ice blocks and junctions between the sea ice and the sea water by using a watershed method.
2.1 Frame selection is carried out on the interested area in the original image and the interested area is copied to obtain a copied image, and then only the copied image is processed;
2.2 Detecting a horizontal edge region and a vertical edge region of the image by a first derivative-based filter Sobel method, and respectively carrying out difference operation on the horizontal direction and the longitudinal direction to obtain gradient values of the image in two directions;
2.3 The low-pass filter Gaussian takes two-dimensional positive-to-positive distribution as a convolution template, and performs equal-weight averaging on the whole image pixel in a form of normal distribution in the probability sense to achieve the effect of image blurring, so that detail information is suppressed, and block surface information with larger scale is highlighted;
2.4 Expand all gray values of the image by a certain multiple (generally 3-5 times) to increase the gray range, because the boundary gradient of the sea ice and the sea ice or the sea ice and the sea water is larger, the boundary gradient value is larger after the integral gray value is increased by a certain multiple. So that rough independent sea ice block boundaries and sea ice and sea water boundaries can be obtained on the image. At the moment, sea ice and sea water are not distinguished, and each gray scale line with large gradient change encloses a closed area in a certain range to serve as a local gray scale area;
2.5 Computing the minimum value of the pixel in the local gray area, wherein the minimum value is defined as the gray value of all the pixels in the direct neighborhood of the connected pixel set with similar gray values, namely the local gray area, which is strictly smaller than the gray value of all the pixels in the direct neighborhood of the set;
2.6 In different gray scale regions, the boundary of the gray scales of two local regions is identified by using a bottom-to-top flooding algorithm with the minimum value of a pixel as a starting point to obtain a more accurate segmentation boundary;
and thirdly, performing binarization processing on the sea ice and sea water images after accurate segmentation. Because the color of the water area is pure, the pixel value change is not as large as that of sea ice, calculation is carried out by counting the average value and the maximum value of the gray level in each block, and variance calculation is considered when necessary, so that the sea water and the sea ice are distinguished;
and fourthly, performing boundary optimization on the segmentation result by adopting a corrosion-expansion method, and overlapping the segmentation result after the copy image is optimized with the original image.
4.1 Processing the binary converted image by erosion-expansion method, scanning the pixels of the binary image one by using matrix operator composed of 0 and 1, and making logic operation on the image and the matrix operator,
and (3) corrosion:
Figure BDA0002587784120000021
and the structural element in the B and the binary image A are used for carrying out AND operation, if the structural element in the B and the binary image A are both 1, the pixel of the point is 1, and otherwise, the pixel is 0.
Expansion:
Figure BDA0002587784120000022
and performing OR operation on the structural elements in the B and the binary image A, wherein if the structural elements in the B and the binary image A are both 0, the pixel of the point is 0, and otherwise, the pixel is 1.
Wherein A is a binary image; b is a matrix operator formed by 0 and 1, and the distribution of the matrix structural elements can have different shapes, such as a circle, a square, a rhombus, a hexagon, a line segment and the like. The operation results of the structural elements with different shapes have certain differences, and the selection should be performed according to the geometric shape of the image to be processed.
4.2 The copy image of the optimized boundary information is superposed to the original image converted into the gray scale image, and obvious ice floes independent of sea ice and boundaries of the sea ice and the sea water are obtained on the original image.
And fifthly, extracting the size of each floating ice pixel, solving the ice block area according to the pixels by the first step of calibration result, calculating the equivalent diameter and the distribution interval thereof, and fitting to obtain the sea ice size probability density distribution function.
5.1 Each pixel calibrated by the known dimension in the first step represents the size of the actual area, and the actual area is obtained according to the pixels of each piece of sea ice. Each piece of sea ice is equivalent to a circle with the same area, and then the diameter of the equivalent circle is as follows:
Figure BDA0002587784120000031
where d is the equivalent diameter and A is the sea ice area converted from the known pixels.
5.2 The equivalent diameter is divided into certain sections from small to large, the number of the equivalent diameters of ice cubes in each section is calculated, and the probability density of the ice cube size in each section is obtained according to the total number of the sea ice cubes after division.
5.3 The sea ice size distribution is researched by adopting an exponential Weibull (Weibull) distribution, and the research is specifically as follows: f (d) =10 a d b And obtaining a specific fitting form by an exponential function after the logarithm change: y = a + bx, where x denotes the equivalent diameter and y denotes the probability density of the ice cube corresponding to the equivalent diameter interval. And performing global optimal fitting according to the probability distribution data of the floating ice size to obtain a and b, and further obtaining a floating ice size distribution function.
The invention has the beneficial effects that: the invention uses the watershed segmentation method of pixel optimization to effectively prevent the ice block region from being cracked and relieve the interference of edge noise; the comparison of the sea ice satellite remote sensing image and the ice cube size identification result can show that the sea ice size information obtained by inversion calculation can accurately represent the size and shape information of the ice cube distribution in the sea area, and the technical problem that the sea ice and the sea water in the high-definition satellite image are difficult to distinguish and count the sea ice size distribution is effectively solved.
Drawings
FIG. 1 is a flow chart of the present invention.
FIG. 2 is a schematic diagram of the sea ice and sea water satellite image segmentation of the present invention. FIG. 2 (a) is an original diagram of sea ice and sea water; FIG. 2 (b) is a schematic diagram of sea ice distribution by watershed method.
Fig. 3 is a partial enlarged view of the lowest point of the satellite image.
FIG. 4 is a diagram of a bottom-up flood watershed segmentation process. FIG. 4 (a) is a three-dimensional map with elevations represented in pixels; FIG. 4 (b) is a schematic view of flooding from the lowest point of a region; FIG. 4 (c) is a schematic view of the gradual formation of the region boundary lines; fig. 4 (d) is a schematic diagram of the final region dividing line.
Fig. 5 shows an image after the image expansion and erosion process.
FIG. 6 is a graph of a Weibull distribution logarithmic transformation fit function.
Detailed Description
Aiming at the requirements of the problems, the invention provides a watershed algorithm based on pixel optimization, which can effectively identify crushed ice cubes and calculate the size distribution.
In order to achieve the purpose, the technical scheme of the invention comprises the following steps (refer to fig. 1):
firstly, adding a high-definition satellite image with a carrier representative resolution up to 2 meters, and calibrating the actual sizes of the sea ice and seawater satellite images by utilizing the longitude and latitude to obtain the actual area represented by each pixel. And converts the original image into a grayscale image.
And secondly, identifying and segmenting sea ice blocks and boundaries between the sea ice and the sea water by utilizing a watershed method (refer to fig. 2).
2.1 Frame selection is carried out on the interested area in the original image and the interested area is copied to obtain a copied image, and then only the copied image is processed;
2.2 Detecting a horizontal edge region and a vertical edge region of the image by a first derivative-based filter Sobel method, and respectively carrying out difference operation on the horizontal direction and the longitudinal direction to obtain gradient values of the image in two directions;
2.3 The low-pass filter Gaussian takes two-dimensional positive-phase distribution as a convolution template, and performs equal-weight averaging on the whole image pixel in a form of normal distribution in the probability sense to achieve the effect of image blurring, so that detail information is suppressed, and block surface information with larger scale is highlighted;
2.4 All gray values of the image are enlarged by 3 times of a large gray range, and the boundary gradient value is larger after the integral gray is increased by a certain multiple because the boundary gradient of the sea ice and the sea ice or the sea ice and the sea water is larger. So that rough independent sea ice block boundaries and sea ice and sea water boundaries can be obtained on the image. At the moment, sea ice and seawater are not distinguished, and each gray scale line with large gradient change encloses a closed area in a certain range to serve as a local gray scale area;
2.5 Calculate the minimum value of the pixels in the local gray area (refer to fig. 3), where the local minimum value is defined as the gray value strictly smaller than the gray values of all pixels in the direct neighborhood of the local gray area in the connected pixel set with similar gray values, i.e., in the local gray area;
2.6 In different gray scale regions, the boundary of the gray scales of two local regions is identified by using a bottom-up flooding algorithm with the minimum value of a pixel as a starting point to obtain a more accurate segmentation boundary (refer to fig. 4);
and thirdly, performing binarization processing on the sea ice and sea water images after accurate segmentation. Because the color of the water area is pure, the pixel value change is not as large as that of sea ice, calculation is carried out by counting the average value and the maximum value of the gray level in each block, and variance calculation is considered when necessary, so that the sea water and the sea ice are distinguished;
and fourthly, performing boundary optimization on the segmentation result by using a corrosion-expansion method (refer to fig. 5), and overlapping the segmentation result after the copy image is optimized with the original image.
4.1 Processing the binary converted image by erosion-expansion method, scanning the pixels of the binary image one by using matrix operator composed of 0 and 1, and making logic operation on the image and the matrix operator,
and (3) corrosion:
Figure BDA0002587784120000041
and the structural element in the B and the binary image A are used for carrying out AND operation, if the structural element in the B and the binary image A are both 1, the pixel of the point is 1, and otherwise, the pixel is 0.
Expansion:
Figure BDA0002587784120000042
and performing OR operation on the structural elements in the B and the binary image A, wherein if the structural elements in the B and the binary image A are both 0, the pixel of the point is 0, and otherwise, the pixel is 1.
Wherein A is a binary image; b is a matrix operator formed by 0 and 1, and the distribution of matrix structural elements can have different shapes, such as a circle, a square, a rhombus, a hexagon, a line segment and the like. The operation results of the structural elements with different shapes have certain differences, and the selection should be performed according to the geometric shape of the image to be processed.
4.2 The copy image of the optimized boundary information is superposed to the original image converted into the gray scale image, and obvious ice floes independent of sea ice and boundaries of the sea ice and the sea water are obtained on the original image.
And fifthly, extracting the size of each floating ice pixel, solving the ice block area according to the pixels by the first step of calibration result, calculating the equivalent diameter and the distribution interval thereof, and fitting to obtain the sea ice size probability density distribution function.
5.1 Each pixel calibrated by the known dimension in the first step represents the size of the actual area, and the actual area is obtained according to the pixels of each piece of sea ice. Each piece of sea ice is equivalent to a circle with the same area, and the diameter of the equivalent circle is as follows:
Figure BDA0002587784120000051
wherein d is the equivalent diameter, and A is the sea ice area obtained by conversion of known pixels.
5.2 The equivalent diameter is divided into certain sections from small to large, the number of the equivalent diameters of ice cubes in each section is calculated, and the probability density of the ice cube size in each section is obtained according to the total number of the sea ice cubes after division.
5.3 The sea ice size distribution is researched by adopting an exponential Weibull (Weibull) distribution, and the research is specifically as follows: f (d) =10 a d b . Obtaining a specific fitting form by an exponential function after logarithmic change: y = a + bx, where x represents the equivalent diameter and y represents the probability density of the ice cube corresponding to the equivalent diameter interval. Global best fit was performed based on probability distribution data of ice floe size to obtain a =7.29 and b = -3.53 (see fig. 6).
Finally determining a sea ice size distribution function: f (d) =10 7.29 d -3.53 Wherein d is greater than or equal to d min ＝31.4。
The above-mentioned embodiments only express the embodiments of the present invention, but not should be understood as the limitation of the scope of the invention patent, it should be noted that those skilled in the art can make several variations and modifications without departing from the concept of the present invention, and these all fall into the protection scope of the present invention.

Claims (1)

1. A sea ice block watershed segmentation method based on pixel optimization is characterized by comprising the following steps:
firstly, loading a high-definition satellite image, and calibrating the actual sizes of the sea ice and the sea satellite image by utilizing longitude and latitude to obtain the actual area represented by each pixel; converting the original image into a gray image;
secondly, identifying and segmenting sea ice blocks and junctions between the sea ice and the sea water by a watershed method;
2.1 Frame selection is carried out on the interested area in the original image and the interested area is copied to obtain a copied image, and then only the copied image is processed;
2.2 Detecting a horizontal edge region and a vertical edge region of the image by a first derivative-based filter Sobel method, and respectively carrying out difference operation on the horizontal direction and the longitudinal direction to obtain gradient values of the image in two directions;
2.3 The low-pass filter takes two-dimensional positive-space distribution as a convolution template, and performs equal-weight averaging on the pixels of the whole image in a normal distribution mode to achieve the effect of image blurring, so that detail information is suppressed, and block surface information with larger scale is highlighted;
2.4 All gray values of the image are expanded, and the gray range is enlarged, so that rough independent sea ice block boundaries and sea ice and sea water boundaries can be obtained on the image; at the moment, sea ice and sea water are not distinguished, and each gray line with large gradient change encloses a closed area in a certain range to serve as a local gray area;
2.5 Computing the minimum value of the pixels in the local gray area, wherein the local minimum value is defined as the gray value of a connected pixel set with similar gray levels, namely the local gray area, which is strictly smaller than the gray values of all pixels in the direct neighborhood of the set;
2.6 In different gray scale regions, the boundary of the gray scales of two local regions is identified by using a bottom-to-top flooding algorithm with the minimum value of a pixel as a starting point to obtain a more accurate segmentation boundary;
thirdly, performing binarization processing on the sea ice and sea water images after accurate segmentation; because the color of the water area is pure, the pixel value change is smaller than that of the sea ice, and the sea water and the sea ice are distinguished by calculating the average value and the maximum value of the gray level in each block;
fourthly, performing boundary optimization on the segmentation result by adopting a corrosion-expansion method, and overlapping the segmentation result after the optimization of the copied image with the original image;
4.1 Processing the image after the binarization conversion by using a corrosion-expansion method, scanning pixels of the binarized image one by using a matrix operator consisting of 0 and 1, and performing logic operation on the image and the matrix operator;
4.2 The copy image of the optimized boundary information is superposed to the original image converted into the gray scale image, and obvious ice floations independent of the sea ice and boundaries of the sea ice and the sea water are obtained on the original image;
fifthly, extracting the size of each floating ice pixel, solving the ice block area according to the pixels by the first step of calibration result, calculating the equivalent diameter and the distribution interval thereof, and obtaining the sea ice size probability density distribution function through fitting;
5.1 Each pixel calibrated by the known dimension in the first step represents the size of the actual area, and the actual area is calculated according to the pixel of each piece of sea ice; each piece of sea ice is equivalent to a circle with the same area, and then the diameter of the equivalent circle is as follows:
Figure FDA0002587784110000011
wherein d is the equivalent diameter, and A is the sea ice area obtained by conversion of known pixels;
5.2 Dividing the equivalent diameter into certain intervals from small to large, calculating the number of the equivalent diameters of ice cubes in each interval, and calculating the probability density of the size of the ice cubes in each interval according to the number of the total sea ice cubes after division;
5.3 Employing exponential Weibull distribution to research sea ice size distribution, wherein the specific research form is as follows: f (d) =10 a d b And obtaining a specific fitting form by an exponential function after the logarithm change: y = a + bx, where x represents the equivalent diameter and y represents the probability density of the ice cube corresponding to the equivalent diameter interval; and performing global optimal fitting according to the probability distribution data of the floating ice size to obtain a and b, and further obtaining a floating ice size distribution function.
CN202010686569.3A 2020-07-16 2020-07-16 Sea ice block watershed segmentation method based on pixel optimization Active CN111862117B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010686569.3A CN111862117B (en) 2020-07-16 2020-07-16 Sea ice block watershed segmentation method based on pixel optimization

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010686569.3A CN111862117B (en) 2020-07-16 2020-07-16 Sea ice block watershed segmentation method based on pixel optimization

Publications (2)

Publication Number Publication Date
CN111862117A CN111862117A (en) 2020-10-30
CN111862117B true CN111862117B (en) 2022-10-18

Family

ID=72983666

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010686569.3A Active CN111862117B (en) 2020-07-16 2020-07-16 Sea ice block watershed segmentation method based on pixel optimization

Country Status (1)

Country Link
CN (1) CN111862117B (en)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115358497B (en) * 2022-10-24 2023-03-10 湖南长理尚洋科技有限公司 GIS technology-based intelligent panoramic river patrol method and system
CN117890068B (en) * 2024-03-14 2024-05-24 中国空气动力研究与发展中心低速空气动力研究所 Icing cloud particle size reconstruction method and computer readable storage medium

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102708369B (en) * 2012-05-14 2014-06-25 大连理工大学 Sea ice parameter extraction method on basis of satellite image
CN103500451B (en) * 2013-10-10 2016-02-17 中国科学院上海技术物理研究所 A kind of independent floating ice extracting method for satellite data
CN106296675A (en) * 2016-08-04 2017-01-04 山东科技大学 A kind of dividing method of the uneven image of strong noise gray scale
CN111353371A (en) * 2019-11-22 2020-06-30 中国电子科技集团公司电子科学研究院 Coastline extraction method based on satellite-borne SAR image

Also Published As

Publication number Publication date
CN111862117A (en) 2020-10-30

Similar Documents

Publication Publication Date Title
Cao et al. Mapping changes in coastlines and tidal flats in developing islands using the full time series of Landsat images
Chen et al. Temporal and spatial variation of coastline using remote sensing images for Zhoushan archipelago, China
Fan et al. Shoreline dynamics of the active Yellow River delta since the implementation of Water-Sediment Regulation Scheme: A remote-sensing and statistics-based approach
Burvingt et al. Classification of beach response to extreme storms
Li et al. Coastline change detection with satellite remote sensing for environmental management of the Pearl River Estuary, China
Cham et al. An analysis of shoreline changes using combined multitemporal remote sensing and digital evaluation model
CN110956101B (en) Remote sensing image yellow river ice detection method based on random forest algorithm
Diez et al. Vulnerability to sea-level rise on the coast of the Buenos Aires Province
Sunder et al. Inter-comparison of remote sensing sensing-based shoreline mapping techniques at different coastal stretches of India
Liu et al. Detecting coastline change from satellite images based on beach slope estimation in a tidal flat
Hu et al. Monitoring coastline variations in the Pearl River Estuary from 1978 to 2018 by integrating Canny edge detection and Otsu methods using long time series Landsat dataset
CN111862117B (en) Sea ice block watershed segmentation method based on pixel optimization
Hayakawa et al. GIS analysis of fluvial knickzone distribution in Japanese mountain watersheds
Xu et al. Analysis of the characteristics and causes of coastline variation in the Bohai Rim (1980–2010)
Behling et al. Spatiotemporal shoreline dynamics of Namibian coastal lagoons derived by a dense remote sensing time series approach
CN110020607B (en) Method for searching similar watershed based on space dimension division theory
CN112509134A (en) Tidal flat digital elevation model construction method and system
Papanicolaou et al. The use of fractals to quantify the morphology of cluster microforms
Gong et al. Parameterizing the Yellow River Delta tidal creek morphology using automated extraction from remote sensing images
Eom et al. Tidal channel distribution in relation to surface sedimentary facies based on remotely sensed data
Kang et al. Topographic evolution of tidal flats based on remote sensing: an example in Jiangsu coast, Southern Yellow Sea
Toorani et al. Monitoring shoreline shift under rapid sea-level change on the Caspian Sea observed over 60 years of satellite and aerial photo records
Wu et al. Coastal reclamation shaped narrower and steeper tidal flats in Fujian, China: Evidence from time-series satellite data
CN113283437B (en) Submarine sand wave feature identification method
Ping et al. Application of a sea surface temperature front composite algorithm in the Bohai, Yellow, and East China Seas

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant