Autofocusing is crucially important for automated microscopy, where fully automatic acquisition of microscope images in unattended operation is required. In addition to providing automation of imaging for scanning microscopy applications, it facilitates objective, accurate, and consistent image measurements for quantitative analysis. Autofocusing algorithms determine the in-focus position for an image based on maximization of a focus function, which represents the measure of focus as a function of the axial z position, and is sampled at different positions along the z-axis. The value of the focus function is computed from an image captured at that z position. Research on microscope autofocusing dates back over a quarter of a century, and many autofocusing functions and algorithms have been proposed in the literature [1][2][3][4]. A commonality of existing autofocusing methods is that all of them are based on single-resolution image analysis, and have inherent limitations. Recent developments in signal processing and wavelet transform theory, however, suggest an alternative approach to tackle the problem and to overcome the limitations by resorting to multiresolution image analysis.

No abstract