CN107679476A - A kind of Sea Ice Types Classification in Remote Sensing Image method - Google Patents
A kind of Sea Ice Types Classification in Remote Sensing Image method Download PDFInfo
- Publication number
- CN107679476A CN107679476A CN201710877666.9A CN201710877666A CN107679476A CN 107679476 A CN107679476 A CN 107679476A CN 201710877666 A CN201710877666 A CN 201710877666A CN 107679476 A CN107679476 A CN 107679476A
- Authority
- CN
- China
- Prior art keywords
- data
- sea ice
- classified
- training
- classification
- 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
Links
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
- G06V20/13—Satellite images
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/24323—Tree-organised classifiers
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/08—Feature extraction
- G06F2218/10—Feature extraction by analysing the shape of a waveform, e.g. extracting parameters relating to peaks
Abstract
The present invention relates to a kind of Sea Ice Types Classification in Remote Sensing Image method, and polar region sea ice is classified using the data of satellite radar altimeter Cryosat 2, realizes based on the sea ice classification in new tool, large scale.Its step includes obtaining the latitude and longitude coordinates of measurement point and corresponding radar return waveform from the altimeter datas of CryoSat 2, extracts the relevant feature of echo waveform；AARI Sea Ice Types data are downloaded, extract the Sea Ice Types information at corresponding measurement point position；The data message of said extracted is subjected to geographical matching, the point data with latitude and longitude coordinates is converted into by MATLAB programmings；Wave character and corresponding Sea Ice Types at acquired point position are trained using random forest grader as training data, sea ice classification are carried out to data to be sorted；Sorted data are handled in ArcGIS by projective transformation, grid etc., obtain the Sea Ice Types data set that spatial resolution is 25km.
Description
Technical Field
The invention relates to a sea ice type remote sensing classification method, and belongs to the technical field of remote sensing application.
Technical Field
The sea ice type has important influence on global climate, and is one of important factors influencing sea ice thickness inversion and ship navigation. Early sea ice type data can only be obtained through field investigation, and classification methods based on optical and SAR remote sensing images are developed later, but the method is limited by the cost of the remote sensing images, small space coverage and strict use conditions, and large-scale sea ice classification is difficult to realize. With the development of satellite radar altimeters, more scholars began studying the use of altimeter data to classify sea ice because altimeters have greater spatial coverage and similarly higher temporal resolution. The classification of sea ice using altimeters is different from the application of common altimeter data, as altimeters are often used to invert sea ice thickness. Since 1990, multiple scholars successively found that different types of sea ice had some difference in radar echo waveform, and the difference can be used to classify the sea ice. However, more advanced radar altimeters are often required to achieve more accurate classification of sea ice.
The CryoSat-2 satellite transmitted by the European space Bureau in 2010, 4 months, is the most advanced satellite radar altimeter so far, has an observation data range reaching 88 degrees N in northern latitude, and has a full period of 369 days and a small period of 30 days. The vertical measurement precision of a synthetic aperture interference radar altimeter (SIRAL) carried on a satellite reaches 1-3 cm, and the ground footprint of the satellite is reduced to about 0.3km along the track and about 1.5km across the track by adopting a delay Doppler radar altimeter (DDA) technology. In addition, SIRAL performs multi-view processing on the surface points to reduce noise caused by radar spots, and the accuracy of measuring sea level height data is about 2 times that of a traditional radar altimeter.
The arctic and south Pole institute (AARI) provides ice condition maps of the south and north sea areas once a week. Its sea ice type product is obtained using optical, near infrared, SAR satellite data and vessel navigation data. The data product adopting the shape format provides five types of sea ice data: nascent ice, new ice, annual ice, perennial ice, and firm ice. Because Cryosat-2 has a large spatial coverage and annual and perennial ice are the predominant ice types in polar regions, we only used the data for annual and perennial ice.
Disclosure of Invention
The technical problem to be solved by the invention is as follows: provides a sea ice classification method based on CryoSat-2 satellite height measurement data. In view of the high spatial and temporal resolution of the altimeter data, this data greatly improves the diversity of polar sea ice classification methods, while providing a novel, fast and large-scale sea ice classification method.
In order to solve the technical problems, the technical scheme provided by the invention is as follows: the remote sensing classification method of sea ice comprises the following steps:
the first step is as follows: the method comprises the following steps of preparing training data and data to be classified, and reading information of the data, wherein the method specifically comprises the following contents in the following aspects:
1) Downloading Cryosat-2L1b data in a certain period as training data, downloading Cryosat-2L1b data in a period to be classified as data to be classified, wherein the Cryosat-2L1b data is Cryosat-2 satellite SAR mode L1 b-level data; reading an original file in a DBL format, and extracting longitude and latitude and waveform information from training data and data to be classified respectively;
2) Downloading ice condition map (sea ice type) data of An AARI (AARI) in the same period as the training data, reading an original file in a shape format, screening the data, extracting multi-year ice and annual ice data, and obtaining corresponding sea ice type information data;
secondly, extracting waveform characteristics, extracting corresponding waveform characteristics as training characteristics of a classifier for subsequent sea ice classification according to radar echo waveforms of all measurement points in training data and data to be classified, wherein the waveform characteristics mainly comprise pulse width, leading edge width, trailing edge width, stack standard deviation, maximum energy value and backscattering coefficient, the stack standard deviation and the backscattering coefficient can be directly obtained from Cryosat-2L1b data, and other waveform characteristics are obtained through waveform data calculation;
thirdly, respectively carrying out space matching on radar waveform characteristics extracted from the training data and the data to be classified and corresponding longitude and latitude coordinates, and converting the well-matched data into vector point data with the longitude and latitude coordinates;
fourthly, before classification, only using waveforms with leading edge widths less than or equal to 14, and eliminating waveforms with pulse widths less than 0.3 and stack standard deviations greater than 4;
fifthly, generating a training sample, carrying out space matching on the vector points with longitude and latitude coordinates and waveform characteristics in a certain period and sea ice type information data in the same period so as to obtain sea ice type information of each vector point, and taking the data as the training sample;
sixthly, training a random forest classifier by using the training sample, and classifying vector point data to be classified by using the random forest classifier after training is finished;
seventhly, obtaining classified sea ice type grid data with uniform spatial resolution under the same coordinate system, and specifically comprising the following steps:
a. defining the classified vector points as a WGS _1984 geographical coordinate system, and projecting polar region three-dimensional directions;
b. and gridding the classified vector points into grid data with the spatial resolution of 25km multiplied by 25km, and taking the mode of the sea ice types of all vector points in the grid as the sea ice type of the grid.
The remote sensing classification method of sea ice types of the invention also has the following improvement:
1. in the second step, the pulse width is the ratio of the maximum energy to the sum of the cumulative energies:
P i is the echo energy at the ith range gate, P max The maximum energy in the echo waveform;
the leading edge width is a number of distance gates between 1% and 99% of a maximum energy value;
the trailing edge width is the number of distance gates between 99% and 1% of the maximum energy value;
the maximum energy value refers to the maximum value of energy in the waveform.
2. And in the third step, converting the matched data into vector point data with longitude and latitude coordinates by using a shape write function in MATLAB software.
3. In the fourth step, the elimination part is greatly influenced by noise and the influence of an ice water channel, so that the data quantity to be processed is reduced, and the classification precision is improved.
4. The seventh step is realized by means of ArcGIS software.
5. The relevant parameters of the random forest classifier are set as follows: 300 classification trees, each classification node is provided with 6 feature random for selection.
The CryoSat-2 satellite data adopted by the invention belongs to satellite radar altimeter data, which is generally used for estimating the thickness of sea ice and belongs to innovation of data application in the classification of the sea ice.
The invention utilizes CryoSat-2 satellite data and AARI sea ice type data, and has relatively simple data acquisition and convenient operation. CryoSat-2 can provide accurate waveform information, and AARI sea ice type data can provide sea ice type data with extremely high space-time resolution. The random forest classifier is utilized to effectively combine the sea ice and the sea ice, so that the random forest classifier can be used for classifying the sea ice on a large scale, and a better theoretical basis is provided for the subsequent sea ice thickness estimation and climate change research.
The related data extraction and vector conversion processes are automatically realized through MATLAB programming, so that the manual participation is reduced, and the classification efficiency is improved.
In conclusion, the method has the advantages of simple and feasible execution steps and good classification effect. At present, sea ice type large-area observation data are few, and the research range of the traditional field investigation and the classification method based on optics and SAR images is extremely limited. According to the method, the Cryosat-2 data of the satellite radar altimeter with high precision and large space coverage range is used, the relation between radar echo waveforms and sea ice types is explored through a classical machine learning method, namely a random forest classifier, and sea ice classification on a large space scale is achieved. The method has profound scientific significance for accurately quantifying the balance change of the substances of the north-south polar sea ice and researching the influence of the sea ice type on the global climate.
Drawings
The invention is further described below with reference to the accompanying drawings.
FIG. 1 is a flow chart of the sea ice classifying method of the present invention.
Fig. 2 is a graph of AARI ice conditions at 3 months of 2016.
Fig. 3 shows the latitude and longitude coordinates of the extracted vector points and the data information of 6 waveform characteristics.
FIG. 4 contains information about training data for sea ice types after matching with the AARI ice condition map.
Fig. 5 shows classification results of 2016 year 3 month sea ice.
Detailed Description
The technical route and the operation steps of the present invention will be more clearly understood from the following detailed description of the present invention with reference to the accompanying drawings. The data adopted by the embodiment of the invention are CryoSat-2 satellite L1 b-level SAR mode baseline C data and AARIshapefile format data. The training data acquisition time is 2015 year 3 month 1 day-2015 year 3 month 30 days, the data to be classified acquisition time is 2016 year 3 month 1 day-2016 year 3 month 30 days, and the AARI data acquisition time is 2016 year 3 month 15 days.
Fig. 1 is a flow chart of a sea ice remote sensing classification method, which specifically comprises the following steps:
the first step is as follows: preparing training data and data to be classified, and reading information of the data, wherein the method specifically comprises the following contents in the following aspects:
1) And sequentially reading the downloaded original files in the L1 b-level DBL format of the Cryosat-2 satellite SAR mode, and respectively obtaining training data (1/3/2015-1/2015-30/2015) and data to be classified (3/1/2016-3/2016-30/2016). And extracting longitude and latitude information and radar echo waveform information of each measuring point from the two data sets.
2) Reading an AARI ice condition diagram (sea ice type) shapefile format file, screening data, extracting multi-year ice and one-year ice data, obtaining corresponding sea ice type information data, and obtaining a sea ice type distribution diagram of a research area. The AARI ice condition map at 3 months in 2016 is shown in fig. 2.
And secondly, extracting waveform characteristics. Extracting corresponding waveform characteristics as training characteristics of a classifier for subsequent sea ice classification according to the training data and radar echo waveforms of all measuring points in the data to be classified, wherein the waveform characteristics mainly comprise pulse width, leading edge width, trailing edge width, stack standard deviation, maximum energy value and backscattering coefficient, the stack standard deviation and the backscattering coefficient can be directly obtained from Cryosat-2L1b data, and other waveform characteristics are obtained through waveform data calculation. Other waveform feature calculations are as follows:
i, pulse width is the ratio of maximum energy to cumulative energy sum:
P i is the echo energy at the ith range gate, P max The maximum energy in the echo waveform.
II, leading edge width is the number of distance gates between 1% and 99% of maximum energy.
III number of gates with trailing edge width between 99% and 1% of maximum energy
IV, the maximum energy value refers to the maximum value of energy in the waveform.
And thirdly, respectively carrying out space matching on the radar waveform characteristics extracted from the training data and the data to be classified and the corresponding longitude and latitude coordinates. And converting the matched data into vector point data with longitude and latitude coordinates by using a shapewrite function in MATLAB software. The generated vector file contains longitude and latitude information and 6 waveform feature data, and the attribute values of the vector data are shown in fig. 3.
And fourthly, pre-processing in classification. Only waveforms with leading edge widths less than or equal to 14 are used because waveforms greater than 14 are more heavily affected by noise. Also, to remove possible effects of ice water channels, waveforms with pulse widths less than 0.3 and stack standard deviations greater than 4 are rejected.
And fifthly, generating a training sample. And carrying out space matching on the vector points of the training data and the AARI in the same period, extracting sea ice type information at positions corresponding to the longitude and the latitude of the vector points in the AARI ice condition map, matching the sea ice type information with corresponding point vectors, further obtaining the sea ice type information of each vector point, and taking the data as a training sample. The attribute values of the vector data are shown in fig. 4, with 0 representing years of ice and 1 representing years of ice.
And sixthly, training a random forest classifier by using the training sample, and classifying the vector point data to be classified by using the random forest classifier after the training is finished. The relevant parameters of the random forest classifier are set as follows: 300 classification trees, 6 feature random are set at each classification node for selection, and 2/3 training data can be put back for selection.
And seventhly, acquiring classified sea ice type grid data with uniform spatial resolution under the same coordinate system by using ArcGIS software. The method specifically comprises the following aspects.
a. The vector file (classified vector point data) is loaded into ArcGIS, the WGS _1984 geographical coordinate system is defined, and polar stereo azimuth projection is performed.
b. And gridding the classified vector points into grid data with the spatial resolution of 25km multiplied by 25km, and taking the mode of the sea ice types of all vector points in the grid as the sea ice type of the grid.
The specific implementation process of the step b in ArcGIS is as follows: 1. conversion Tools- > To rate- > Point To rate, frequency is selected in a calculation mode, the mode of sea ice types of all vector points in the grid is used as the sea ice type of the grid, and the size of the grid is a default value. Sea ice type grid data is obtained. 2. The spatial resolution of the above raster data is unified to 25km × 25km by spatial resampling. Data Management Tools- > scanner Processing- > sample, the nearest distribution method is selected, and the grid Data are unified to 25km multiplied by 25km spatial resolution. The classification results are shown in fig. 5.
In addition to the above embodiments, the present invention may have other embodiments. All technical solutions formed by adopting equivalent substitutions or equivalent transformations fall within the protection scope of the claims of the present invention.
Claims (6)
1. A remote sensing classification method for sea ice types comprises the following steps:
the first step is as follows: preparing training data and data to be classified, and reading information of the data, wherein the method specifically comprises the following contents in the following aspects:
1) Downloading Cryosat-2L1b data in a certain period as training data, downloading Cryosat-2L1b data in a period to be classified as data to be classified, wherein the Cryosat-2L1b data is Cryosat-2 satellite SAR mode L1 b-level data; reading an original file in a DBL format, and extracting longitude and latitude and waveform information from training data and data to be classified respectively;
2) Downloading ice condition graph data of a north pole research institute and a south pole research institute in the same period as the training data, reading an original file in a shape file format, carrying out data screening, extracting perennial ice and annual ice data, and obtaining corresponding sea ice type information data;
secondly, extracting waveform characteristics, extracting corresponding waveform characteristics as training characteristics of a classifier for subsequent sea ice classification according to radar echo waveforms of all measurement points in training data and data to be classified, wherein the waveform characteristics mainly comprise pulse width, leading edge width, trailing edge width, stack standard deviation, maximum energy value and backscattering coefficient, the stack standard deviation and the backscattering coefficient can be directly obtained from Cryosat-2L1b data, and other waveform characteristics are obtained through waveform data calculation;
thirdly, respectively carrying out space matching on radar waveform characteristics extracted from the training data and the data to be classified and corresponding longitude and latitude coordinates, and converting the well-matched data into vector point data with the longitude and latitude coordinates;
fourthly, before classification, only using waveforms with leading edge widths less than or equal to 14, and eliminating waveforms with pulse widths less than 0.3 and stack standard deviations greater than 4;
fifthly, generating a training sample, carrying out space matching on the vector points with longitude and latitude coordinates and waveform characteristics in a certain period and sea ice type information data in the same period so as to obtain sea ice type information of each vector point, and taking the data as the training sample;
sixthly, training a random forest classifier by using the training sample, and classifying vector point data to be classified by using the random forest classifier after training is finished;
seventhly, obtaining classified sea ice type grid data with uniform spatial resolution under the same coordinate system, and specifically comprising the following steps:
a. defining the classified vector points as a WGS _1984 geographical coordinate system, and projecting polar region three-dimensional directions;
b. and gridding the classified vector points into grid data with the spatial resolution of 25km multiplied by 25km, and taking the mode of the sea ice types of all vector points in the grid as the sea ice type of the grid.
2. The remote sensing classification method for sea ice types according to claim 1, characterized in that: in the second step, the pulse width is the ratio of the maximum energy to the sum of the cumulative energies:
P i is the echo energy at the ith range gate, P max The maximum energy in the echo waveform;
the leading edge width is the number of distance gates between 1% and 99% of the maximum energy value;
the trailing edge width is the number of distance gates between 99% and 1% of the maximum energy value;
the maximum energy value refers to the maximum value of energy in the waveform.
3. The remote sensing classification method for sea ice types according to claim 1, characterized in that: and in the third step, converting the matched data into vector point data with longitude and latitude coordinates by using a shape write function in MATLAB software.
4. The remote sensing classification method for sea ice types according to claim 1, characterized in that: and in the fourth step, the elimination part is greatly influenced by noise and the influence of an ice water channel, so that the data quantity to be processed is reduced, and the classification precision is improved.
5. The sea ice type remote sensing classification method according to claim 1, characterized in that: the seventh step is realized by means of ArcGIS software.
6. The remote sensing classification method for sea ice types according to claim 1, characterized in that: the relevant parameters of the random forest classifier are set as follows: 300 classification trees, 6 feature random are set at each classification node for selection.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710877666.9A CN107679476B (en) | 2017-09-26 | 2017-09-26 | Sea ice type remote sensing classification method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710877666.9A CN107679476B (en) | 2017-09-26 | 2017-09-26 | Sea ice type remote sensing classification method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107679476A true CN107679476A (en) | 2018-02-09 |
| CN107679476B CN107679476B (en) | 2020-10-09 |
Family
ID=61137356
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710877666.9A Active CN107679476B (en) | 2017-09-26 | 2017-09-26 | Sea ice type remote sensing classification method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107679476B (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109211798A (en) * | 2018-11-16 | 2019-01-15 | 福州大学 | A kind of annual sea ice distributed intelligence extracting method based on remote sensing image spectral signature |
| CN109344737A (en) * | 2018-09-12 | 2019-02-15 | 中国农业大学 | One-dimensional DCNN land use classes method based on SAR polarization decomposing |
| CN109359631A (en) * | 2018-11-30 | 2019-02-19 | 南京大学 | A kind of Sea Ice Types Classification in Remote Sensing Image method based on convolutional neural networks |
| CN109447089A (en) * | 2018-10-16 | 2019-03-08 | 同济大学 | High-resolution Sea Ice Model type-collection method based on oversubscription technology |
| CN109919042A (en) * | 2019-02-18 | 2019-06-21 | 中国科学院上海光学精密机械研究所 | A kind of oceanographic lidar echo waveform classification method, system and terminal |
| CN110956101A (en) * | 2019-11-19 | 2020-04-03 | 广东省城乡规划设计研究院 | Remote sensing image yellow river ice detection method based on random forest algorithm |
| CN112504144A (en) * | 2020-12-04 | 2021-03-16 | 南京大学 | Remote sensing estimation method for accumulated snow thickness on sea ice surface |
| CN113378766A (en) * | 2021-06-25 | 2021-09-10 | 南通大学 | Marine large-scale wind power station monitoring system based on synthetic aperture radar |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1301968A (en) * | 1999-12-30 | 2001-07-04 | 中国科学院空间科学与应用研究中心 | Land and sea compatible and three-dimensional imaging radar altimeter system and its design method |
| CN103712606A (en) * | 2013-12-27 | 2014-04-09 | 大连海事大学 | Sea ice monitoring system and monitoring method |
| US20140313072A1 (en) * | 2013-04-23 | 2014-10-23 | Conocophillips Company | Ice keel prediction from sar, optical imagery and upward looking sonars |
| CN104407338A (en) * | 2014-12-12 | 2015-03-11 | 国家卫星海洋应用中心 | Chinese HY-2 satellite microwave scatterometer-based polar sea ice identification method |
| CN105116464A (en) * | 2015-08-12 | 2015-12-02 | 南京大学 | Polar sea ice melting pool extraction method based on neural network model |
| CN106871877A (en) * | 2017-02-13 | 2017-06-20 | 国家卫星海洋应用中心 | Sea ice mark determines method and device |
| CN107064890A (en) * | 2017-04-11 | 2017-08-18 | 南京信息工程大学 | A kind of pulse radar sea ice detectivity appraisal procedure |
-
2017
- 2017-09-26 CN CN201710877666.9A patent/CN107679476B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1301968A (en) * | 1999-12-30 | 2001-07-04 | 中国科学院空间科学与应用研究中心 | Land and sea compatible and three-dimensional imaging radar altimeter system and its design method |
| US20140313072A1 (en) * | 2013-04-23 | 2014-10-23 | Conocophillips Company | Ice keel prediction from sar, optical imagery and upward looking sonars |
| CN103712606A (en) * | 2013-12-27 | 2014-04-09 | 大连海事大学 | Sea ice monitoring system and monitoring method |
| CN104407338A (en) * | 2014-12-12 | 2015-03-11 | 国家卫星海洋应用中心 | Chinese HY-2 satellite microwave scatterometer-based polar sea ice identification method |
| CN105116464A (en) * | 2015-08-12 | 2015-12-02 | 南京大学 | Polar sea ice melting pool extraction method based on neural network model |
| CN106871877A (en) * | 2017-02-13 | 2017-06-20 | 国家卫星海洋应用中心 | Sea ice mark determines method and device |
| CN107064890A (en) * | 2017-04-11 | 2017-08-18 | 南京信息工程大学 | A kind of pulse radar sea ice detectivity appraisal procedure |
Non-Patent Citations (1)
| Title |
|---|
| 赵庆平等: "SAR海冰图像分割研究", 《唐山师范学院学报》 * |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109344737A (en) * | 2018-09-12 | 2019-02-15 | 中国农业大学 | One-dimensional DCNN land use classes method based on SAR polarization decomposing |
| CN109447089A (en) * | 2018-10-16 | 2019-03-08 | 同济大学 | High-resolution Sea Ice Model type-collection method based on oversubscription technology |
| CN109447089B (en) * | 2018-10-16 | 2021-09-14 | 同济大学 | High-resolution arctic sea ice type extraction method based on super-resolution technology |
| CN109211798A (en) * | 2018-11-16 | 2019-01-15 | 福州大学 | A kind of annual sea ice distributed intelligence extracting method based on remote sensing image spectral signature |
| CN109359631A (en) * | 2018-11-30 | 2019-02-19 | 南京大学 | A kind of Sea Ice Types Classification in Remote Sensing Image method based on convolutional neural networks |
| CN109919042A (en) * | 2019-02-18 | 2019-06-21 | 中国科学院上海光学精密机械研究所 | A kind of oceanographic lidar echo waveform classification method, system and terminal |
| CN110956101A (en) * | 2019-11-19 | 2020-04-03 | 广东省城乡规划设计研究院 | Remote sensing image yellow river ice detection method based on random forest algorithm |
| CN112504144A (en) * | 2020-12-04 | 2021-03-16 | 南京大学 | Remote sensing estimation method for accumulated snow thickness on sea ice surface |
| CN112504144B (en) * | 2020-12-04 | 2021-10-29 | 南京大学 | Remote sensing estimation method for accumulated snow thickness on sea ice surface |
| CN113378766A (en) * | 2021-06-25 | 2021-09-10 | 南通大学 | Marine large-scale wind power station monitoring system based on synthetic aperture radar |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107679476B (en) | 2020-10-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107679476B (en) | Sea ice type remote sensing classification method | |
| Hodúl et al. | Satellite derived photogrammetric bathymetry | |
| CN109597074B (en) | SAR image geometric positioning parameter correction method and system | |
| CN114740180A (en) | Soil organic carbon estimation method and device based on multi-source remote sensing data | |
| CN108766203B (en) | Compact polarization rice mapping method and system | |
| Jeong et al. | Improved multiple matching method for observing glacier motion with repeat image feature tracking | |
| Leong et al. | DeepBedMap: a deep neural network for resolving the bed topography of Antarctica | |
| CN107688777A (en) | A kind of urban green space extracting method for cooperateing with multi-source Remote Sensing Images | |
| CN109100719A (en) | Combine plotting method with the topographic map of optical image based on satellite-borne SAR image | |
| Park et al. | Conceptual design of the CZMIL data processing system (DPS): algorithms and software for fusing lidar, hyperspectral data, and digital images | |
| CN105160665A (en) | Double-circle sub-template underwater terrain matching method | |
| CN111444616B (en) | Sea surface chlorophyll transmission track simulation method | |
| Stylianidis et al. | FORSAT: A 3D forest monitoring system for cover mapping and volumetric 3D change detection | |
| Halimi et al. | Including antenna mispointing in a semi-analytical model for delay/Doppler altimetry | |
| CN109359631A (en) | A kind of Sea Ice Types Classification in Remote Sensing Image method based on convolutional neural networks | |
| Xiao et al. | Re-delineating mountainous areas with three topographic parameters in Mainland Southeast Asia using ASTER global digital elevation model data | |
| CN112926468A (en) | Tidal flat elevation automatic extraction method | |
| Włodarczyk-Sielicka et al. | The process of modelling the elevation surface of a coastal area using the fusion of spatial data from different sensors | |
| CN112285808A (en) | Method for reducing scale of APHRODITE precipitation data | |
| Muslim et al. | DEM and bathymetry estimation for mapping a tide‐coordinated shoreline from fine spatial resolution satellite sensor imagery | |
| Xie et al. | Machine-Learning-Method-Based Inversion of Shallow Bathymetric Maps Using ICESat-2 ATL03 Data | |
| CN110927765A (en) | Laser radar and satellite navigation fused target online positioning method | |
| Hapep et al. | Comparison of Different DEM Generation Methods based on Open Source Datasets. | |
| CN112052720B (en) | High-space-time normalization vegetation index NDVI fusion model based on histogram clustering | |
| He et al. | An integrated use of topography with RSI in gully mapping, Shandong Peninsula, China |
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 |