Optimal Parametric Analysis of Non‐Adaptive and Adaptive Sinc‐Based Image Zoom Techniques

ABSTRACT Image zoom techniques play a crucial role in numerous applications, notably in biomedical imaging, where high‐resolution visualization of fine details is essential. Traditional interpolation techniques often suffer from suboptimal image quality when the sampling rate is not adequately high. In this study, we systematically test and validate both adaptive and non‐adaptive Sinc‐based zoom technologies, utilizing the Whittaker–Shannon interpolation formula (Sinc interpolation) and the Nyquist–Shannon sampling theorem (cardinal theorem of interpolation). Through experiments exploring various sampling rate and bandwidth combinations, we identify optimal parameter ranges for successful zoom and empirically elucidate the underlying mechanisms. Our results show that adaptive and non‐adaptive Sinc‐based zoom can achieve robust zoom, with the adaptive technique demonstrating superior performance, particularly when sampling rates significantly exceed the bandwidth. Here, we present the key contributions of our research: (1) determination of optimal parameter combinations for high‐quality zoom, (2) explanation of the zoom mechanism through empirical analysis and (3) comparison of adaptive and non‐adaptive Sinc‐based zoom techniques. For the non‐adaptive technique, the ideal bandwidth range is between 0.2 and 0.5 spatial frequency units (sfu). The corresponding sampling rate is between 0.4 and 1.4 sfu. The optimal range for the adaptive technique is attained with sampling rates in between 0.4 and 10 sfu. The implications of this study are discussed in the context of MRI reconstruction, where Sinc‐based zoom technology can significantly enhance the visualization of small anatomical features.