Calibrating the orientation between a microlens array and a sensor based on projective geometry

• A geometric model containing nine unknown parameters is utilized to describe the optical system including a microlens array (MLA) and a sensor component.
• A calibration procedure is presented with a experimental setup for the optimization algorithm.
• The experimental results verify that the presented method can quantitatively estimate the orientation parameters between a MLA and a sensor for quality evaluation and applications of plenoptic system.

We demonstrate a method for calibrating a microlens array (MLA) with a sensor component by building a plenoptic camera with a conventional prime lens. This calibration method includes a geometric model, a setup to adjust the distance (L) between the prime lens and the MLA, a calibration procedure for determining the subimage centers, and an optimization algorithm. The geometric model introduces nine unknown parameters regarding the centers of the microlenses and their images, whereas the distance adjustment setup provides an initial guess for the distance L. The simulation results verify the effectiveness and accuracy of the proposed method. The experimental results demonstrate the calibration process can be performed with a commercial prime lens and the proposed method can be used to quantitatively evaluate whether a MLA and a sensor is assembled properly for plenoptic systems.