A new adaptive image fusion technique for IKONOS satellite imagery

Abstract

Preservation of spectral information and the enhancement of spatial resolution are regarded as very important in satellite image fusion. In previous research, many algorithms simultaneously unsolved these problems, or needed experimental parameters to enhance fusion performance. This paper proposed a new fusion method based on fast intensity-huesaturation (FIHS) to merge a high-resolution panchromatic image with a low-resolution multispectral image. It is conducted by multiple regressions for generating synthetic image and statistical ratio-based image enhancement, which is presented as solving the spectral distortion and conserving the spatial information of the panchromatic image. IKONOS datasets were employed in the evaluation. The results showed that the proposed method was better than the widely used image fusion methods, including the FIHS-based method and the Pan Sharpening module in PCI Geomatica. We compared widely used algorithms with adaptive FIHS image fusion using various fusion quality Indexes such as ERGAS, RASE, correlation, and the Q4 index. The images obtained from the proposed algorithm present higher spectral and spatial quality than the results from using other fusion methods. Therefore, the proposed algorithm is very efficient for high-resolution satellite image fusion with an automatic process.