GENERALIZED RÉNYI IMAGE ENTROPY: A NEW NOISE MEASURE
Salvador Gabarda, Gabriel Cristóbal · Fluctuation and Noise Letters · 2007
Noise cannot be distinguished from information by simply measuring entropy. Nevertheless, the Rényi entropy can be used to calculate the entropy in a pixel-wise basis. Consequently, entropy histograms of images can be obtained. Hence, histograms of entropy can be used to quantify differences in the information contents of images. In this paper, entropy histograms of noisy images has been analyzed and results have been applied to define an index that measures the noise contents of natural images. The pixel-wise entropy of digital images has been calculated through the use of a spatial/spatial-frequency distribution. Therefore generalized Rényi entropy and a normalized windowed pseudo-Wigner distribution (PWD) have been selected to define a particular pixel-wise entropy. In this way, the shape of such a distribution of entropy indicates the amount of noise present in the image. Some examples are presented and discussed.