Figure - available via license: Creative Commons Attribution 4.0 International
Content may be subject to copyright.
Source publication
Cloud extraction is a vital step in remote sensing image processing. Although many advanced cloud extraction methods have been proposed and confirmed to be effective in recent years, there are still difficulties in cloud extraction in areas of high brightness reflectivity covered. High brightness reflectivity cover can have similar spectral charact...
Context in source publication
Context 1
... PC3, the brightness of the clouds is evident, which is exactly the opposite of the case of PC1. Quantitatively, Table 1 records the statistical output of the PCA. It is evident that PC1 reflects the presence of clouds because it is positive in Band 3 and negative in Band 6 and Band 9. ...Citations
... The threshold-based cloud screening method aims to set appropriate thresholds analyzing the reflectance and brightness temperature values of clouds. Since the thr old method is simple and easy to implement, the current cloud screening metho mainly based on the threshold method, but it depends on the threshold setting and extraction effect will be disturbed by the subsurface [2][3][4]. The short-wave infrared b is widely used in cloud screening because the reflectivity of snow in this spectral rang usually lower than that of clouds. ...
... The threshold-based cloud screening method aims to set appropriate thresholds by analyzing the reflectance and brightness temperature values of clouds. Since the threshold method is simple and easy to implement, the current cloud screening method is mainly based on the threshold method, but it depends on the threshold setting and the extraction effect will be disturbed by the subsurface [2][3][4]. The short-wave infrared band is widely used in cloud screening because the reflectivity of snow in this spectral range is usually lower than that of clouds. ...
The first step in the application of Landsat 9 imagery is cloud screening, and the International Satellite Cloud Climatology Project (ISCCP) has made cloud screening an important part of the World Climate Research Program. The accurate identification of clouds in remote sensing satellite images containing snow and ice on the subsurface has been a challenging task in the cloud screening process. It is imperative to fully exploit the characteristic heterogeneous information of the cloud and snow subsurface, to solve the problem of cloud–snow confusion in the snow and ice environment, and to carry out research on cloud screening technology without interference from the snow and ice subsurface. In view of this, this paper will systematically carry out research on cloud screening methods in snow and ice environments. In this paper, we propose the building of a cloud screening algorithm that takes into account the difficulty of the fact that snow and ice on the subsurface can easily interfere with cloud recognition, and the influence of an empirical threshold or statistical threshold that makes its application less effective, and then establish a dynamic threshold cloud screening algorithm that is suitable for snow and ice environments. The research results will provide new ideas and perspectives to solve the problem of surface-type interference that most of the existing cloud screening algorithms contend with.
Detection of clouds and shadows in remote sensing imagery is important due to its wide range of applications. There are a lot of applications in remote sensing images such as monitoring of the environment, change detection etc. It is an important and booming research area. Ineffective and inaccurate cloud and cloud shadow masking will cause undesirable effects on different task that can be performed by using remote sensing images. Because of high spectral conglomeration and the spectral and temperature discrepancy of the underlying surface the detection of clouds and associated shadows is not candid. In this paper, we propose CSDUNet a modified U-Net network for precise pixel-wise semantic segmentation of cloud and its associated shadow from optical remote sensing images. It uses an encoder network and a decoder network. This method concatenated feature maps at different scales. We have proposed a novel network for cloud detection, which extract features corresponding cloud and shadow at different scales from multilevel layers to generate sharp boundaries. Which will help to detect clouds in heterogeneous landscape, under complex underlying surfaces with varying geometry. Experimental analysis on the Landsat satellite dataset proves that the proposed CSDUNet achieves a dice coefficient of 95.05%. Our method got 95.93% precision, recall of 94.71% and Jaccard index of 97.29%. CSDUNet achieves accurate detection accuracy and surpass several traditional methods.
The land use and land cover map plays a significant role in agricultural, water resources planning, management, and monitoring programs at regional and national levels and is an input to various hydrological models. Land use and land cover maps prepared using satellite remote sensing techniques in conjunction with landform-soil-vegetation relationships and ground truth are popular for locating suitable sites for the construction of water harvesting structures, soil and water conservation measures, runoff computations, irrigation planning and agricultural management, analyzing socio-ecological concerns, flood controlling, and overall watershed management. Here we use a novel approach to analyze Sentinel–2 multispectral satellite data using traditional and principal component analysis based approaches to evaluate the effectiveness of maximum likelihood estimation, random forest tree, and support vector machine classifiers to improve land use and land cover categorization for Soil Conservation Service Curve Number model. Additionally, we use stratified random sampling to evaluate the accuracies of resulted land use and land cover maps in terms of kappa coefficient, overall accuracy, producer's accuracy, and user's accuracy. The classifiers were used for classifying the data into seven major land use and land cover classes namely water, built-up, mixed forest, cultivated land, barren land, fallow land with vertisols dominance, and fallow land with inceptisols dominance for the Vishwamitri watershed. We find that principal component analysis with support vector machine is able to produce highly accurate land use and land cover classified maps. Principal component analysis extracts the useful spectral information by compressing redundant data embedded in each spectral channel. The study highlights the use of principal component analysis with support vector machine classifier to improve land use and land cover classification from which policymakers can make better decisions and extract basic information for policy amendments.
(Download Link in comment)
