K-means clustering for support construction in diffractive imaging

Shunsuke Hattanda; Hiroyuki Shioya; Yosuke Maehara; Kazutoshi Gohara

doi:10.1364/JOSAA.31.000470

K-means clustering for support construction in diffractive imaging

J Opt Soc Am A Opt Image Sci Vis. 2014 Mar 1;31(3):470-4. doi: 10.1364/JOSAA.31.000470.

Authors

Shunsuke Hattanda, Hiroyuki Shioya, Yosuke Maehara, Kazutoshi Gohara

PMID: 24690641
DOI: 10.1364/JOSAA.31.000470

Abstract

A method for constructing an object support based on K-means clustering of the object-intensity distribution is newly presented in diffractive imaging. This releases the adjustment of unknown parameters in the support construction, and it is well incorporated with the Gerchberg and Saxton diagram. A simple numerical simulation reveals that the proposed method is effective for dynamically constructing the support without an initial prior support.