12 October 2006 Cloud detection and height estimation through registration of Disaster Monitoring Constellation imagery
Author Affiliations +
For the purpose of detecting clouds over large areas it is necessary to use satellite imagery. Although a variety of techniques for cloud detection and cloud height estimation exist, they often make assumptions concerning the radiometry and spectral coverage available to the sensor payload. This paper explores the use of registration shifts observed between dual-bank single-band image pairs from the DMC multi-spectral imager to detect and estimate the height of clouds. The Disaster Monitoring Constellation (DMC) comprises a network of five disaster-monitoring micro-satellites that have been built by Surrey Satellite Technology Ltd (SSTL). Each DMC satellite has a multi-spectral imager (MSI) consisting of 2 banks of 3 channels pairs. The proposed technique uses a narrow angle between imagers to discern altitude and is comparable to stereo imaging but able to distinguish absolute cloud height without reference to the ground surface, using satellite telemetry. Simulations have shown that with a sub-degree angle between imagers and appropriate sub-pixel level registration scheme, vertical accuracies in the order of a few hundred metres maybe extracted. Preliminary results using phase correlation registration to ensure sub-pixel accuracies between DMC imagery have helped confirm the viability of the technique and will be presented alongside simulations.
© (2006) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
David C. Bamber, Stephen Mackin, Philip Palmer, "Cloud detection and height estimation through registration of Disaster Monitoring Constellation imagery", Proc. SPIE 6362, Remote Sensing of Clouds and the Atmosphere XI, 63621T (12 October 2006); doi: 10.1117/12.690003; https://doi.org/10.1117/12.690003

Back to Top