Kohonen SOM for image slides sequencing

0Citations
Citations of this article
1Readers
Mendeley users who have this article in their library.

This article is free to access.

Abstract

In this paper we present a new approach to sequencing of 2-D slide images with the goal of 3-D object reconstructions. We focus our attention mostly on microscope images obtained from serial sections of biological tissues. For adequate 3-D body reconstruction the correct sequence of section should be available. We propose a modified SOM Kohonen networks with open chain topology for true sequence estimation based on based on 2-D image features invariant to section image translation and rotation, but sensitive to a scale of an object on the slide. We provide an experimental validation of the proposed method using artificial section images of a mechanical device and real microscopic histological section images.

Cite

CITATION STYLE

APA

Skubalska-Rafajłowicz, E., & Górniak, A. (2016). Kohonen SOM for image slides sequencing. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 9842 LNCS, pp. 352–365). Springer Verlag. https://doi.org/10.1007/978-3-319-45378-1_32

Register to see more suggestions

Mendeley helps you to discover research relevant for your work.

Already have an account?

Save time finding and organizing research with Mendeley

Sign up for free