The classification of forest vegetation types plays an important role in the land management agencies for natural resource inventory information, especially for federally protected national forests in China. The classification results are widely used in the calculation and inversion of parameters such as forest storage volume, biomass and coverage. Forest canopy density response the extent to which the canopy is connected to each other in the forest. It can be used to observe vegetation growth. In recent years, deep learning convolutional neural networks have made significant progress in the task of remote sensing image classification and recognition. Considering that the spectral characteristics of forests in different seasons in Jilin Province of China are quite different, this paper used the optical image data of Sentinel-2A in summer, spring and autumn as the data source to calculate the normalized difference vegetation index (NDVI), bare index (BI), perpendicular vegetation index (PVI) and shadow index (SI). Next use the four vegetation indexes combined with weighted overlay analysis method to calculate forest canopy density. In this paper, the convolutional neural network (CNN) was used as the forest vegetation type classifier. The classification indexes were the spectral data and the spectral data combined with the forest canopy density information, respectively. The experiment shows that the forest canopy density can significantly improve the classification accuracy and the overall accuracy is increased from 85.58% to 90.41%.
With the development of space satellite technology, a large number of high resolution remote sensing images have emerged. Deep learning has become an effective way to handle big data. Crop classification can estimate crop planting area and structure, and classification result is an important input parameter for crop yield model. Crop classification based on deep learning can further improve the estimation accuracy of production. In this paper, multi-temporal Sentinel-2 data and GF2 data are used as data sources. Sentinel-2 data is used as training data, and GF-2 data is used as validation data. Jilin Province in Northeast of China is selected as the experimental area. The experimental area is classified as rice, towns, corn and soy. Firstly, the multi-temporal Sentinel-2 data and GF-2 data is preprocessed. Then, the Sentinel-2 data is used to classify crops based on convolutional neural network (CNN) and visual geometry group (VGG). The red edge band, multiple indexes including normalization difference vegetation index (NDVI), normalized difference water index (NDWI) and difference vegetation index (DVI) are added respectively to compare with the classification results of original multitemporal Sentinel-2 data. The final classification results using CNN and VGG are compared with the other two machine learning algorithms including support vector machine (SVM) and random forest (RF). The experimental results show that the VGG performs best in crop classification accuracy. The overall classification accuracy of the crop can reach 94.8878%, and the Kappa coefficient can reach 0.9253, which is superior to the two traditional machine learning algorithms.
As an important factor for global climate change, snow affects local and global radiative balances of the earth. Excessive snow can cause destroy for global hydrological cycle and climate system. In recent years, the use of passive microwave remote sensing to retrieval snow has made greatly progress. Snow deep retrieval algorithms and snow-covered products can provide spatial and temporal information on snow cover distribution, which is an important data source for snow monitoring. The accuracy validation and contrastive analysis of snow deep retrieval algorithms are helpful to further development of snow retrieval in China. Northern Xinjiang, Qinghai-Tibet Plateau and Inner Mongolia-Northeast China are stable snow areas in China. Relying on the survey project of snow cover characteristics and distribution in China, the snow survey route has been carefully designed to continuously observe whole dry snow period (December 2017 to March). FengYun3B microwave radiation imager (FY3B-MWRI) brightness temperature data and MODIS land cover product data are used in this paper. The accuracy of snow depth retrieval algorithms, including FY operational algorithm, NASA series algorithm and GlobSnow snow water equivalent product algorithm, shows that the FY operational algorithm has the best result, and the root mean square error and deviation are 8.91cm, 6.4cm, respectively. However, the accuracy of NASA series algorithms and GlobSnow snow water equivalent product algorithm is seriously influenced by land cover type.
Active microwave remote sensing is one of the important methods to monitor snow cover and retrieve snow parameters such as snow water equivalent (SWE), snow particle size and snow density. SWE provides useful information for hydrological, climatic and meteorological applications. Retrieving SWE from active microwave remote sensing data has attracted widespread attention in the academic community. This paper took Nongan County, Jilin Province, China as the study area, and selected two scenes of Sentinel-1 interference wide-mode (IW) SAR images to retrieve the SWE in the shallow dry snow area in winter. A mathematical expression that defines the relationship between the backscattering coefficient ratio and the thermal resistance calculated from the measured snow data was established. By using the relationship between the measured thermal resistance and the measured SWE, the value of SWE in the study area can be retrieved by the backscattering coefficient ratio. The result shows that the method used in this paper has high precision (MAE=0.198cm, RMSE=0.24cm). It is feasible to use the C-band Sentinel-1 SAR data to retrieve SWE.
Crop classification can accurately estimate crop area, structure, and spatial distribution, as well as provide important input parameters for crop yield models. The crop yield information is an important basis for the country to formulate food policies and economic plans, so the study of crop classification is of great significance. Traditional optical remote sensing is susceptible to sunlight and clouds, and Synthetic Aperture Radar (SAR) can be used all-time and all-weather. Compared to single- polarization SAR, full-polarization SAR has more abundant information. In this paper, C-band GF- 3(GaoFen-3) satellite data and multi-temporal Sentinel-1 data were used as data sources. Changchun City in northeastern China is selected as the experimental area and the scattering characteristics of typical crops in this area are analyzed. Firstly, the GF-3 and multi-temporal Sentinel-1 SAR data were preprocessed. Then, polarization decomposition of GF-3 data was performed to obtain three polarization characteristics: scattering angle, entropy and anisotropy (H/α/A). The Supports vector machine (SVM) algorithm was implemented as the classifier. Polarization characteristics, multi-source and multi-temporal SAR were used for classification features. The overall accuracy reached 91.9537%, nearly 10% higher than using full-polarization information alone, and the kappa coefficient was 0.8827. It shows that multi-source and multi-temporal SAR has obvious advantages in crop identification.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.