Title : Human vision and its transpositions to image processing
Taking inspiration from the
significantly superior performance of humans to extract and interpret
visual information, the exploitation of biological visual mechanisms can
contribute to the improvement of the performance of computational image
processing systems. We propose a novel algorithm for automatic image
segmentation that combines elements of human visual attention with
Legendre moments applied on the probability density function of color
histograms. We also show how we can integrate entropy on top of
intensity, orientation and color to compute saliency maps and obtain
better quality segmentation.