A new IHS and wavelet based pansharpening algorithm for high spatial resolution satellite imagery

In remote sensing applications, pan-sharpening or image fusion focuses on improving the low spatial resolution (LSR) of multispectral (MS) images by injecting the details information selected from the panchromatic (PAN) images with high spatial resolution (HSR). This paper presents a new pansharpening alternative based on the combination of the well-known Intensity-Hue-Saturation (IHS) method and multiscale analysis by using wavelets. The multiscale wavelet decomposition has been achieved by using the "a trous" wavelet transform (ATWT). In this novel approach, a new approximation of the weighting coefficients of the multispectral bands based on the combination of energy and correlation coefficients of the MS and PAN images is proposed. Therefore, the intensity element is extracted from the up-sampled MS images and the weighting coefficients of the intensity component are computed by using the average energy ratio and the correlation coefficients between the wavelet-based decomposed PAN image and the different MS bands in order to achieve more detail information. The detail information is then injected to the MS bands to obtain a fused image with high spectral and spatial information. The experimental results confirmed that a superior trade-off between spectral and spatial quality can be achieved by using the proposed approach. Moreover, compared to numerous classical and state-of-the-art pansharpening methods, the proposed scheme can inject more spatial information and guarantee more color preservation. (C) 2020 COSPAR. Published by Elsevier Ltd. All rights reserved.