Enhancing Image Resolution with Pixel-Shift and Precision Motion Control

With the rapid advancement of digital imaging and display technologies, the demand for high-quality images has grown significantly across various fields. Image super-resolution (SR) techniques, which reconstruct high-resolution (HR) images from low-resolution (LR) inputs, offer a practical alternative to traditional hardware upgrades. These techniques provide enhanced visual detail for applications such as small target detection, medical imaging, industrial inspection, and satellite remote sensing. This paper focuses on multi-frame super-resolution reconstruction and proposes a hardware-assisted approach using a high-precision piezoelectric micro-positioning stage to drive a CCD sensor with subpixel-level shifts. This enables the acquisition of a sequence of LR images with well-defined shift relationships. By leveraging these precise shifts, SR algorithms can effectively reconstruct higher-resolution images. Experimental results demonstrate that, for scale factors of ×2 and ×3, the reconstructed images exhibit significantly improved detail compared to the original LR images. However, as the scaling factor increases further, the improvement becomes limited due to the modulation transfer function (MTF) of the imaging system, mounting errors, and noise, which can even lead to distortion in the reconstructed results. This study provides an efficient and reliable solution for hardware-assisted SR imaging.