Reservoir Computing for Prediction of the Spatially-Variant Point Spread Function

A new method is presented which provides prediction of the spatially variant point spread function for the restoration of astronomical images, distorted by atmospheric turbulence when viewed using ground-based telescopes. Our approach uses reservoir computing to firstly learn the spatio-temporal evolution of aberrations caused by turbulence, and secondly, predicts the space-varying point spread function (PSF) for application of widely-used deconvolution algorithms, resulting in the restoration of astronomical images. In this article, a reservoir-based, recurrent neural network is used to predict modal aberrations that comprise the spatially variant PSF over a wide field-of-view using a time-series ensemble from multiple reference beacons.

[1]  José Carlos Príncipe,et al.  Analysis and Design of Echo State Networks , 2007, Neural Computation.

[2]  R. Y. Webb,et al.  Dynamic Artificial Neural Networks for Centroid Prediction in Astronomy , 2006, 2006 Sixth International Conference on Hybrid Intelligent Systems (HIS'06).

[3]  James G. Burnett,et al.  Optical propagation through low-level turbulence , 1998, Remote Sensing.

[4]  Jochen J. Steil,et al.  Online reservoir adaptation by intrinsic plasticity for backpropagation-decorrelation and echo state learning , 2007, Neural Networks.

[5]  Michel Tallon,et al.  Focusing on a Turbulent Layer: Principle of the “Generalized SCIDAR” , 1998 .

[6]  D. Fried.  Anisoplanatism in adaptive optics , 1982 .

[7]  David G. Sandler,et al.  Adaptive optics: Neural network wavefront sensing, reconstruction, and prediction , 1999 .

[8]  Michael Lloyd-Hart.  SPATIO-TEMPORAL PREDICTION FOR ADAPTIVE OPTICS WAVEFRONT RECONSTRUCTORS , 2007 .

[9]  Roberto Ragazzoni,et al.  Adaptive-optics corrections available for the whole sky , 2000, Nature.

[10]  Danilo P. Mandic,et al.  Recurrent Neural Networks for Prediction: Learning Algorithms, Architectures and Stability , 2001 .

[11]  Russell Y. Webb,et al.  MULTI-LAYER CORRECTIVE CASCADE ARCHITECTURE FOR ON-LINE PREDICTIVE ECHO STATE NETWORKS , 2008, Appl. Artif. Intell..

[12]  J. Conan,et al.  Efficient phase estimation for large-field-of-view adaptive optics. , 1999, Optics letters.

[13]  Russell Webb.  A Cascade Architecture for On-line Predictive Echo State Networks , 2007, MLMTA.

[14]  Herbert Jaeger,et al.  Overview of Reservoir Recipes , 2007 .

[15]  Danilo P. Mandic,et al.  Recurrent Neural Networks for Prediction: Learning Algorithms, Architectures and Stability , 2001 .

[16]  R. Foy,et al.  Adaptive telescope with laser probe : isoplanatism and cone effect , 1990 .

[17]  A. Wolf,et al.  Determining Lyapunov exponents from a time series , 1985 .

[18]  J. Goodman.  Introduction to Fourier optics , 1969 .

[19]  E. Marchetti,et al.  Modal tomography for adaptive optics , 1999 .

[20]  R. Lane,et al.  Fast simulation of a kolmogorov phase screen. , 1999, Applied optics.

[21]  S. Engelberg.  Random Signals and Noise , 2006 .

[22]  Harald Haas,et al.  Harnessing Nonlinearity: Predicting Chaotic Systems and Saving Energy in Wireless Communication , 2004, Science.

[23]  George J. M. Aitken,et al.  Temporal analysis of stellar wave-front-tilt data , 1997 .

[24]  Claire E. Max,et al.  Introduction to Adaptive Optics , 2004 .

[25]  Mathieu Aubailly,et al.  Prediction of the space-varying point spread function for reconstruction of anisoplanatic adaptive optics images , 2005, SPIE Optics + Photonics.

[26]  L. Cao.  Practical method for determining the minimum embedding dimension of a scalar time series , 1997 .

[27]  B. Welsh,et al.  Imaging Through Turbulence , 1996 .

[28]  M C Roggemann,et al.  Prediction of wave-front sensor slope measurements with artificial neural networks. , 1997, Applied optics.

[29]  Richard G Lane,et al.  Extended analysis of curvature sensing. , 2002, Journal of the Optical Society of America. A, Optics, image science, and vision.

[30]  T. Fusco,et al.  Characterization of adaptive optics point spread function for anisoplanatic imaging. Application to stellar field deconvolution , 2000 .

[31]  George J. M. Aitken,et al.  Genetic algorithm design of complexity-controlled time-series predictors , 2003, 2003 IEEE XIII Workshop on Neural Networks for Signal Processing (IEEE Cat. No.03TH8718).

[32]  D. A. de Wolf.  Optical propagation through turbulent air , 1977 .

[33]  Robert K. Tyson.  Introduction to Adaptive Optics , 2000 .

[34]  M.K. Sundareshan,et al.  Training multilayer perceptron and radial basis function neural networks for wavefront sensing and restoration of turbulence-degraded imagery , 2004, 2004 IEEE International Joint Conference on Neural Networks (IEEE Cat. No.04CH37541).

[35]  S. Weddell,et al.  A Neural Network Architecture for Reconstruction of Turbulence Degraded Point Spread Functions , 2007 .

[36]  R. Noll.  Zernike polynomials and atmospheric turbulence , 1976 .

[37]  A. Quirrenbach,et al.  Adaptive Optics with Laser Guide Stars: Basic Concepts and Limitations , 2000 .

[38]  Francois Roddier,et al.  Adaptive Optics in Astronomy: Imaging through the atmosphere , 2004 .

[39]  J. Sprott.  Chaos and time-series analysis , 2001 .