Improving Spectral Quality of IHS-Pansharpening Result by Integrating Equalization Process using SVE-DWT for Satellite Imagery Data

Abstract

Remote Sensing satellite such as WorldView-4 has limitations in producing images that have advantages on both spectral and spatial resolution. This problem makes it necessary to divide the images into high spectral resolution multispectral (MS) and high spatial resolution panchromatic (PAN) images. The method of image fusing such as intensity-hue-saturation (IHS) pansharpening causes spectral distortion. This drawback is caused by overusing the pixel values from a PAN image. In this study, a new method was proposed that integrating IHS-pansharpening with Singular Value Equalization (SVE) as equalization process and Discrete Wavelet Transform (DWT) for preserving edge and detail information of the PAN image. The SVE method equalizes values on PAN images by intensity values in the IHS transformation process. The spectral quality produced from the proposed method shows the superiority to the original method when tested using spectral quality metrices like: RMSE, Spectral Angle Mapper (SAM) and Relative Dimensionless Synthesis Global Error (ERGAS).