Image formation is the process of computing or refining an image from both raw sensor data and prior information. A basic task of image formation is the extraction of the information contained in the sensor data. The information theory provides a mathematical framework to develop measures and algorithms in that process. Based on an information channel between the luminosity and composition of an image, we present three measures to quantify the saliency, specific information, and entanglement of this image associated with its luminance values and regions. The evaluation of these measures could be potentially used as a criterion to achieve more aesthetic or enhanced images.
展开▼
