Cloud Extraction Scheme for Multi-Spectral Images Using Landsat-8 OLI Images With High Brightness Reflectivity Covered

Abstract

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 characteristics as clouds, and thus, it is easily confused with clouds in cloud extraction schemes. This work presents a novel scheme designed to extract clouds in satellite imagery with high brightness reflectivity covered. The fractal summation method and spatial analysis are used to extract the clouds in the Landsat 8 Operational Land Imager (OLI) images containing high brightness reflectivity covered. The scheme consists of three main steps: cloud extraction based on pixel values, Anselin Local Moran's I value, and anisotropy. Pixel values were applied to extract the clouds associated with anomalies, and the last two steps were conducted to eliminate false anomalies. The findings showed that the cloudassociated anomaly pixel-values well approximate a power-law function, but both the real and fake anomaly patches (e.g., snow/ice, desert, etc.) routinely coexist within the same (fractal) scaleless segments, and that the latter seems to be more significant than the former. Consequently, these results indicate that the diagnostic difference between true and false anomalies must lie in their spatial distribution patterns. Furthermore, experiments confirmed that the fractal dimension and spatial distribution (i.e. Anselin Local Moran's I index and anisotropy) difference between the real and false anomalies displayed a certain universality. The proposed scheme effectively reduces the confusion and misclassification caused by cloud, snow and the highlighted underlying surface. It is of great significance for cloud restoration processing, image analysis, image matching, target detection and extraction, and effective extraction and utilization of remote sensing data.

Cloud Screening Method in Complex Background Areas Containing Snow and Ice Based on Landsat 9 Images

“…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.…”

“…Although each of these methods has achieved certain results, they can only give a relatively coarse classification result when the image is located in a complex terrain area or the subsurface contains snow and ice [ 10 ]. The threshold method helps to reduce misclassification to some extent, but it is difficult to accurately determine the threshold, and so achieving high precision remains challenging [ 4 ].…”