TY - JOUR
T1 - CloudX-net
T2 - A robust encoder-decoder architecture for cloud detection from satellite remote sensing images
AU - Kanu, Sumit
AU - Khoja, Rohit
AU - Lal, Shyam
AU - Raghavendra, B. S.
AU - CS, Asha
N1 - Funding Information:
This publication is the outcome of R & D work undertaken in the Young Faculty Research Fellowship project under Visvesvaraya PhD Scheme of Ministry of Electronics & Information Technology (MeitY), Government of India in the National Institute of Technology Karnataka, Surathkal being implemented by Digital India Corporation (formerly Media Lab Asia), New Delhi, Grant No. DIC/MUM/GA/10(37)D, Dated 24-01-2019.
Publisher Copyright:
© 2020 Elsevier B.V.
PY - 2020/11
Y1 - 2020/11
N2 - Cloud Detection is an important pre-processing step for any application involving remote sensing data. This paper presents a deep learning based CloudX-Net architecture, that can detect cloud cover with improved accuracy in comparison to the benchmark from satellite remote sensing images. The proposed CloudX-Net model reduces the number of parameters needed for accurate predictions and thus make deep learning based cloud detection method very efficient. Atrous Spatial Pyramid Pooling (ASPP) and Separable convolution are used to optimize the network. For experimentation, we have used Landsat 8 images and 38-Cloud dataset and trained the architectures using Soft Jaccard loss function. Comparing several quantifying metrics result from various recent deep learning architectures proves the efficiency and effectiveness of the proposed CloudX-Net model for cloud detection from satellite images. The source code and data are available at https://github.com/shyamfec/CloudXNet.
AB - Cloud Detection is an important pre-processing step for any application involving remote sensing data. This paper presents a deep learning based CloudX-Net architecture, that can detect cloud cover with improved accuracy in comparison to the benchmark from satellite remote sensing images. The proposed CloudX-Net model reduces the number of parameters needed for accurate predictions and thus make deep learning based cloud detection method very efficient. Atrous Spatial Pyramid Pooling (ASPP) and Separable convolution are used to optimize the network. For experimentation, we have used Landsat 8 images and 38-Cloud dataset and trained the architectures using Soft Jaccard loss function. Comparing several quantifying metrics result from various recent deep learning architectures proves the efficiency and effectiveness of the proposed CloudX-Net model for cloud detection from satellite images. The source code and data are available at https://github.com/shyamfec/CloudXNet.
UR - https://www.scopus.com/pages/publications/85092145086
UR - https://www.scopus.com/pages/publications/85092145086#tab=citedBy
U2 - 10.1016/j.rsase.2020.100417
DO - 10.1016/j.rsase.2020.100417
M3 - Article
AN - SCOPUS:85092145086
SN - 2352-9385
VL - 20
JO - Remote Sensing Applications: Society and Environment
JF - Remote Sensing Applications: Society and Environment
M1 - 100417
ER -