The image compression based on principal component analysis is a well-established fundamental technique. Organs, cells and micro-structures in cells dealt with in biomedical image analysis are volumetric images. Tensor expression of volumetric images allows us to derive unified frameworks to process and analyse these volumetric images dealt with in medical and biological sciences. We develop a fast approximate tensor principal component analysis for volumetric images by using tensor decomposition.

This presentation is part of Contributed Presentation “CP7 - Contributed session 7”

Authors:

Atsushi Imiya (IMIT, Chiba University)

Keywords:

computer vision, image compression, image representation, medical and biological images, tenosr principal component analysis, tensor decomposition, volumetric images