Article ID Journal Published Year Pages File Type
533231 Pattern Recognition 2016 6 Pages PDF
Abstract

•Mistakes are pointed out in Xiao et al. [1].•Corrected IEFMs are proposed as a response to Xiao et al. [1].•Experiments and theoretical analysis on corrected IEFMs.

In this paper, we demonstrate that Improved Exponent–Fourier moments mentioned in Xiao et al. (2015) [1] are with crucial mistakes in its definition, an image cannot be decomposed and reconstructed by these incorrect Improved Exponent–Fourier moments. The corrected definition of Improved Exponent–Fourier moments is thus given in this paper to solve problems about numerical instability involved in the process of calculating the Exponent–Fourier moments and values of radial functions approach infinity at origin. Experimental results show the superiority of the corrected Improved Exponent–Fourier moments.

Related Topics
Physical Sciences and Engineering Computer Science Computer Vision and Pattern Recognition
Authors
, , ,