CN111767813A - Multi-feature and long-time-sequence probability synergistic mangrove forest extraction method - Google Patents

Multi-feature and long-time-sequence probability synergistic mangrove forest extraction method Download PDF

Info

Publication number
CN111767813A
CN111767813A CN202010565391.7A CN202010565391A CN111767813A CN 111767813 A CN111767813 A CN 111767813A CN 202010565391 A CN202010565391 A CN 202010565391A CN 111767813 A CN111767813 A CN 111767813A
Authority
CN
China
Prior art keywords
mangrove forest
mangrove
probability
index
data
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.)
Granted
Application number
CN202010565391.7A
Other languages
Chinese (zh)
Other versions
CN111767813B (en
Inventor
杨刚
黄可
孙伟伟
孟祥珍
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.)
Ningbo University
Original Assignee
Ningbo University
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 Ningbo University filed Critical Ningbo University
Priority to CN202010565391.7A priority Critical patent/CN111767813B/en
Publication of CN111767813A publication Critical patent/CN111767813A/en
Application granted granted Critical
Publication of CN111767813B publication Critical patent/CN111767813B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/10—Terrestrial scenes
    • G06V20/188—Vegetation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/23—Clustering techniques
    • G06F18/232—Non-hierarchical techniques
    • G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
    • G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/20—Image preprocessing
    • G06V10/30—Noise filtering
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/40—Extraction of image or video features
    • G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • General Physics & Mathematics (AREA)
  • Multimedia (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Evolutionary Computation (AREA)
  • Evolutionary Biology (AREA)
  • General Engineering & Computer Science (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Artificial Intelligence (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Probability & Statistics with Applications (AREA)
  • Health & Medical Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Image Processing (AREA)

Abstract

The invention relates to a mangrove forest extraction method with multi-feature and long time sequence probability synergy, which comprises the following steps: screening all Landsat image surface reflectivity data according to time and regions by relying on a GEE platform; and calculating the normalized vegetation index, the normalized water index and the mangrove forest comprehensive identification index. The invention has the beneficial effects that: the method fully excavates and utilizes Landsat long time sequence data, is convenient to acquire data and easy to realize the process, can provide mangrove forest data with long time sequence and full space coverage, is favorable for analyzing the time-space variation trend of mangroves in China, and provides scientific support and decision reference suggestions for implementing large-scale mangrove forest ecosystem protection and recovery actions in China. The method provided by the invention fully excavates and utilizes Landsat long time sequence data, fuses spatial information and time information, effectively supplements the existing mangrove forest extraction method, is convenient for data acquisition and easy in process realization, and is beneficial to fine monitoring of the mangrove forest and ecological system protection of the mangrove forest.

Description

Multi-feature and long-time-sequence probability synergistic mangrove forest extraction method
Technical Field
The invention relates to the technical field of classification extraction of remote sensing images, in particular to a mangrove forest extraction method with cooperation of multiple features and long time sequence probability.
Background
The mangrove forest is a woody plant community, mainly grows on intertidal zone beach of tropical and subtropical coast, and has unique ecological function and great social and economic value. Under the influence of human activities, the mangrove forest area worldwide is decreasing at a rate of 1% per year, and the mangrove forest in China is also subjected to multiple damages, resulting in dramatic area reduction.
The mangrove remote sensing monitoring technology is applied to the fields of region extraction, interspecies classification, community structure, biomass, disaster situations, dynamic change, driving mechanism, mangrove wetland protection and management and the like, most of researches aim at typical regions such as natural protected areas, and the national and global scales are less. The extraction of the range of the mangrove forest is usually based on a single vegetation index (including a normalized vegetation index, a ratio vegetation index, a greenness vegetation index and the like), spectral characteristics and textural characteristics, the classification methods commonly used for extracting the mangrove forest comprise visual interpretation, an object-oriented classification method, a decision tree method, a support vector machine method and the like, and the information extraction is performed by combining the classification methods commonly used. Most researches have differences in data sources, time phases, classification methods, interpretation scales and the like, so that large differences exist among different research results, and uncertainty is caused for analyzing the spatiotemporal change condition of mangroves in a certain region.
At present, mangrove remote sensing monitoring is limited to a small space range of a certain province or a certain area, and monitoring is rarely carried out from the national scale. Meanwhile, only single-time phase data or a plurality of time phase data are used for monitoring the mangrove forest, and the advantage of long-time remote sensing data is ignored. Although the development of the satellite remote sensing technology has accumulated a large amount of remote sensing data, due to the reasons of large calculation amount, difficult acquisition and the like, the remote sensing monitoring in a long time sequence and a large range is difficult to realize, and the data potential cannot be well played. The GEE platform provides complete Landsat, MODIS and Sentinel series images, can perform online batch processing on the images, and can well meet the requirement of long-time sequence monitoring of mangroves in large-scale areas.
Disclosure of Invention
The invention aims to overcome the defects in the prior art and provide a mangrove forest extraction method with multi-feature and long time sequence probability synergy.
The mangrove forest extracting method with the cooperation of multiple characteristics and long time sequence probability comprises the following steps:
step 1, screening all Landsat image surface reflectivity data according to time and regions by relying on a GEE platform;
step 2, calculating a normalized vegetation index, a normalized water index and a mangrove forest comprehensive identification index based on the selected Landsat image surface reflectivity data:
Figure BDA0002547646790000021
Figure BDA0002547646790000022
CMRI＝NDVI-NDWI (3)
in the above formula (1) to the above formula (3), NDVI is a normalized vegetation index, NDWI is a normalized water index, CMRI is a mangrove forest comprehensive identification index, NIR is a near infrared band reflectivity, R is a red light band reflectivity, and G is a green light band reflectivity;
step 3, carrying out unsupervised classification on the Landsat image by adopting a K-means algorithm, and selecting NDVI, NDWI, NIR, SWIR1 and SWIR2 to participate in unsupervised classification to distinguish a water body from a land;
step 4, according to unsupervised classification results of the land and the water body, extracting land and water borderlines through a Canny edge detection algorithm, generating a buffer zone by taking a certain distance from the land and water borderlines as a limit, and then overlapping land masks generated by unsupervised classification to obtain a possible mangrove forest growing zone;
step 4.1, smoothing the image by Gaussian filtering to remove noise; the Gaussian filter two-dimensional kernel function calculation formula is as follows:
Figure BDA0002547646790000023
in the formula, sigma is set according to the size of the generated template, x represents the difference value between the horizontal coordinate of the current element and the horizontal coordinate of the center of the template, and y represents the difference value between the vertical coordinate of the current element and the vertical coordinate of the center of the template;
step 4.2, calculating gradient values and gradient directions:
Figure BDA0002547646790000024
Figure BDA0002547646790000025
wherein: g (m, n) is the integrated gradient value of the pixel point of the position (m, n), theta is the gradient direction, Gx(m, n) and gy(m, n) are x-direction gradient values and y-direction gradient values obtained by multiplying a sobel or other operators by pixel points at the positions (m, n) respectively;
step 4.3, non-maximum suppression: filtering out points which are not edges, and setting the gray value as 0; if a pixel belongs to the edge, the gradient value of the pixel in the gradient direction is maximum; if a pixel point does not belong to the edge, the gradient value of the pixel point in the gradient direction is minimum;
step 4.4, double-threshold screening: setting two thresholds as maxVal and minVal respectively, wherein pixel points larger than the threshold maxVal are detected as edges, pixel points lower than the threshold minVal are detected as non-edges, and for pixel points positioned between the threshold maxVal and the threshold minVal, if the pixel points are adjacent to the pixel points determined as the edges, the pixel points are judged as the edges, otherwise, the pixel points are non-edges;
step 5, extracting the mangrove forest by adopting a multi-feature decision tree classification method based on index and spectral information;
and 6, calculating the mangrove forest growth probability based on the mangrove forest length time sequence data extracted in the step 5, and synthesizing annual data according to a probability threshold value to determine the mangrove forest growth probability.
Preferably, the method for extracting the mangrove forest by adopting the multi-feature decision tree classification method based on the index and the spectral information in the step 5 comprises the following steps: selecting mangrove sample data based on-site sampling data in combination with an existing mangrove reference data set, counting distribution histograms of original single bands and spectral indexes of the Landsat images in the sample area, acquiring the maximum value and the minimum value of the mangrove in each band and index, setting the maximum value and the minimum value as the upper threshold and the lower threshold of mangrove extraction, constructing a decision tree based on the combination of the index characteristics and the original spectral bands, and taking the Landsat images of which the band values and the indexes are simultaneously positioned between the upper threshold and the lower threshold as the mangrove:
Figure BDA0002547646790000031
in the above formula, the first and second carbon atoms are,
Figure BDA0002547646790000032
the intersection symbols represent that all the wave bands and the indexes simultaneously meet the conditions; b isi(m, n) is the value of the ith band or index at position (m, n); mini、maxiLower and upper thresholds for the ith band or index, respectively.
Preferably, the annual data synthesis is performed according to the probability threshold in the step 6, and the method for determining the mangrove forest growth probability comprises the following steps:
firstly, according to the decision tree classification result, the probability that the position of each pixel is a mangrove forest in the year is counted:
Figure BDA0002547646790000033
in the above formula, P (m, N) is the mangrove forest probability of the position (m, N) in the year, N' (m, N) is the number of times the position (m, N) is identified as the mangrove forest, N (m, N) is the total number of classification times of the position (m, N), and P is the mangrove forest growth probability threshold;
obtaining annual probability data of mangroves, and then carrying out sensitivity analysis on the extracted results to different probability values; and determining the mangrove forest growth probability threshold value P through an experiment with the minimum comparison difference between the extracted mangrove forest area result and the existing data set and the highest precision evaluation of the on-site sampling points.
Preferably, when the sensitivities with different probability values are analyzed in the step 6, the value range of the probability value is 0-1, and the step length is 0.1.
The invention has the beneficial effects that: the method fully excavates and utilizes Landsat long time sequence data, is convenient to acquire data and easy to realize the process, can provide mangrove forest data with long time sequence and full space coverage, is favorable for analyzing the time-space variation trend of mangroves in China, and provides scientific support and decision reference suggestions for implementing large-scale mangrove forest ecosystem protection and recovery actions in China. The method provided by the invention fully excavates and utilizes Landsat long time sequence data, fuses spatial information and time information, effectively supplements the existing mangrove forest extraction method, is convenient for data acquisition and easy in process realization, and is beneficial to fine monitoring of the mangrove forest and ecological system protection of the mangrove forest.
Drawings
FIG. 1 is a flow chart of an embodiment of a mangrove forest extraction method;
FIG. 2 is a diagram of an embodiment of a true color composite image and an unsupervised classification result;
FIG. 3 is a view of a buffer area 2300m in the embodiment;
FIG. 4 is a diagram of a possible growing area of mangrove forest in the example;
FIG. 5 is a chart of annual mangrove forest probability in the example (taking New Yingwan of Hainan province as an example);
FIG. 6 is a diagram of the result of mangrove forest extraction under different probabilities in the example.
FIG. 7 is a graph showing the results of the example (the distribution of the Chinese mangrove forest in 2015).
Detailed Description
The present invention will be further described with reference to the following examples. The following examples are set forth merely to aid in the understanding of the invention. It should be noted that, for a person skilled in the art, several modifications can be made to the invention without departing from the principle of the invention, and these modifications and modifications also fall within the protection scope of the claims of the present invention.
The invention provides a mangrove forest extraction method with multi-feature and long time sequence probability synergy, aiming at the problems of remote sensing monitoring of the mangrove forest at present. Firstly, all available Landsat images in one year are processed in batches through a GEE platform, a multi-feature decision tree classification method based on index and spectral information is provided to extract a rough mangrove forest growing range, finally, the mangrove forest probability is calculated based on long-term data and annual data synthesis is carried out according to the mangrove forest growing range, mangrove forest annual probability data are generated, and the mangrove forest range is finely extracted by determining the mangrove forest growing probability.
As an example, as shown in fig. 1, the mangrove forest extraction method based on the GEE platform includes:
step 1, screening all available Landsat surface reflectivity data according to time and regions, wherein the specific operation is the prior art;
step 2, calculating three spectral indexes of NDVI, NDWI and CMRI based on the selected Landsat surface reflectivity data, wherein the calculation formula is the prior art and is as follows:
Figure BDA0002547646790000041
Figure BDA0002547646790000051
CMRI＝NDVI-NDWI (3)
wherein: b3, B4 and B5 are respectively the reflectivities of Landsat-8 images in the B3, B4 and B5 wave bands. The purpose of this step is to obtain the time series data set of each index feature required for extracting mangrove forest.
And 3, carrying out unsupervised classification on the images by adopting a K-means algorithm, selecting two spectral indexes of NDVI and NDWI and three infrared wave bands of NIR, SWIR1 and SWIR2 to participate in unsupervised classification, wherein the classification result is water and land, and the specific operation is the prior art. The purpose of this step is to distinguish between water and land to facilitate the extraction of the land and water lines in step 4. The true color composite image and the unsupervised classification result are shown in fig. 2.
Step 4, according to unsupervised classification results of land and water bodies, extracting land and water edges by a Canny edge detection algorithm, generating a buffer zone (as shown in fig. 3) by taking 2300m away from the land and water edges as a limit, and then overlapping land masks generated by unsupervised classification in the step 3 to obtain a possible mangrove forest growing zone as shown in fig. 4, wherein the specific operation is the prior art, and the sigma value of Gaussian filtering in the first step of the Canny edge detection algorithm is set to be 1.4;
step 5, extracting the mangrove forest by adopting a multi-feature decision tree classification method based on index and spectral information, and specifically operating as follows: selecting mangrove sample data based on-site sampling data and an existing mangrove reference data set, counting distribution histograms of original single bands and spectral indexes of Landsat images in a sample area, obtaining the maximum value and the minimum value of the mangrove in each band and index through multiple experiments, setting the maximum value and the minimum value as threshold upper limits and lower limits of mangrove extraction, constructing a decision tree based on the combination of index features and original spectral bands, and determining that the mangrove is obtained when each band value and each index value simultaneously meet the upper and lower limit ranges of the threshold, wherein the formula is (7):
Figure BDA0002547646790000052
in the above formula, the first and second carbon atoms are,
Figure BDA0002547646790000053
the intersection symbols represent that all the wave bands and indexes simultaneously meet the conditions; b isi(m, n) is the value of the ith band or index at position (m, n); mini、maxiRespectively, of the i-th band or indexLower and upper threshold limits of (d);
the spectral and exponential characteristics and thresholds are shown in table 1.
TABLE 1 mangrove forest classification threshold
Figure BDA0002547646790000061
And 6, calculating the mangrove forest growth probability based on the mangrove forest length time sequence data extracted in the step 5, synthesizing annual data according to a probability threshold, determining the mangrove forest growth probability threshold, and realizing the fine extraction of the mangrove forest range. The specific operation is as follows: according to the decision tree classification result, the probability that the position of each pixel is the mangrove forest in the year is counted to obtain a probability map of the annual mangrove forest as shown in fig. 5 (taking new gulf of Hainan province as an example), and the calculation formula is (8) to obtain the annual probability data of the mangrove forest. And then, carrying out sensitivity analysis of the extracted result on different probability values (0.1-1, step length of 0.1), namely determining a probability threshold value P to be 0.5 by an experiment with the minimum comparison difference between the extracted mangrove forest area result and the existing data set and the highest precision evaluation of the field sampling point, and realizing fine extraction of the mangrove forest range. The mangrove forest extraction results under different probabilities are shown in fig. 6.
Figure BDA0002547646790000062
Wherein: p (m, N) is the mangrove forest probability for the location (m, N) in the year, N' (m, N) is the number of times the location is classified as a mangrove forest, N (m, N) is the total number of classifications for the location, and P is the mangrove forest growth probability threshold. The mangrove forest map shown in fig. 7 (the chinese mangrove forest map in 2015) was finally obtained.

Claims (4)

1. A mangrove forest extraction method with multi-feature and long time sequence probability synergy is characterized by comprising the following steps:
step 1, screening all Landsat image surface reflectivity data according to time and regions by relying on a GEE platform;
step 2, calculating a normalized vegetation index, a normalized water index and a mangrove forest comprehensive identification index based on the selected Landsat image surface reflectivity data:
Figure FDA0002547646780000011
Figure FDA0002547646780000012
CMRI＝NDVI-NDWI (3)
in the above formula (1) to the above formula (3), NDVI is a normalized vegetation index, NDWI is a normalized water index, CMRI is a mangrove forest comprehensive identification index, NIR is a near infrared band reflectivity, R is a red light band reflectivity, and G is a green light band reflectivity;
step 3, carrying out unsupervised classification on the Landsat image by adopting a K-means algorithm, and selecting NDVI, NDWI, NIR, SWIR1 and SWIR2 to participate in unsupervised classification to distinguish a water body from a land;
step 4, according to unsupervised classification results of the land and the water body, extracting land and water borderlines through a Canny edge detection algorithm, generating a buffer zone by taking a certain distance from the land and water borderlines as a limit, and then overlapping land masks generated by unsupervised classification to obtain a possible mangrove forest growing zone;
step 4.1, smoothing the image by Gaussian filtering to remove noise; the Gaussian filter two-dimensional kernel function calculation formula is as follows:
Figure FDA0002547646780000013
in the formula, sigma is set according to the size of the generated template, x represents the difference value between the horizontal coordinate of the current element and the horizontal coordinate of the center of the template, and y represents the difference value between the vertical coordinate of the current element and the vertical coordinate of the center of the template;
step 4.2, calculating gradient values and gradient directions:
Figure FDA0002547646780000014
Figure FDA0002547646780000015
wherein: g (m, n) is the integrated gradient value of the pixel point of the position (m, n), theta is the gradient direction, Gx(m, n) and gy(m, n) are x-direction gradient values and y-direction gradient values obtained by multiplying a sobel or other operators by pixel points at the positions (m, n) respectively;
step 4.3, non-maximum suppression: filtering out points which are not edges, and setting the gray value as 0; if a pixel belongs to the edge, the gradient value of the pixel in the gradient direction is maximum; if a pixel point does not belong to the edge, the gradient value of the pixel point in the gradient direction is minimum;
step 4.4, double-threshold screening: setting two thresholds as maxVal and minVal respectively, wherein pixel points larger than the threshold maxVal are detected as edges, pixel points lower than the threshold minVal are detected as non-edges, and for pixel points positioned between the threshold maxVal and the threshold minVal, if the pixel points are adjacent to the pixel points determined as the edges, the pixel points are judged as the edges, otherwise, the pixel points are non-edges;
step 5, extracting the mangrove forest by adopting a multi-feature decision tree classification method based on index and spectral information;
and 6, calculating the mangrove forest growth probability based on the mangrove forest length time sequence data extracted in the step 5, and synthesizing annual data according to a probability threshold value to determine the mangrove forest growth probability.
2. The method for extracting mangrove forest with multi-feature and long time-series probability coordination according to claim 1, characterized in that: the method for extracting the mangrove forest by adopting the multi-feature decision tree classification method based on the index and the spectral information in the step 5 comprises the following steps: selecting mangrove sample data based on-site sampling data in combination with an existing mangrove reference data set, counting distribution histograms of original single bands and spectral indexes of the Landsat images in the sample area, acquiring the maximum value and the minimum value of the mangrove in each band and index, setting the maximum value and the minimum value as the upper threshold and the lower threshold of mangrove extraction, constructing a decision tree based on the combination of the index characteristics and the original spectral bands, and taking the Landsat images of which the band values and the indexes are simultaneously positioned between the upper threshold and the lower threshold as the mangrove:
Figure FDA0002547646780000021
in the above formula, the first and second carbon atoms are,
Figure FDA0002547646780000022
the intersection symbols represent that all the wave bands and the indexes simultaneously meet the conditions; b isi(m, n) is the value of the ith band or index at position (m, n); mini、maxiLower and upper thresholds for the ith band or index, respectively.
3. The method for extracting mangrove forest with multi-feature and long time-series probability coordination according to claim 1, characterized in that: in the step 6, annual data synthesis is carried out according to a probability threshold, and the method for determining the mangrove forest growth probability comprises the following steps:
firstly, according to the decision tree classification result, the probability that the position of each pixel is a mangrove forest in the year is counted:
Figure FDA0002547646780000023
in the above formula, P (m, N) is the mangrove forest probability of the position (m, N) in the year, N' (m, N) is the number of times the position (m, N) is identified as the mangrove forest, N (m, N) is the total number of classification times of the position (m, N), and P is the mangrove forest growth probability threshold;
obtaining annual probability data of mangroves, and then carrying out sensitivity analysis on the extracted results to different probability values; and determining the mangrove forest growth probability threshold value P through an experiment with the minimum comparison difference between the extracted mangrove forest area result and the existing data set and the highest precision evaluation of the on-site sampling points.
4. The method for extracting mangrove forest cooperated with multi-feature and long time-series probability as claimed in claim 3, wherein: and 6, when the sensitivities of different probability values are analyzed, the value range of the probability value is 0-1, and the step length is 0.1.
CN202010565391.7A 2020-06-19 2020-06-19 Multi-feature and long-time-sequence probability synergistic mangrove forest extraction method Active CN111767813B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010565391.7A CN111767813B (en) 2020-06-19 2020-06-19 Multi-feature and long-time-sequence probability synergistic mangrove forest extraction method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010565391.7A CN111767813B (en) 2020-06-19 2020-06-19 Multi-feature and long-time-sequence probability synergistic mangrove forest extraction method

Publications (2)

Publication Number Publication Date
CN111767813A true CN111767813A (en) 2020-10-13
CN111767813B CN111767813B (en) 2021-09-28

Family

ID=72721074

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010565391.7A Active CN111767813B (en) 2020-06-19 2020-06-19 Multi-feature and long-time-sequence probability synergistic mangrove forest extraction method

Country Status (1)

Country Link
CN (1) CN111767813B (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112329829A (en) * 2020-10-27 2021-02-05 宁波大学 Hyperspectral data-based mangrove forest extraction method
CN115620133A (en) * 2022-10-08 2023-01-17 宁波大学 Mangrove extraction method considering phenology and water level time sequence characteristics

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8181386B1 (en) * 2008-03-13 2012-05-22 Riley Jr Robert W Encasement devices and methods for planting mangroves
CN103000077A (en) * 2012-11-27 2013-03-27 中国科学院东北地理与农业生态研究所 Method for carrying out mangrove forest map making on intermediate resolution remote sensing image by utilizing object-oriented classification method
CN108986116A (en) * 2018-07-16 2018-12-11 中国科学院东北地理与农业生态研究所 A kind of mangrove extracting method and system based on remote sensing image
CN110852381A (en) * 2019-11-11 2020-02-28 四川航天神坤科技有限公司 Forest fire burned area extraction method and system
WO2020063461A1 (en) * 2018-09-30 2020-04-02 广州地理研究所 Urban extent extraction method and apparatus based on random forest classification algorithm, and electronic device

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8181386B1 (en) * 2008-03-13 2012-05-22 Riley Jr Robert W Encasement devices and methods for planting mangroves
CN103000077A (en) * 2012-11-27 2013-03-27 中国科学院东北地理与农业生态研究所 Method for carrying out mangrove forest map making on intermediate resolution remote sensing image by utilizing object-oriented classification method
CN108986116A (en) * 2018-07-16 2018-12-11 中国科学院东北地理与农业生态研究所 A kind of mangrove extracting method and system based on remote sensing image
WO2020063461A1 (en) * 2018-09-30 2020-04-02 广州地理研究所 Urban extent extraction method and apparatus based on random forest classification algorithm, and electronic device
CN110852381A (en) * 2019-11-11 2020-02-28 四川航天神坤科技有限公司 Forest fire burned area extraction method and system

Non-Patent Citations (4)

* Cited by examiner, † Cited by third party
Title
KAUSHIK GUPTA等: "An index for discrimination of mangroves from non-mangroves using LANDSAT 8 OLI imagery", 《METHODSX》 *
刘 凯等: "基于Google Earth Engine的红树林年际变化监测研究", 《地球信息科学》 *
李澜等: "《光学设计实用技术》", 31 January 2019 *
赵谦等: "《智能视频图像处理技术与应用》", 30 September 2016 *

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112329829A (en) * 2020-10-27 2021-02-05 宁波大学 Hyperspectral data-based mangrove forest extraction method
CN112329829B (en) * 2020-10-27 2021-08-27 宁波大学 Hyperspectral data-based mangrove forest extraction method
CN115620133A (en) * 2022-10-08 2023-01-17 宁波大学 Mangrove extraction method considering phenology and water level time sequence characteristics

Also Published As

Publication number Publication date
CN111767813B (en) 2021-09-28

Similar Documents

Publication Publication Date Title
Huang et al. A new building extraction postprocessing framework for high-spatial-resolution remote-sensing imagery
Ke et al. A review of methods for automatic individual tree-crown detection and delineation from passive remote sensing
Luo et al. Mapping species of submerged aquatic vegetation with multi-seasonal satellite images and considering life history information
Lu et al. Land cover classification in a complex urban-rural landscape with QuickBird imagery
Carvalho et al. Long-term evaluation of three satellite ocean color algorithms for identifying harmful algal blooms (Karenia brevis) along the west coast of Florida: A matchup assessment
CN106650812B (en) A kind of urban water-body extracting method of satellite remote-sensing image
CN104951799B (en) A kind of SAR remote sensing image oil spilling detection recognition method
Watkins et al. Automating field boundary delineation with multi-temporal Sentinel-2 imagery
CN111767813B (en) Multi-feature and long-time-sequence probability synergistic mangrove forest extraction method
CN105719283A (en) Road surface crack image detection method based on Hessian matrix multi-scale filtering
CN107403433A (en) A kind of complicated cloud infrared small target in background detection method
CN107085708A (en) High-resolution remote sensing image change detecting method based on multi-scale division and fusion
Lu et al. Detection of urban expansion in an urban-rural landscape with multitemporal QuickBird images
Shutler et al. Coccolithophore bloom detection in the north east Atlantic using SeaWiFS: Algorithm description, application and sensitivity analysis
Zhang et al. Mapping freshwater marsh species in the wetlands of Lake Okeechobee using very high-resolution aerial photography and lidar data
Yang et al. Evaluating AISA+ hyperspectral imagery for mapping black mangrove along the South Texas Gulf Coast
Luo et al. A new technique for quantifying algal bloom, floating/emergent and submerged vegetation in eutrophic shallow lakes using Landsat imagery
CN110443139A (en) A kind of target in hyperspectral remotely sensed image noise wave band detection method of Classification Oriented
Wang et al. Object-based change detection for vegetation disturbance and recovery using Landsat time series
Xu et al. Automatic detection of algal blooms using sentinel-2 MSI and Landsat OLI images
Vukadinov et al. An algorithm for coastline extraction from satellite imagery
Haest et al. An object-based approach to quantity and quality assessment of heathland habitats in the framework of Natura 2000 using hyperspectral airborne AHS images
CN108447045B (en) SAR remote sensing image water area detection method based on SAT integral graph
Tamim et al. Detection of Moroccan coastal upwelling fronts in SST images using the microcanonical multiscale formalism
CN111339909B (en) Surface linear water body extraction algorithm based on multi-scale region growth

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