skip to main content

Title: Automated detection of cloud and cloud-shadow in single-date Landsat imagery using neural networks and spatial post-processing

Use of Landsat data to answer ecological questions is contingent on the effective removal of cloud and cloud shadow from satellite images. We develop a novel algorithm to identify and classify clouds and cloud shadow, \textsc{sparcs}: Spacial Procedures for Automated Removal of Cloud and Shadow. The method uses neural networks to determine cloud, cloud-shadow, water, snow/ice, and clear-sky membership of each pixel in a Landsat scene, and then applies a set of procedures to enforce spatial rules. In a comparison to FMask, a high-quality cloud and cloud-shadow classification algorithm currently available, \textsc{sparcs} performs favorably, with similar omission errors for clouds (0.8% and 0.9%, respectively), substantially lower omission error for cloud-shadow (8.3% and 1.1%), and fewer errors of commission (7.8% and 5.0%). Additionally, textsc{sparcs} provides a measure of uncertainty in its classification that can be exploited by other processes that use the cloud and cloud-shadow detection. To illustrate this, we present an application that constructs obstruction-free composites of images acquired on different dates in support of algorithms detecting vegetation change.
 [1] ;  [2]
  1. University of Tennessee, Knoxville (UTK)
  2. ORNL
Publication Date:
OSTI Identifier:
DOE Contract Number:
Resource Type:
Journal Article
Resource Relation:
Journal Name: Remote Sensing; Journal Volume: 6; Journal Issue: 6
Research Org:
Oak Ridge National Laboratory (ORNL)
Sponsoring Org:
SC USDOE - Office of Science (SC)
Country of Publication:
United States
remote sensing; cloud masking; Landsat