Backscattering polarimetric images derive structural information on scattering media

Optical polarization is the orientation of the planes of oscillation of the electric field vectors for many light waves. For over three centuries, physicists have attempted to measure this phenomenon. To date, the Stokes-vector/Mueller-matrix polarimetry has shown positive results and consequently has been adopted widely. So far, this elegant and concise mathematical marvel that contains an abundance of polarimetric information, if suitably parametrized, could enable the optical and physical features of the probed system to be extracted. Nonetheless, if the probed system is not a pure retarder or polarizer/diattenuator, the resultant Perrin-Mueller (PM) matrix is difficult to decipher and the system cannot be characterized in a straightforward fashion.

Recently, attempts have been made on possibilities of extracting polarimetric properties from the PM matrix by developing mathematical factorization/decomposition techniques. Unfortunately, owing to the complexity of the matrix, the developed approaches do not necessarily do justice. Worse off, the manner in which the intricate interaction of the light with multiple scattering systems manifests itself in polarimetric data is scarcely understood.

To this effect, University of Bern scientists: Manes Hornung (PhD candidate), Arushi Jain (PhD candidate), Professor Martin Frenz and Dr. Günhan Akarçay from the Institute of Applied Physics in Switzerland, presented a study where they demonstrated that polarimetric images recorded in the backscattering geometry could be interpreted by analyzing the spatial variations of the backscattered light’s Stokes vectors and using symmetry/geometry arguments. Of great significance, they focused on exploring the possibility of interpreting polarimetric images recorded in the backscattering geometry without resorting to a decomposition of the PM matrix. Their work is currently published in the research journal, Optics Express.

To begin with, the Swiss research team first elucidated on how the macroscopic polarimetric properties of a suspension could be quantified by analyzing the spatial variations of backscattered Stokes vectors and by using symmetry arguments. This was closely followed by the introduction of an analytical, semi-microscopic light propagation model that had the capability to qualitatively reproduce the measurements. In general, they related the semi-microscopic model parameters to the previously calculated macroscopic properties with the purpose of making mathematically explicit the interconnections between diattenuation, degree of polarization and retardance.

The physicists reported that their model further enlightened on the interdependence between the macroscopic diattenuation, depolarization and retardance; attributes that had often been taken as polarimetric parameters when interpreting measurements. In addition, following the introduction of the analytical model that qualitatively reproduced their measurements, a need to redefine the macroscopic polarimetric properties that were actually relevant to characterize non-crystallic multiply scattering systems was exposed.

In a nutshell, with the objective in mind to interpret polarimetric images recorded in the backscattering geometry and gain insight into the underlying physics and with the example of measurements on colloidal suspensions, the researchers articulately demonstrated that with a thorough cross-examination of the information contained in the backscattered Stokes vectors separately for different input states, one could comprehend the polarization altering properties of multiply scattering systems. Overall, the team presented an analytical model based on the coherency matrix and the geometric phase to describe the polarimetric behavior of the probed samples. This study provides insights that will be of great importance in analyzing and differentiating for example brain tissue structures from glioma tumors thus enhancing the diagnosing performance of these polarimetric images.