CN113780232B - Urban wetland dynamic monitoring method - Google Patents
Urban wetland dynamic monitoring method Download PDFInfo
- Publication number
- CN113780232B CN113780232B CN202111113769.0A CN202111113769A CN113780232B CN 113780232 B CN113780232 B CN 113780232B CN 202111113769 A CN202111113769 A CN 202111113769A CN 113780232 B CN113780232 B CN 113780232B
- Authority
- CN
- China
- Prior art keywords
- time sequence
- image
- urban
- images
- sample
- 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
Links
- 238000012544 monitoring process Methods 0.000 title claims abstract description 28
- 238000000034 method Methods 0.000 title claims abstract description 25
- 239000013598 vector Substances 0.000 claims abstract description 29
- 238000007781 pre-processing Methods 0.000 claims abstract description 7
- 238000004458 analytical method Methods 0.000 claims abstract description 6
- 238000013508 migration Methods 0.000 claims abstract description 6
- 230000005012 migration Effects 0.000 claims abstract description 6
- 238000003066 decision tree Methods 0.000 claims abstract description 5
- 238000010191 image analysis Methods 0.000 claims abstract description 5
- 238000001228 spectrum Methods 0.000 claims description 15
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims description 11
- 238000007637 random forest analysis Methods 0.000 claims description 5
- 230000003595 spectral effect Effects 0.000 claims description 5
- 238000004422 calculation algorithm Methods 0.000 claims description 4
- 238000001308 synthesis method Methods 0.000 claims description 3
- 230000001131 transforming effect Effects 0.000 claims description 2
- 238000005516 engineering process Methods 0.000 abstract description 5
- 238000000605 extraction Methods 0.000 abstract description 4
- 238000011160 research Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 4
- 238000012545 processing Methods 0.000 description 4
- 238000013507 mapping Methods 0.000 description 3
- 238000011161 development Methods 0.000 description 2
- 241000120622 Rhizophoraceae Species 0.000 description 1
- 238000009360 aquaculture Methods 0.000 description 1
- 244000144974 aquaculture Species 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000015556 catabolic process Effects 0.000 description 1
- 238000006731 degradation reaction Methods 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000002372 labelling Methods 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 150000003839 salts Chemical class 0.000 description 1
Classifications
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Investigating Or Analysing Materials By Optical Means (AREA)
- Image Processing (AREA)
Abstract
The invention discloses a dynamic monitoring method of urban wetland, belonging to the technical field of remote sensing extraction and solving the problem that the existing remote sensing technology can not rapidly and accurately monitor the wetland in the urban range, and the method comprises the following steps: acquiring Landsat series images based on a global scale remote sensing cloud computing platform, preprocessing the acquired images, and establishing a time sequence image set according to the preprocessed images; acquiring feature vectors of the images, and establishing a time sequence feature vector image set; sample migration is carried out on the time series images through time series image analysis and a decision tree model, and a time series sample set suitable for each image is obtained; based on the time sequence feature vector data set and the time sequence sample set, urban wetland extraction is carried out year by year, and a time sequence urban wetland data set is established; and carrying out superposition analysis on the time sequence urban wetland data set, and analyzing the dynamic change of the urban wetland range in the time sequence to realize the monitoring of the dynamic change of the urban wetland.
Description
Technical Field
The invention relates to the technical field of urban wetland remote sensing extraction and dynamic monitoring, in particular to a urban wetland dynamic monitoring method based on a time sequence Landsat satellite image and a cloud computing platform.
Background
Urban wetland provides a series of precious ecosystem services and has huge economic and social values. However, with the increase in population and rapid economic development, urban wetlands have been subject to rapid loss, degradation and associated environmental disasters during the last three decades. With the remarkable progress of remote sensing technology and the continuous increase of available data sources, a great deal of meaningful researches on wetlands in the national and regional fields are performed by utilizing the remote sensing technology, including mangroves, aquaculture ponds, beaches, salt and biogas and the like, however, researches on quick drawing and dynamic monitoring of urban wetlands are relatively less, and effective urban wetland drawing and dynamic monitoring methods suitable for different scales cannot be provided. In classification and mapping research of urban wetlands, sample acquisition and updating are one of the most critical links, and the high space-time dynamic characteristics of the wetlands make reliable sample selection more challenging, so that the traditional method consumes a great amount of time and effort of researchers. In the conventional urban wetland mapping and dynamic monitoring research, reliable sample data cannot be obtained, so that the urban wetland dynamic monitoring with large scale and long time is difficult to realize. Therefore, a rapid and effective method is researched to carry out long-time and large-scale drawing and dynamic detection on the urban wetland, and has important significance for reasonably utilizing and protecting the wetland in the urban range. Landsat is the middle resolution optical remote sensing satellite system with the longest running time in the world at present, and is the most widely used data source for large-scale and long-term dynamic monitoring of land coverage types. Meanwhile, with further development of cloud computing and cloud storage technologies, more and more researches realize large-scale and long-time remote sensing ground object monitoring by utilizing a cloud computing platform, and a common geoscience data processing cloud computing platform, such as a global scale remote sensing cloud computing platform (Google Earth Engine), can provide Pb-level mass data products and algorithms for efficient data processing, classification and the like, and has important significance for remote sensing dynamic monitoring. By combining the Landsat images of the time sequence and the cloud computing platform, the long-time and large-scale urban wetland drawing and monitoring become possible, and the method has important significance for management, protection and reasonable utilization of the urban wetland.
Disclosure of Invention
Aiming at the problem that the existing remote sensing technology cannot rapidly and accurately monitor the wetland in the urban area, the invention provides a method for monitoring the urban wetland dynamically.
The invention discloses a dynamic monitoring method of urban wetland, which comprises the following steps:
s1, acquiring Landsat series images based on a global scale remote sensing cloud computing platform, preprocessing the acquired images, and establishing a time sequence image set according to the preprocessed images;
s2, obtaining feature vectors of the images, and establishing a time sequence feature vector image set according to the feature vectors;
s3, performing sample migration on the time sequence images in the time sequence image set through time sequence image analysis and a decision tree model to obtain a time sequence sample set suitable for each image;
s4, extracting urban wetlands year by year based on the time sequence feature vector data set and the time sequence sample set, and establishing a time sequence urban wetland data set;
s5, carrying out superposition analysis on the time sequence urban wetland data set, and analyzing dynamic change of the urban wetland range in the time sequence to realize dynamic change monitoring of the urban wetland.
Preferably, S3 includes:
s31, calculating the standard deviation of the humidity component of the time series images in the time series image set, comparing the standard deviation with a set threshold value, and dividing the time series images in the time series image set into a changed sample and an unchanged sample;
s32, using unchanged samples to create reference spectrum curves of water bodies, marshes and non-wetlands;
s33, reclassifying each change sample to obtain reclassifying samples:
calculating a spectrum angle distance SAD of the change sample and the reference spectrum curve created in the step S32, and selecting the ground object type corresponding to the largest spectrum angle distance SAD to give the change sample;
s34, combining the unchanged samples and the reclassified samples to be used as a time sequence sample set.
In the step S4, a time sequence feature vector data set and a time sequence sample set are adopted, and a random forest classification algorithm is adopted to extract the urban wetlands year by year, so that a time sequence urban wetland data set is established.
Preferably, the Landsat series of images includes Landsat5, landsat7 and Landsat8 image products
Preferably, in S1, preprocessing the acquired image includes:
(1) Removing invalid observation values such as clouds, shadows and the like in the image;
(2) Removing the topographic shadows in the images;
(3) And removing the banding error in the Landsat7 image.
Preferably, in S1, a median image synthesis method provided by a global scale remote sensing cloud computing platform is used to synthesize processed image images year by year, and a time series image set is established.
Preferably, the feature vector in S2 includes: spectral index feature vector, normalized vegetation index NDVI, normalized water index NDWI, normalized water index mNDWI, normalized building index NDBI, K-T transform principal components, humidity, greenness, brightness, texture features, second moment, contrast, and information entropy.
Compared with the traditional urban wetland drawing and dynamic monitoring method, the method has the beneficial effects that a great amount of time and labor are saved in the manufacturing process of sample data, and the quality and effectiveness of the sample data can be ensured. Meanwhile, the realization of the method is based on a Google Earth Engine cloud computing platform, compared with the traditional processing method, the efficiency of urban wetland drawing and dynamic monitoring is greatly improved, and finally, the quick and automatic drawing and dynamic monitoring of the urban wetland with large scale and long time are realized.
Drawings
FIG. 1 is a schematic diagram of the present invention;
FIG. 2 is a schematic flow chart of step three in the present invention;
FIG. 3 is a diagram of obtaining XX year high quality Landsat remote sensing images of a certain urban area;
FIG. 4 is a schematic diagram of feature vectors;
FIG. 5 is a chart showing the standard deviation of the humidity component Wetness of a time-series image of an urban area for 10 years;
FIG. 6 is a spectral plot of different types of sample data over spectral bands B1 to B7, NDVI, NDWI, mNDWI, wetness, greenness, brightness;
fig. 7 is a schematic diagram of urban wetland coverage superposition for a city of 10 years.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
It should be noted that, without conflict, the embodiments of the present invention and features of the embodiments may be combined with each other.
The invention is further described below with reference to the drawings and specific examples, which are not intended to be limiting.
The urban wetland dynamic monitoring method in the embodiment comprises the following steps:
step one, acquiring Landsat series images based on a global scale remote sensing cloud computing platform, preprocessing the acquired images, and establishing a time sequence image set according to the preprocessed images;
the first step of this embodiment specifically includes:
based on Google Earth Engine platform, landsat series of images, including Landsat5, landsat7 and Landsat8 image products, were acquired.
Pretreatment: based on Google Earth Engine platform, the acquired images are processed as follows: (1) removing invalid observed values such as clouds, shadows and the like in the images; (2) removing the topographic shadows in the image; (3) removing banding errors in the Landsat7 image.
Establishing a time sequence image set: finally, to create a long time series high quality dataset, high quality images were synthesized year by year using the median image synthesis method provided by the Google Earth Engine platform, creating a time series image set.
Step two, obtaining feature vectors of the images, and establishing a time sequence feature vector image set according to the feature vectors;
the second step of this embodiment specifically includes:
based on Google Earth Engine platform, calculate the eigenvector of image, mainly include: (1) Spectral index feature vector, normalized vegetation index NDVI, normalized water index NDWI, modified normalized water index mNDWI, and normalized building index NDBI; (2) K-T transforming principal components, humidity Wetness, greenness and Brightness Brightness; (3) Texture features, second moment ASM, contrast, and Entropy. Finally, a time series feature vector dataset is established.
Thirdly, performing sample migration on time sequence images in the time sequence image set through time sequence image analysis and a decision tree model to obtain a time sequence sample set suitable for each image;
in the third step of the embodiment, the time series sample set is created based on the Google Earth Engine platform through time series image analysis and a decision tree model. As shown in fig. 2, the method specifically includes:
1. calculating standard deviation of humidity components of the time series images in the time series image set, comparing the standard deviation with a set threshold value, and dividing the time series images in the time series image set into a changed sample and an unchanged sample;
2. creating reference spectrum curves for water, swamps and non-wetlands using the unchanged samples;
3. reclassifying each change sample to obtain a reclassifying sample:
calculating a spectrum angle distance SAD of the change sample and the reference spectrum curve created in the step S32, and selecting the ground object type corresponding to the largest spectrum angle distance SAD to give the change sample;
4. the unchanged samples are combined with the reclassified samples and then used as a time series sample set.
By 1-4 above, a set of time series samples suitable for each image is established.
Step four, extracting urban wetlands year by year based on the time sequence feature vector data set and the time sequence sample set, and establishing a time sequence urban wetland data set;
the fourth step of this embodiment specifically includes:
based on a Google Earth Engine platform, a time sequence feature vector data set and a time sequence sample set are used, a Random Forest classification algorithm Random Forest is adopted to extract the urban wetland year by year, and finally a time sequence urban wetland data set is established.
And fifthly, carrying out superposition analysis on the time sequence urban wetland data set, and analyzing dynamic change of the urban wetland range in the time sequence based on a Google Earth Engine platform so as to realize dynamic change monitoring of the urban wetland.
Specific examples:
(1) Fig. 3 shows an image acquisition and preprocessing process. Taking a certain urban area range as an example, taking a yellow range in a left image as a certain urban area in 2020, searching 18 Landsat8 remote sensing images in each period according to the certain urban area range, and obtaining 2020 high-quality Landsat remote sensing images in certain urban area through cloud mask processing, topographic shadow eliminating processing and image median composition (R: C: B=wave band 4:3: 2);
(2) Fig. 4 shows a feature vector extraction process for calculating feature vectors of an image based on a processed Landsat image of a certain urban area, including: normalized vegetation index NDVI, normalized water index NDWI, improved normalized water index mNDWI, second moment ASM, contrast, entropy, humidity component Wetness, greenness component Greenness, luminance component Brightness, and the calculation results of the feature vectors are shown in the following figures:
(3) FIG. 5 shows a sample migration process, calculates the standard deviation of the humidity component Wetness of a time-series image of a city 1990-2020 (left image), and analyzes the feature type changes of samples with different labeling differences in the time-series image;
(4) FIG. 6 shows a sample migration process, and according to sample data of a certain city 1990-2020 ground feature type, spectrum curves of different types of sample data on spectrum bands B1 to B7 and NDVI, NDWI, mNDWI, wetness, greenness, brightness are calculated as reference spectrum curves;
(5) FIG. 7 shows a process of urban wetland mapping and dynamic monitoring, wherein urban wetland in a certain urban area of 1990-2020 is further extracted by classifying by a random forest classifier based on feature vector images and sample sets of the certain urban area of 1990-2020, and the urban wetland range in the certain urban area of 1990-2020 is overlapped by a superposition analysis method for dynamic monitoring analysis;
although the invention herein has been described with reference to particular embodiments, it is to be understood that these embodiments are merely illustrative of the principles and applications of the present invention. It is therefore to be understood that numerous modifications may be made to the illustrative embodiments and that other arrangements may be devised without departing from the spirit and scope of the present invention as defined by the appended claims. It should be understood that the different dependent claims and the features described herein may be combined in ways other than as described in the original claims. It is also to be understood that features described in connection with separate embodiments may be used in other described embodiments.
Claims (4)
1. A method for dynamically monitoring urban wetlands, the method comprising:
s1, acquiring Landsat series images based on a global scale remote sensing cloud computing platform, preprocessing the acquired images, and establishing a time sequence image set according to the preprocessed images;
s2, obtaining feature vectors of the images, and establishing a time sequence feature vector image set according to the feature vectors;
s3, performing sample migration on the time sequence images in the time sequence image set through time sequence image analysis and a decision tree model to obtain a time sequence sample set suitable for each image;
s4, extracting urban wetlands year by year based on the time sequence feature vector data set and the time sequence sample set, and establishing a time sequence urban wetland data set;
s5, carrying out superposition analysis on the time sequence urban wetland data set, and analyzing dynamic change of the urban wetland range in the time sequence to realize dynamic change monitoring of the urban wetland;
in S1, preprocessing the acquired image includes:
(1) Removing invalid observation values such as clouds, shadows and the like in the image;
(2) Removing the topographic shadows in the images;
(3) Removing a stripe error in the Landsat7 image;
s3 comprises the following steps:
s31, calculating the standard deviation of the humidity component of the time series images in the time series image set, comparing the standard deviation with a set threshold value, and dividing the time series images in the time series image set into a changed sample and an unchanged sample;
s32, using unchanged samples to create reference spectrum curves of water bodies, marshes and non-wetlands;
s33, reclassifying each change sample to obtain reclassifying samples:
calculating a spectrum angle distance SAD of the change sample and the reference spectrum curve created in the step S32, and selecting the ground object type corresponding to the largest spectrum angle distance SAD to give the change sample;
s34, merging the unchanged samples and the reclassified samples to be used as a time sequence sample set;
the feature vector of the image in S2 includes: (1) Spectral index feature vectors, normalized vegetation index, normalized water index, improved normalized water index, and normalized building index; (2) K-T transforming principal components, humidity, greenness and brightness; (3) texture features, second moment, contrast, and entropy.
2. The urban wetland dynamic monitoring method according to claim 1, wherein in S4, a time-series feature vector data set and a time-series sample set are adopted, and a random forest classification algorithm is adopted to extract urban wetlands year by year, so as to establish a time-series urban wetland data set.
3. The method of claim 2, wherein the Landsat series of images includes Landsat5, landsat7 and Landsat8 image products.
4. The method for dynamically monitoring urban wetlands according to claim 1, wherein in S1, the processed image images are synthesized year by using a median image synthesis method provided by a global scale remote sensing cloud computing platform, and a time-series image set is established.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111113769.0A CN113780232B (en) | 2021-09-23 | 2021-09-23 | Urban wetland dynamic monitoring method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111113769.0A CN113780232B (en) | 2021-09-23 | 2021-09-23 | Urban wetland dynamic monitoring method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN113780232A CN113780232A (en) | 2021-12-10 |
CN113780232B true CN113780232B (en) | 2024-02-02 |
Family
ID=78852849
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202111113769.0A Active CN113780232B (en) | 2021-09-23 | 2021-09-23 | Urban wetland dynamic monitoring method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113780232B (en) |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106650689A (en) * | 2016-12-30 | 2017-05-10 | 厦门理工学院 | Coastal city time sequence land utilization information extracting method |
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 |
CN111062368A (en) * | 2019-12-31 | 2020-04-24 | 中山大学 | City update region monitoring method based on Landsat time sequence remote sensing image |
CN113327214A (en) * | 2021-05-19 | 2021-08-31 | 中国科学院地理科学与资源研究所 | Continuous time series water body remote sensing mapping method |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
DE102011076780B4 (en) * | 2011-05-31 | 2021-12-09 | Airbus Operations Gmbh | Method and device for condition monitoring, computer program product |
-
2021
- 2021-09-23 CN CN202111113769.0A patent/CN113780232B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106650689A (en) * | 2016-12-30 | 2017-05-10 | 厦门理工学院 | Coastal city time sequence land utilization information extracting method |
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 |
CN111062368A (en) * | 2019-12-31 | 2020-04-24 | 中山大学 | City update region monitoring method based on Landsat time sequence remote sensing image |
CN113327214A (en) * | 2021-05-19 | 2021-08-31 | 中国科学院地理科学与资源研究所 | Continuous time series water body remote sensing mapping method |
Non-Patent Citations (3)
Title |
---|
基于Google Earth Engine的红树林年际变化监测研究;刘凯;彭力恒;李想;谭敏;王树功;;地球信息科学学报(05);全文 * |
基于决策树模型的上海城市湿地遥感提取与分类;黄颖;周云轩;吴稳;况润元;李行;;吉林大学学报(地球科学版)(06);全文 * |
遥感监测东北地区典型湖泊湿地变化的方法研究;李晓东等;《遥感技术与应用》;第36卷(第4期);第728-239页 * |
Also Published As
Publication number | Publication date |
---|---|
CN113780232A (en) | 2021-12-10 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Mao et al. | National wetland mapping in China: A new product resulting from object-based and hierarchical classification of Landsat 8 OLI images | |
Li et al. | Mapping Moso bamboo forest and its on-year and off-year distribution in a subtropical region using time-series Sentinel-2 and Landsat 8 data | |
Sun et al. | Classification mapping and species identification of salt marshes based on a short-time interval NDVI time-series from HJ-1 optical imagery | |
Wang et al. | Comparison of IKONOS and QuickBird images for mapping mangrove species on the Caribbean coast of Panama | |
CN111738144B (en) | Surface water product generation method and system based on Google Earth Engine cloud platform | |
CN113033670B (en) | Rice planting area extraction method based on Sentinel-2A/B data | |
US8548248B2 (en) | Correlated land change system and method | |
Rendenieks et al. | Half a century of forest cover change along the Latvian-Russian border captured by object-based image analysis of Corona and Landsat TM/OLI data | |
CN112164062A (en) | Wasteland information extraction method and device based on remote sensing time sequence analysis | |
Zhou et al. | Individual tree parameters estimation for plantation forests based on UAV oblique photography | |
Zhang et al. | Extraction of tree crowns damaged by Dendrolimus tabulaeformis Tsai et Liu via spectral-spatial classification using UAV-based hyperspectral images | |
CN109584284B (en) | Hierarchical decision-making coastal wetland ground object sample extraction method | |
CN112085781B (en) | Method for extracting winter wheat planting area based on spectrum reconstruction technology | |
Xu et al. | Spatial pattern analysis of Haloxylon ammodendron using UAV imagery-A case study in the Gurbantunggut Desert | |
Calva et al. | Assessing Google Earth Pro images for detailed conservation diagnostics of mangrove communities | |
Liu et al. | Thick cloud removal under land cover changes using multisource satellite imagery and a spatiotemporal attention network | |
Guo et al. | A flexible object-level processing strategy to enhance the weight function-based spatiotemporal fusion method | |
CN113780232B (en) | Urban wetland dynamic monitoring method | |
Adepoju et al. | Improved landsat-8 OLI and sentinel-2 MSI classification in mountainous terrain using machine learning on google earth engine | |
Xiao et al. | A Novel Image Fusion Method for Water Body Extraction Based on Optimal Band Combination. | |
Zhang et al. | A method for estimating fractal dimension of tree crowns from digital images | |
CN112766090B (en) | Method and system for rapidly identifying suburb idle cultivated land by utilizing multi-season-phase Sentinel-2 image | |
Hong et al. | Mangrove extraction from super-resolution images generated by deep learning models | |
CN112633155B (en) | Natural conservation place human activity change detection method based on multi-scale feature fusion | |
CN117079059B (en) | Tree species automatic classification method based on multi-source satellite image |
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 |