作者: Lahouari Ghouti , Ahmed Bouridane
DOI:
关键词:
摘要: In this paper, we propose a new perceptual model for balanced multiwavelet (BMW) transforms. The latter transform achieves simultaneous orthogonality and symmetry without requiring any input prefiltering. proposed is derived using multiresolution domain extensions of Chou's model. depends only on the image activity not multifilter sets used by transform, unlike those developed scalar wavelets. redundancy, present in image, efficiently quantified through just-noticeable distortion (JND) profile. model, visibility threshold assigned to each BMW subband coefficient. Therefore, perceptually insignificant components can be clearly disc requirement often encountered watermarking applications.riminated from significant ones. For instance, discrimination constructively achieve imperceptibility