摘要： A limitation to the potential use of inverse methods in identifying the spatially varying properties of biological tissues consists in obtaining the required set of experimental data. In fact, whereas on the numerical side there are almost no limits to complexity in terms of geometry and material behavior of the model, on the experimental side, the collection of full-field displacement and strain data over the full surface of complex 3D geometries still represents a challenge. To address this issue, this work proposes an omnidirectional-DIC method capable of obtaining full-field information over the entire surface of native intact biological parts undergoing large deformation. This is made possible by the use of a concave conical mirror that enlarges the field of view of a conventional binocular stereo-DIC system up to 360° × 320°. The pair of highly distorted panoramic scenes reflected by the conical mirror are then matched by using a novel iterative image deformation scheme. In particular, first, a robust feature-based algorithm is used to calculate the gross disparity between images, then, an iterative NURBS-based intensity interpolation scheme refines the correlation through a hybrid local-global DIC approach. The accuracy of the developed image matching scheme was evaluated on synthetically generated images simulating a large and heterogeneous deformation field. Then, the potential of the proposed omnidirectional-DIC method was tested with rubber analogs reproducing the most representative geometries and loading states involved in biological materials characterization. Finally, as an illustrative example of application, full-field DIC shape and deformation measurements was performed over a whole porcine eye tested in-vitro under reproduced physiological load.