Underwater vision is one of the dominant senses and has shown great prospects in ocean investigations. In this paper, a hierarchical Independent Component Analysis (ICA) framework has been established to explore and understand the functional roles of the higher order statistical structures towards the visual stimulus in the underwater artificial vision system. The model is inspired by characteristics such as the modality, the redundancy reduction, the sparseness and the independence in the early human vision system, which seems to respectively capture the Gabor-like basis functions, the shape contours or the complicated textures in the multiple layer implementations. The simulation results have shown good performance in the effectiveness and the consistence of the approach proposed for the underwater images collected by autonomous underwater vehicles (AUVs).
References
[1]
Mostofsky, D.I. The Behavior of Fish and Other Aquatic Animals; Academic Press: New York, NY, USA, 1978.
[2]
Keenleyside, M.H.A. Diversity and Adaptation in Fish Behaviour. Zoophysiology and Ecology; Springer-Verlag: London, UK, 1979.
[3]
Fréon, P.; Dagorn, L. Review of fish associative behaviour: Toward a generalization of the meeting point hypothesis. Rev. Fish Biol. Fish. 2000, 10, 183–207.
[4]
Lowry, M.; Folpp, H.; Gregson, M.; Mckenzie, R. A comparison of methods for estimating fish assemblages associated with estuarine artificial reefs. Braz. J. Oceanogr. 2011, 59, 119–131.
[5]
Corke, P.; Detweiler, C.; Dunbabin, M.; Hamilton, M.; Rus, D.; Vasilescu, I. Experiments with Underwater Robot Localization and Tracking. Proceedings of the IEEE International Conference on Robotics and Automation, Roma, Italy, 10–14 April 2007; pp. 4556–4561.
[6]
Dudek, G.; Jenkin, M.; Prahacs, C.; Hogue, A.; Sattar, J.; Giguere, P.; German, A.; Liu, H.; Saunderson, S.; Ripsman, A.; et al. A Visually Guided Swimming Robot. Proceedings of 2005 IEEE/RSJ International Conference on Intelligent Robots and Systems, Edmonton, AB, Canada, 2–6 August 2005; pp. 3604–3609.
[7]
Balasuriya, A.; Ura, T. Vision-Based Underwater Cable Detection and Following Using AUVs. Proceedings of the MTS/IEEE OCEANS, Biloxi, MS, USA, 29–31 October 2002; pp. 1582–1587.
[8]
He, B.; Liang, Y.; Feng, X.; Nian, R.; Yan, T.; Li, M.; Zhang, S. AUV SLAM and experiments using a mechanical scanning forward-looking sonar. Sensors 2012, 12, 9386–9410.
[9]
He, B.; Zhang, H.; Li, C.; Zhang, S.; Liang, Y.; Yan, T. Autonomous navigation for autonomous underwater vehicles based on information filters and active sensing. Sensors 2011, 11, 10958–10980.
[10]
Simon, J.T.; Michele, F.-T. Seeking categories in the brain. Science 2001, 291, 260–263.
[11]
Barlow, H.B. Possible principles underlying the transformation of sensory messages. Sens. Commun. 1961, 217–234.
[12]
Anthony, J.B.; Sejnowski, T.J. The “independent components” of natural scenes are edge filters. Vision Res. 1997, 37, 3327–3338.
Herault, J.; Jutten, C. Space or time adaptive signal processing by neural network models. AIP Conf. Proc. 1986, 151, 206–211.
[15]
Christian, J.; Jeanny, H. Blind separation of sources, Pt. I: An adaptive algorithm based on neuromimetic architecture. Signal Proc. 1991, 24, 1–10.
[16]
Pierre, C. Blind separation of sources, Pt. II: Problem statement. Signal Proc. 1991, 24, 11–20.
[17]
Donckers, N.; Lendasse, A.; Wertz, V.; Verleysen, M. Extraction of Intrinsic Dimension Using CCA Application to Blind Sources Separation. In ESANN; D-Facto Public.: Bruges, Belgium, 1999; pp. 339–344.
[18]
Teh, Y.W.; Welling, M.; Osindero, S.; Hinton, G.E. Energy-based models for sparse overcomplete representations. J. Mach. Learn. Res. 2003, 4, 1235–1260.
[19]
Bell, A.J.; Sejonowski, T.J. An information–maximization approach to blind separation and blind deconvolution. Neural Comput. 1995, 7, 1129–1159.
[20]
Bell, A.J.; Sejnowski, T.J. A non-linear information maximization algorithm that performs blind separation. Adv. Neural Inf. Proc. Syst. 1995, 7, 467–474.
[21]
Amari, S.I. New learning in structural parameter space – natural Riemannian gradient. Adv. Neural Inf. Proc. Syst. 1997, 9, 127–133.
[22]
Amari, S.I. Natural gradient works efficiently in learning. Neural Component 1998, 10, 251–276.
[23]
Hyvarinen, A.; Oja, E. A fast fixed-point algorithm for independent component analysis. Neural Comput. 1997, 9, 1483–1492.
[24]
Hyvarinen, A. Fast and robust fixed-point algorithm for independent component analysis. IEEE Trans. Neural Netw. 1999, 10, 626–634.
[25]
Amari, S.; Cichocki, A.; Yang, H.H. A new learning algorithm for blind signal separation. Adv. Neural Inf. Proc. Syst. 1996, 8, 757–763.
[26]
Schwartz, O.; Simoncelli, E.P. Natural signal statistics and sensory gain control. Nat. Neurosci. 2001, 4, 819–825.
[27]
Shan, H.H.; Zhang, L.Y.; Cottrell, G.W. Recursive ICA. In Advances in Neural Information Processing Systems; MIT Press: Vancouver, BC, Canada, 2006; pp. 1273–1280.
[28]
Shan, H.H.; Cottrell, G.W. Looking Around the Backyard Helps to Recognize Faces and Digits. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Anchorage, AK, USA, 23–28 June 2008.
[29]
Hyvarinen, A.; Hoyer, P.O. A two-layer sparse coding model learns simple and complex cell receptive fields and topography from natural images. Vision Res. 2001, 41, 2413–2423.
[30]
Wainwright, M.J.; Simoncelli, E.P. Scale Mixtures of Gaussians and the Statistics of Natural Images. In Advances in Neural Information Processing Systems; MIT Press: Cambridge, MA, USA, 2000; Volume 12, pp. 855–861.
[31]
Karklin, Y.; Lewicki, M.S. A hierarchical bayesian model for learning non-linear statistical regularities in non-stationary natural signals. Neural Comput. 2005, 17, 397–423.
[32]
Osindero, S.; Welling, M.; Hinton, G.E. Topographic product models applied to natural scene statistics. Neural Comput. 2005, 18, 381–414.
[33]
Finney, E.M.; Fine, I.; Dobkins, K.R. Visual stimuli activate auditory cortex in the deaf. Nat. Neurosci. 2001, 4, 1171–1173.
[34]
Hubel, D.H.; Wiesel, T.N. Receptive fields and functional architecture of monkey striate cortex. J. Physiol. 1968, 195, 215–243.
[35]
Olshausen, B.A.; Field, D.J. Emergence of simple-cell receptive field properties by learning a sparse code for natural images. Nature 1996, 381, 607–609.
[36]
Olshausen, B.A.; Field, D.J. Sparse coding with an over-complete basis set: A strategy employed by v1? Vision Res. 1997, 37, 3311–3325.
[37]
Lee, H.; Battle, A.; Raina, R.; Ng, A.Y. Efficient sparse coding algorithms. In Advances in Neural Information Processing Systems; MIT Press: Vancouver, Canada, 2006; pp. 801–808.
[38]
Lewicki, M.S.; Olshausen, B.A. A probabilistic framework for the adaptation and comparison of image codes. J. Opt. Soc. Am. A 1999, 16, 1587–1601.
[39]
Attwell, D.; Laughlin, S.B. An energy budget for signaling in the grey matter of the brain. J. Cereb. Blood Flow Metab. 2001, 21, 1133–1145.
[40]
Jaffe, J.S. Computer modeling and the design of optimal underwater imaging systems. IEEE J. Ocean. Eng. 1990, 15, 101–111.
[41]
Jaffe, J.S.; Moore, K.D.; McLean, J.; Strand, M.P. Underwater optical imaging: Status and prospects. Oceanography 2001, 14, 66–76.
[42]
Schettini, R.; Corchs, S. Underwater image processing: State of the art of restoration and image enhancement methods. EURASIP J. Adv. Signal Proc. 2010, 2010, 746025.
[43]
Field, David J. What is the goal of sensory coding? Neural Comput. 1994, 6, 559–601.
[44]
Song, K.-S. A globally convergent and consistent method for estimating the shape parameter of a generalized Gaussian distribution. IEEE Trans. Inf. Theory 2006, 52, 510–527.
