Multisensor Data Fusion with Singular Value Decomposition

The present study aims at multi-sensor data fusion with Singular Value Decomposition (SVD). Earth observations imaging systems collect data at different spatial and radiometric resolutions due to transmission bandwidth and other technical constraints. Fusion of multi-sensor images enables a synergy of complementary information obtained by sensors of different spectral ranges. The study illustrates the excellent potential of Singular Value Decomposition for image fusion with Quick bird panchromatic and multi spectral data. The study brings out that this fusion process outscores conventional techniques used in operational environments and is illustrated with a second example by merging IRS1C panchromatic data with IRSP6 multi spectral data.