부산대학교 도서관

Summary: ``In a large number of measuring and characterizationmethods, the sphere size distribution of particles embedded in amedium has to be estimated from profile radii observed in crosssections or thin slices. On the one hand, this is a typical problem ofdensity estimation. For that reason, kernel estimators may beapplied. On the other hand, the computation of the sphere sizedistribution from the profile size distribution requires the inversionof an integral equation. In the case of cross sections or infinitelythin slices, this is an ill-posed problem which can be solved withspecific methods as, e.g., Tikhonov regularization. In thiscontribution we propose a method for unfolding the sphere sizedistribution given a sample of profile radii which combines theadvantages of kernel estimators with those of regularizationmethods. In order to study and test this method, Monte Carlosimulations have been performed. The results demonstrate that themethod is reliable and leads to properly unfolded sphere sizedistributions. Finally, the method has been applied to experimentaldata obtained from transmission electron microscopy images of severalpolymer blends. The results demonstrate that the new method is avaluable tool for data analysis.''