Image-processing approach to vibration isolation of a scanning electron microscope

摘要

We propose a new approach to estimating a pointing error of the electron probe of a scanning electron microscope. The approach is formulated to estimate the pointing errors using a specimen image. Specimen images are numerically simulated by a mathematical model and identified with measured images of the specimen using a least-squares procedure to determine the pointing errors. The pointing errors are estimated by the proposed approach and used to design a controller for vibration isolation of a scanning electron microscope. Acceleration sensors are located at the root of the specimen chamber and are used to detect any environmental disturbance into the microscope. The designed controller is based on a transfer function from the sensor outputs to the pointing errors, and the transfer function is determined by sinusoidal excitation tests for the microscope. The controller is implemented as a digital filter on a PC and is used to move the electron probe to cancel the pointing errors using image-shifting coils. The pointing errors are successfully reduced by the controller in a lower frequency region that contains the first four natural frequencies