Abstract
With the rapid advancement of biomedical imaging device technology, the role of image processing coupled with pattern recognition, which is the science of classification of objects into a number categories or classes, is increasingly becoming important for the study of biological and medical images. As a matter of fact, the term Bioimage Informatics has been coined in the community of bioinformatics to emphasize the essential application of computational methods in assisting life-science researchers in the quantitative analysis of large volumes of images. This paper presents the formulation of the Kolmogorov-Sinai entropy under the mechanism of fuzzy uncertainty as an effective image feature for complex pattern classification. In this study, the particular focus is on classifying biological images relating to aging. The performance of the proposed method was tested against several current challenging image datasets that are benchmarks for comparing image classification techniques for overcoming the limited capacity of existing automated biological image analysis. The results demonstrate the superiority of the use of the new image feature for classifying difficult image objects in current biology.