A Kurtosis-Based Statistical Measure for Two-Dimensional Processes and Its Applications to Image Sharpness

Published

June 30, 2003

Author(s)

Nien F. Zhang, Andras Vladar, Michael T. Postek, Robert D. Larrabee

Abstract

Fully automated or semiautomatic scanning electron microscopes (SEM) are now commonly used in semiconductor production and other forms of manufacturing. It is required that these automated instruments be routinely capable of 3 nanometer (nm) or better resolution below 1kV accelerating voltage for the measurement of a nominal 70-150 nm size parts of the integrated circuits. The testing and proving that the instrument is performing at this level for production on a day-by-day basis, however, has not been routinely employed. Once a human operator is no longer monitoring the instrument s performance and multiple instruments are concerned, an objective diagnostic procedure must be implemented to ensure data and measurement fidelity.

Citation

Proceedings of the 2003 Section on Physical and Engineering Sciences

Pub Type

Journals

Keywords

autocorrelation, discrete Fourier transform, periodograms, scanning electron microscope, stationary process

Citation

Zhang, N. , Vladar, A. , Postek, M. and Larrabee, R. (2003), A Kurtosis-Based Statistical Measure for Two-Dimensional Processes and Its Applications to Image Sharpness, Proceedings of the 2003 Section on Physical and Engineering Sciences (Accessed October 11, 2025)