Improved cloud detection for Landsat 8 images using a combined neural network model. Issue 3 (3rd March 2020)

• Record Type: Journal Article
• Authors: Ma, Nan; Sun, Lin; Wang, Quan; Yu, Zhenjun; Liu, Sichao
• Is Part Of: Remote sensing letters. Volume 11:Issue 3(2020)
• Page Start: 274
• Page End: 282
• Publication Date: 2020-03-03
• Subjects: Remote sensing -- Periodicals; Remote sensing; Periodicals; 621.3678
• DOI: 10.1080/2150704X.2019.1708503
• Languages: English
• ISSNs: 2150-704X

Abstract

ABSTRACT: High-precision cloud detection is a key step in the processing of remote sensing imagery. However, the existing cloud detection methods struggle to extract high-accuracy cloud pixels, especially for images of thin and fragmented clouds or those over high-brightness surfaces. In this study, we developed a new model by combining the existing models of Fully Convolutional Network-8 sample (FCN-8s) and U-network (U-net) (based on the three visible bands) to take full advantage of spectral and spatial information. In the proposed Fully Convolutional Network Ensembling Learning (FCNEL) model, U-net and FCN-8s initially conduct separate classifications based on their relative strengths, and their outputs are fused by the voting strategy to integrate multi-scale features from both the models. Different surface and cloud types in Landsat 8 Operational Land Imager (OLI) data were used to test the model, which showed an average overall accuracy of 91.68% and an average producer accuracy of 98.52%. Thus, the proposed FCNEL model was superior to FCN-8s or U-net as well as the widely used function of mask algorithm. The proposed method has good adaptability to various cloud types and diverse underlying surface environments.