Comparison of Discrete Wavelet Transform (DWT), Discrete Cosine Transform (DCT) and Stationary Wavelet Transform (SWT) based Satellite Image Fusion Techniques

Abstract

The aim of the image fusion is to combining evidence from different images; Multispectral (MS) and Panchromatic (PAN) images acquired from different sensors of the same interpretation in directive to convey enhancedspectral and spatial information as well.In this paper discrete wavelet transform (DWT) and two specializations of discrete cosine transform (DCT); i)DCT varianc,ii) consistency verification with DCT variance fusion techniques are implemented and compared with the proposed methodology for image fusion named stationary wavelet transform (SWT).Fused results obtained from these fusion approaches are assessed through typical evaluation parameters.Fused outcomes obtained from proposed SWT outperforms DWT and two flavors of DCT based fusion approaches.The shift invariant property of SWT produces improved spectral and spatial evidence in the fused image followed by fused grades accomplished from DCT based fusion approaches.The discrete cosine transforms (DCT) grounded approaches of image fusion are further proper and performance oriented in real time applicationsby means of DCT founded principles of static images.Conclusion through this work is a glowing systematic practice for fusion of multi-focus images based on SWT is presented and proved that SWT based fused results surpass other fusion approaches.