Image similarity index based on moment invariants of approximation level of discrete wavelet transform
Subjective quality measures based on the human visual system for images do not agree well with well-known metrics such as mean squared error and peak signal-to-noise ratio. Recently, the structural similarity measure (SSIM) has received acclaim owing to its ability to produce results on a par with the human visual system. However, experimental results indicate that noise and blur seriously degrade the performance of the SSIM metric. Furthermore, despite the SSIM's popularity, it does not provide adequate insight into how it handles the 'structural similarity' of images. Proposed is a new structural similarity measure based on the approximation level of a given discrete wavelet decomposition that evaluates moment invariants to capture the structural similarity with superior results over the SSIM. 2012 The Institution of Engineering and Technology.