Three-dimensional (3-D) objects classification by means of phase-only digital holographic information using Alex Network

This paper presents the application of the Alex network for binary classification of three-dimensional (3-D) objects in phase-only digital holographic information. The 3-D objects considered for the binary classification task include ‘triangle-square’, ‘circle-square’, ‘square-circle’, and ‘triangle-circle’. The ‘triangle-square’ object is assigned to the TRUE class, while the ‘circle-square’, ‘square-circle’, and ‘triangle-circle’ objects are assigned to the FALSE class. The digital holograms of 3-D objects were generated using the two-step phase-shifting digital holography (PSDH) technique and processed computationally to extract phase-only digital holographic data. Following this, the Alex Network was trained on a dataset comprising 2880 phase-only images to yield the outcomes. The study validates its findings by presenting results such as the confusion matrix, ROC curve, and precision-recall characteristics.