Image Colorfulness Measure for Multilayer Compressed Sensing Model

Colorfulness is an aspect of the visual perception, according to which the color of an object is perceived to be more or less chromatic. However, the existing colorfulness measures face a problem that may fail to measure color distortion for the compressed sensing and even multilayer compressed sensing images. In this paper, we propose a Multilayer Compressed Sensing (Mul-CS) model and a new colorfulness measure (CM-MulCS) for the model. The Mul-CS model is based on an improved Gaussian random measurement matrix, in which the feature map of each layer is visualized by a proposed Gaussian Fitting Visualization (GFV) algorithm. The CM-MulCS is based on a color noise intensity adaptive weighting model which benefits measure of color distortion for the compressed sensing images and the feature maps in Mul-CS model. The experimental results show that GFV and CM-MulCS for Mul-CS model perform better than some popular existing visualization algorithms and colorfulness measures, respectively.