Scanning Electron Microscopy (SEM) leverages electron wavelengths for nanoscale imaging, necessitating precise parameter adjustments like focus, stigmator, and aperture alignment. However, traditional methods depend on skilled personnel and are time-consuming. Existing auto-focus and auto-stigmation techniques face challenges due to interdependent nature of these parameters and sample diversity. We propose a beam kernel estimation method to independently optimize SEM parameters, regardless of sample variations. Our approach untangles parameter influences, enabling concurrent optimization of focus, stigmator x, y, and aperture-align x, y. It achieves robust performance, with average errors of 1.00 μm for focus, 0.30% for stigmators, and 0.79% for aperture alignment, surpassing sharpness-based approach with its average errors of 6.42 μm for focus and 2.32% for stigmators and lacking in aperture-align capabilities. Our approach addresses SEM parameter interplay via blind deconvolution, facilitating rapid and automated optimization, thereby enhancing precision, efficiency, and applicability across scientific and industrial domains.