Comparison of Gini index and Tamura coefficient for holographic autofocusing based on the edge sparsity of the complex optical wavefront

• DOI: 10.48550/arxiv.1708.08055
• Publication Date: 2017-01-01
• Authors: Miu Tamamitsu; Yibo Zhang; Hongda Wang; Yichen Wu; Aydogan Özcan
• Keywords: Region of interest

The Sparsity of the Gradient (SoG) is a robust autofocusing criterion for holography, where gradient modulus complex refocused hologram calculated, on which sparsity metric applied. Here, we compare two different choices metrics used in SoG, specifically, Gini index (GI) and Tamura coefficient (TC), holographic dense/connected or sparse samples. We provide theoretical analysis predicting that uniformly distributed image data, TC GI exhibit similar behavior, while naturally images containing few high-valued signal entries many low-valued noisy background pixels, more sensitive to distribution changes resistive noise. These predictions are also confirmed by experimental results using SoG-based dense connected samples (such as stained breast tissue sections) well highly isolated Giardia lamblia cysts). Through these experiments, found ToG GoG offer almost identical performance samples, whereas should be calculated relatively small region interest (ROI) closely surrounding object, offers flexibility choosing larger ROI pixels.