Abstract
One of the major challenges in biomedical imaging is the extraction of quantified information from the acquired images. Light and tissue interaction leads to the acquisition of images that present inconsistent intensity profiles and thus the accurate identification of the regions of interest is a rather complicated process. On the other hand, the complex geometries and the tangent objects that very often are present in the acquired images, lead to either false detections or to the merging, shrinkage or expansion of the regions of interest. In this paper an algorithm, which is based on alternating sequential filtering and watershed transformation, is proposed for the segmentation of biomedical images. This algorithm has been tested over two applications, each one based on different acquisition system, and the results illustrate its accuracy in segmenting the regions of interest.
© 2009 OSA/SPIE
PDF ArticleMore Like This
Yang Bingqian, Wang Mingquan, Zhang Junsheng, and Gao Jinkai
104526H Education and Training in Optics and Photonics (ETOP) 2017
Nianchao Wang, Linghao Hu, Johanna Webb, and Alex J. Walsh
JTu4A.22 Bio-Optics: Design and Application (BODA) 2021
Neelam Gupta
JTuA49 Conference on Lasers and Electro-Optics (CLEO:S&I) 2007