Contrast enhancement in polarimetric imaging with correlated noise flucutations

Auteurs :
Publication MaxEnt 2014


Contrast enhancement in polarimetric imaging with correlated noise flucutations


application/pdf Contrast enhancement in polarimetric imaging with correlated noise flucutations


74.72 Ko


Creative Commons Aucune (Tous droits réservés)


Sponsors scientifique


Sponsors logistique


Sponsors financier

<resource  xmlns:xsi=""
        <identifier identifierType="DOI">10.23723/9603/11320</identifier><creators><creator><creatorName>Swapnesh Panigrahi</creatorName></creator><creator><creatorName>Julien Fade</creatorName></creator><creator><creatorName>Mehdi Alouini</creatorName></creator></creators><titles>
            <title>Contrast enhancement in polarimetric imaging with correlated noise flucutations</title></titles>
        <resourceType resourceTypeGeneral="Text">Text</resourceType><dates>
	    <date dateType="Created">Sat 30 Aug 2014</date>
	    <date dateType="Updated">Mon 2 Oct 2017</date>
            <date dateType="Submitted">Sun 24 Mar 2019</date>
	    <alternateIdentifier alternateIdentifierType="bitstream">2555734ccf68da4d4d5b196a7e4c54f334d9b822</alternateIdentifier>
            <description descriptionType="Abstract"></description>

Contrast enhancement in polarimetric imaging with correlated noise fluctuations S. Panigrahi, J. Fade, M. Alouini Institut de Physique de Rennes, France Polarimetric imaging has been a topic of research in various fields like medical imaging [1], machine vision, remote sensing and contrast enhancement through turbid media [2]. In general, polarimetric imaging provides a multidimensional image that needs to be reduced to a single image for interpretation. This image is produced such that the contrast of the polarimetric non-uniformities in the imaged scene are maximized. Using information theoretic analysis, we have shown that a polarization sensitive imager can out-perform a simple intensity camera when estimating a parameter on a polarized source embedded in partially polarized and intense background[3]. We assume a gaussian noise model with correlated fluctuations, and compute the gain in measurement precision by taking the ratio of Fisher information (FI) of a parameter estimated by polarimetric imaging to one obtained by simple intensity imager. The analysis indicates that polarimetric imaging always overcomes the precision of a standard intensity camera when the background noise is sufficiently correlated. As a result, we derive gain maps with respect to parameters P (degree of polarization, DOP, of source), β (DOP  of background) , ρ (background correlation) indicating situations that favour the use polarimetric  imager over intensity imager. Further, we choose a favourable situation (with P=1 and β=0) to experimentally implement a long range polarimetric imaging system. The experiment is set up over a distance of about 1.3 km to study imaging through real atmospheric fog [2]. The study of a such a system has tremendous applications in navigation. Thus, to allow for real-time vision enhancement during foggy atmospheric conditions, we use a wollaston prism-based snapshot polarimetric camera to image a polarized source of incoherent light embedded in real atmospheric fog. We have studied various simple representations of the polarimetric image and quantified the contrast obtained by each using contrast-to-noise ratio (CNR). The CNRs from each representation has been compared experimentally in various visibility conditions during foggy weather. It is indeed interesting to compare the maximum gain in contrast obtained experimentally with the theoretically predicted gain in measurement precision. In light of this, we resort to Maximum-likelihood approach to derive an optimal estimator and compare the experimentally obtained contrast gain as a function of background correlation and compare it with theoretical prediction. A preliminary analysis shows a remarkable match and further work is under progress (soon to reported). Finally, from a estimation and detection point of view, it will be a good opportunity for me to widen my knowledge towards the techniques used in field of geometric information. References: [1] M. P. Rowe, J. S. Tyo, N. Engheta, and E. N. Pugh, “Polarization-difference imaging: a biologically inspired technique for observation through scattering media,” Opt. Lett. 20, 608–610 (1995) [2] H. Ramachandran and A. Narayanan, “Two-dimensional imaging through turbid media using a continuous wave light source,” Opt. Commun. 154, 255–260 (1998). [3] J. Fade, S. Panigrahi, and M. Alouini, "Optimal estimation in polarimetric imaging in the presence of correlated noise fluctuations," Opt. Express 22, 4920-4931 (2014). [4] J. FADE, S. Panigrahi, A. Carre, L. Frein, C. Hamel, F. Bretenaker, H. Ramachandran, and M Alouini, "Long range polarimetric imaging through fog", Applied Optics (2014) (accepted)