Automated, Universal Software for Cloud and Cloud Shadow Detection in RS Data

Limitations with existing cloud cover detection techniques for large dataset processing and new challenges presented by the increase in the quantity and quality of data in the commercial realm, offer an opportunity for R&D into new and improved methods for the detection of clouds and cloud shadows in acquired imagery. We propose to develop innovative system for automated pixel-based cloud and cloud shadow detection. The novel, iterative, self-guided approach will rely on spectral, spatial, and contextual information from a limited number of bands (R-G-B or R-G-B-NIR) and will be applicable to large datasets of a wide range of commercial and government space- and air-borne imagery. Phase II will continue the algorithm development and refinement, and include formal algorithm validation and optimization. Further, a prototype turn-key system will be developed. Our innovative algorithms address the need for streamlined acquisition and automated processing of very large volumes of RS data. They fill NASA?s technology needs and fit into the overall NASA mission by: ? Employing rapid analysis methodologies and algorithms, ? Improving the automated process of quality assurance /quality control for science data products, and ? Facilitating the efficient collection of data.