Multiresolution Fully Convolutional Networks to detect Clouds and Snow through Optical Satellite Images

01/07/2022

∙

Clouds and snow have similar spectral features in the visible and near-infrared (VNIR) range and are thus difficult to distinguish from each other in high resolution VNIR images. We address this issue by introducing a shortwave-infrared (SWIR) band where clouds are highly reflective, and snow is absorptive. As SWIR is typically of a lower resolution compared to VNIR, this study proposes a multiresolution fully convolutional neural network (FCN) that can effectively detect clouds and snow in VNIR images. We fuse the multiresolution bands within a deep FCN and perform semantic segmentation at the higher, VNIR resolution. Such a fusion-based classifier, trained in an end-to-end manner, achieved 94.31 for clouds on Resourcesat-2 data captured over the state of Uttarakhand, India. These scores were found to be 30 10 cloud detection purposes, the study also highlights the potential of convolutional neural networks for multi-sensor fusion problems.

READ FULL TEXT

Multiresolution Fully Convolutional Networks to detect Clouds and Snow through Optical Satellite Images

End-to-end Cloud Segmentation in High-Resolution Multispectral Satellite Imagery Using Deep Learning

Data fusion of satellite imagery for generation of daily cloud free images at high resolution level

Part Segmentation for Highly Accurate Deformable Tracking in Occlusions via Fully Convolutional Neural Networks

Object-based Probabilistic Similarity Evidence of Sparse Latent Features from Fully Convolutional Networks

Fusion of Heterogeneous Data in Convolutional Networks for Urban Semantic Labeling (Invited Paper)

Recurrent Multiresolution Convolutional Networks for VHR Image Classification

Using Fully Convolutional Neural Networks to detect manipulated images in videos

Multiresolution Fully Convolutional Networks to detect Clouds and Snow through Optical Satellite Images

Related Research

End-to-end Cloud Segmentation in High-Resolution Multispectral Satellite Imagery Using Deep Learning

Data fusion of satellite imagery for generation of daily cloud free images at high resolution level

Part Segmentation for Highly Accurate Deformable Tracking in Occlusions via Fully Convolutional Neural Networks

Object-based Probabilistic Similarity Evidence of Sparse Latent Features from Fully Convolutional Networks

Fusion of Heterogeneous Data in Convolutional Networks for Urban Semantic Labeling (Invited Paper)

Recurrent Multiresolution Convolutional Networks for VHR Image Classification

Using Fully Convolutional Neural Networks to detect manipulated images in videos