OSA's Digital Library

Journal of the Optical Society of America A

Journal of the Optical Society of America A

| OPTICS, IMAGE SCIENCE, AND VISION

  • Vol. 15, Iss. 2 — Feb. 1, 1998
  • pp: 359–366

Optimum linear combination strategy for an N-channel polarization-sensitive imaging or vision system

J. S. Tyo  »View Author Affiliations


JOSA A, Vol. 15, Issue 2, pp. 359-366 (1998)
http://dx.doi.org/10.1364/JOSAA.15.000359


View Full Text Article

Enhanced HTML    Acrobat PDF (263 KB)





Browse Journals / Lookup Meetings

Browse by Journal and Year


   


Lookup Conference Papers

Close Browse Journals / Lookup Meetings

Article Tools

Share
Citations

Abstract

The optimum linear combination channels for an N-receptor polarization-sensitive imaging or vision system are found by using a principal-components analysis. The channels that are derived are optimum in the sense that their information contents are uncorrelated when considered over the ensemble of possible polarization signals. For a two-receptor system, the optimum channels are shown to be the sum and the difference of the outputs of the individual receptors. As a corollary, the optimal arrangement of the two receptors is shown to be a mosaic of identical, orthogonally aligned linear polarization analyzers. The implications of these results on the development of a representational scheme for polarization information are discussed.

© 1998 Optical Society of America

OCIS Codes
(110.0110) Imaging systems : Imaging systems
(230.5440) Optical devices : Polarization-selective devices
(260.5430) Physical optics : Polarization

History
Original Manuscript: April 9, 1997
Revised Manuscript: September 8, 1997
Manuscript Accepted: September 16, 1997
Published: February 1, 1998

Citation
J. S. Tyo, "Optimum linear combination strategy for an N-channel polarization-sensitive imaging or vision system," J. Opt. Soc. Am. A 15, 359-366 (1998)
http://www.opticsinfobase.org/josaa/abstract.cfm?URI=josaa-15-2-359


Sort:  Author  |  Year  |  Journal  |  Reset  

References

  1. R. Walraven, “Polarization imagery,” in Optical Polarimetry: Instrumentation and Applications, R. M. A. Azzam, D. L. Coffeen, eds., Proc. SPIE112, 164–167 (1977). [CrossRef]
  2. W. G. Egan, W. R. Johnson, V. S. Whitehead, “Terrestrial polarization imagery obtained from the Space Shuttle: Characterization and interpretation,” Appl. Opt. 30, 435–442 (1991). [CrossRef] [PubMed]
  3. L. B. Wolff, T. E. Boult, “Constraining object features using a polarization reflectance model,” IEEE Trans. Pattern Anal. Mach. Intell. 13, 635–657 (1991). [CrossRef]
  4. L. B. Wolff, T. A. Mancini, “Liquid crystal polarization camera,” in Proceedings of the IEEE Workshop on Applications of Computer Vision (IEEE, New York, 1992), pp. 120–127.
  5. L. B. Wolff, “Polarization camera for computer vision with a beam splitter,” J. Opt. Soc. Am. A 11, 2935–2945 (1994). [CrossRef]
  6. G. D. Gilbert, J. C. Pernicka, “Improvement of underwater visibility by reduction of backscatter with a circular polarization technique,” in Underwater Photo Optics I, A. B. Dember, ed., Proc. SPIE7, A-III-I–A-III-10 (1966).
  7. M. P. Rowe, E. N. Pugh, J. S. Tyo, N. Engheta, “Polarization-difference imaging: a biologically inspired technique for imaging in scattering media,” Opt. Lett. 20, 608–610 (1995). [CrossRef] [PubMed]
  8. S. G. Demos, R. R. Alfano, “Temporal gating in highly scattering media by the degree of optical polarization,” Opt. Lett. 21, 161–163 (1996). [CrossRef] [PubMed]
  9. J. S. Tyo, M. P. Rowe, E. N. Pugh, N. Engheta, “Target detection in optically scattering media by polarization-difference imaging,” Appl. Opt. 35, 1855–1870 (1996). [CrossRef] [PubMed]
  10. J. E. Solomon, “Polarization imaging,” Appl. Opt. 20, 1537–1544 (1981). [CrossRef] [PubMed]
  11. G. F. J. Garlick, C. A. Steigman, W. E. Lamb, “Differential optical polarization detectors,” U.S. patent3,992,571 (November16, 1976).
  12. J. S. Tyo, E. N. Pugh, N. Engheta, “Colorimetric representations for use with polarization-difference imaging of objects in scattering media,” J. Opt. Soc. Am. A 15, 367–374 (1998). [CrossRef]
  13. G. D. Bernard, R. Wehner, “Functional similarities between polarization vision and color vision,” Vision Res. 17, 1019–1028 (1977). [CrossRef] [PubMed]
  14. G. Buchsbaum, A. Gottschalk, “Trichromacy, opponent colours coding and optimum colour information transmission in the retina,” Proc. R. Soc. London, Ser. B 220, 89–113 (1983). [CrossRef]
  15. Photoreceptors actually capture radiation in discrete quanta; however, if the intensity of the radiation is strong enough, the discretization can be ignored, and the above formulation can be applied.
  16. S. M. Kay, Modern Spectral Estimation: Theory And Application (Prentice-Hall, Englewood Cliffs, N.J., 1988).
  17. In this section the ensemble is assumed to be uniformly distributed in angle of polarization with a fixed ellipticity described by the parameter ∊. These results can be easily generalized to include other ensembles by simply altering the pdf’s used to calculate the entries in the correlation matrix.
  18. T. W. Anderson, An Introduction to Multivariate Statistical Analysis (Wiley, New York, 1984), Chap. 11.
  19. Z. Zhao, N. H. Farhat, “Tomographic microwave diversity image reconstruction employing unitary compression,” IEEE Trans. Microwave Theory Tech. 40, 315–322 (1992). [CrossRef]
  20. A. Bermann, R. J. Plemmons, Nonnegative Matrices in the Mathematical Sciences (Academic, New York, 1979), Chap. 4.
  21. C. E. Shannon, The Mathematical Theory of Communication (U. of Illinois Press, Urbana, Ill., 1949).
  22. C. H. Papas, Theory of Electromagnetic Radiation (Dover, New York, 1988), pp. 118–134.
  23. Although choosing N>3 does not seem to make sense in the context explored herein, some simple, fast, nonlinear systems may be proposed that operate optimally with N>3.
  24. D. A. Cameron, E. N. Pugh, “Double cones as a basis for a new type of polarization vision in vertebrates,” Nature (London) 353, 161–164 (1991). [CrossRef]
  25. M. P. Rowe, N. Engheta, S. S. Easter, E. N. Pugh, “Graded index model of a fish double cone exhibits differential polarization sensitivity,” J. Opt. Soc. Am. A 11, 55–70 (1994). [CrossRef]
  26. The four Stokes parameters can be determined (up to an ambiguity in sign of one of the parameters) with only three intensity measurements for monochromatic radiation.22 Because circular polarization and unpolarized light are indistinguishable with a system that detects only linear polarization states, their effects on the final representation should be similar, and partially circularly polarized monochromatic radiation is analyzed here for mathematical clarity.
  27. This representation of the incident field is exact up to the choice of a constant. Although the linearly polarized portion of the radiation is constrained to have unit amplitude here, in general it can have arbitrary amplitude. In the general case, A is the ratio between the amplitudes of the circularly polarized and linearly polarized portions of the incident radiation.
  28. Nonbirefringent, nonoptically active photoreceptors are considered for simplicity, since most linear polarization analyzers satisfy this requirement. In a biological PVS, the photoreceptors may in fact be birefringent or optically active, but this can be accounted for by adding appropriate phase delay terms to the P and Q matrices (birefringence) or adding off-diagonal terms to these matrices (optical activity). As mentioned in Section 2, to sense the complete state of polarization, at least one birefringent receptor is needed either to yield the intensity of a circular polarization state or to give a relative phase value.
  29. G. Buchsbaum, J. L. Goldstein. “Optimum probabilistic processing in colour perception. II. Colour vision as template matching,” Proc. R. Soc. London, Ser. B 205, 249–266 (1979). [CrossRef]
  30. The matrix given by Eq. (19) [as well as Eq. (10)] is real symmetric and therefore an example of a self-adjoint matrix. Self-adjoint M×M matrices always produce M orthogonal eigenvectors that span the vector space operated on by the matrix.20 Correlation matrices like the ones treated here are always self-adjoint, even if the random variables are complex.16 The fact that any M×M correlation matrix necessarily spawns M orthogonal eigenvectors is the basis of principal-components analysis.18
  31. In their study Bernard and Wehner13 assumed that the output of the individual receptors was proportional to the logarithm of the input. In this investigation the receptors are assumed to respond linearly within a particular range of intensities.
  32. J. S. Tyo, “Polarization-difference imaging: a means for seeing through scattering media,” Ph.D. dissertation (University of Pennsylvania, Philadelphia, 1997).

Cited By

Alert me when this paper is cited

OSA is able to provide readers links to articles that cite this paper by participating in CrossRef's Cited-By Linking service. CrossRef includes content from more than 3000 publishers and societies. In addition to listing OSA journal articles that cite this paper, citing articles from other participating publishers will also be listed.

Figures

Fig. 1 Fig. 2 Fig. 3
 
Fig. 4
 

« Previous Article  |  Next Article »

OSA is a member of CrossRef.

CrossCheck Deposited