Efficient autofocusing in digital holography using the Tanimoto coefficient

This manuscript proposes a novel, to the best of our knowledge, autofocusing method for digital holography using the Tanimoto coefficient (TC) to assess similarity between dual-channel reconstructions (real and imaginary components). Additionally, connected component analysis is used to extract regions of interest (ROI), improving autofocus accuracy. Unlike existing similarity-based methods, this dual-channel TC approach eliminates the need for RGB information and offers over 90 times faster computation than the TR-MUSIC method. The method's effectiveness is validated through simulations and experimental trials.