Astronomical image reconstruction with convolutional neural networks. Flamary, R. 2017 25th European Signal Processing Conference (EUSIPCO), 2016. doi abstract bibtex State of the art methods in astronomical image reconstruction rely on the resolution of a regularized or constrained optimization problem. Solving this problem can be computationally intensive and usually leads to a quadratic or at least superlinear complexity w.r.t. the number of pixels in the image. We investigate in this work the use of convolutional neural networks for image reconstruction in astronomy. With neural networks, the computationally intensive tasks is the training step, but the prediction step has a fixed complexity per pixel, i.e. a linear complexity. Numerical experiments show that our approach is both computationally efficient and competitive with other state of the art methods in addition to being interpretable.
@article{flamary_astronomical_2016,
title = {Astronomical image reconstruction with convolutional neural networks},
doi = {10.23919/eusipco.2017.8081654},
abstract = {State of the art methods in astronomical image reconstruction rely on the resolution of a regularized or constrained optimization problem. Solving this problem can be computationally intensive and usually leads to a quadratic or at least superlinear complexity w.r.t. the number of pixels in the image. We investigate in this work the use of convolutional neural networks for image reconstruction in astronomy. With neural networks, the computationally intensive tasks is the training step, but the prediction step has a fixed complexity per pixel, i.e. a linear complexity. Numerical experiments show that our approach is both computationally efficient and competitive with other state of the art methods in addition to being interpretable.},
journal = {2017 25th European Signal Processing Conference (EUSIPCO)},
author = {Flamary, Rémi},
year = {2016},
keywords = {Algorithmic efficiency, Artificial neural network, Astrophysics - Instrumentation and Methods for Astrophysics, Computer Science - Computer Vision and Pattern Recognition, Constrained optimization, Constraint (mathematics), Convolution, Convolutional neural network, Experiment, Image resolution, Iterative reconstruction, Mathematical optimization, Numerical analysis, Optimization problem, Performance, Pixel, Quadratic function, Simulation, Statistics - Machine Learning},
pages = {2468--2472},
}
Downloads: 0
{"_id":"mfEcJpKwtq7CcgJvH","bibbaseid":"flamary-astronomicalimagereconstructionwithconvolutionalneuralnetworks-2016","downloads":0,"creationDate":"2019-01-15T11:41:42.135Z","title":"Astronomical image reconstruction with convolutional neural networks","author_short":["Flamary, R."],"year":2016,"bibtype":"article","biburl":"https://bibbase.org/zotero/polyphant","bibdata":{"bibtype":"article","type":"article","title":"Astronomical image reconstruction with convolutional neural networks","doi":"10.23919/eusipco.2017.8081654","abstract":"State of the art methods in astronomical image reconstruction rely on the resolution of a regularized or constrained optimization problem. Solving this problem can be computationally intensive and usually leads to a quadratic or at least superlinear complexity w.r.t. the number of pixels in the image. We investigate in this work the use of convolutional neural networks for image reconstruction in astronomy. With neural networks, the computationally intensive tasks is the training step, but the prediction step has a fixed complexity per pixel, i.e. a linear complexity. Numerical experiments show that our approach is both computationally efficient and competitive with other state of the art methods in addition to being interpretable.","journal":"2017 25th European Signal Processing Conference (EUSIPCO)","author":[{"propositions":[],"lastnames":["Flamary"],"firstnames":["Rémi"],"suffixes":[]}],"year":"2016","keywords":"Algorithmic efficiency, Artificial neural network, Astrophysics - Instrumentation and Methods for Astrophysics, Computer Science - Computer Vision and Pattern Recognition, Constrained optimization, Constraint (mathematics), Convolution, Convolutional neural network, Experiment, Image resolution, Iterative reconstruction, Mathematical optimization, Numerical analysis, Optimization problem, Performance, Pixel, Quadratic function, Simulation, Statistics - Machine Learning","pages":"2468–2472","bibtex":"@article{flamary_astronomical_2016,\n\ttitle = {Astronomical image reconstruction with convolutional neural networks},\n\tdoi = {10.23919/eusipco.2017.8081654},\n\tabstract = {State of the art methods in astronomical image reconstruction rely on the resolution of a regularized or constrained optimization problem. Solving this problem can be computationally intensive and usually leads to a quadratic or at least superlinear complexity w.r.t. the number of pixels in the image. We investigate in this work the use of convolutional neural networks for image reconstruction in astronomy. With neural networks, the computationally intensive tasks is the training step, but the prediction step has a fixed complexity per pixel, i.e. a linear complexity. Numerical experiments show that our approach is both computationally efficient and competitive with other state of the art methods in addition to being interpretable.},\n\tjournal = {2017 25th European Signal Processing Conference (EUSIPCO)},\n\tauthor = {Flamary, Rémi},\n\tyear = {2016},\n\tkeywords = {Algorithmic efficiency, Artificial neural network, Astrophysics - Instrumentation and Methods for Astrophysics, Computer Science - Computer Vision and Pattern Recognition, Constrained optimization, Constraint (mathematics), Convolution, Convolutional neural network, Experiment, Image resolution, Iterative reconstruction, Mathematical optimization, Numerical analysis, Optimization problem, Performance, Pixel, Quadratic function, Simulation, Statistics - Machine Learning},\n\tpages = {2468--2472},\n}\n\n","author_short":["Flamary, R."],"key":"flamary_astronomical_2016","id":"flamary_astronomical_2016","bibbaseid":"flamary-astronomicalimagereconstructionwithconvolutionalneuralnetworks-2016","role":"author","urls":{},"keyword":["Algorithmic efficiency","Artificial neural network","Astrophysics - Instrumentation and Methods for Astrophysics","Computer Science - Computer Vision and Pattern Recognition","Constrained optimization","Constraint (mathematics)","Convolution","Convolutional neural network","Experiment","Image resolution","Iterative reconstruction","Mathematical optimization","Numerical analysis","Optimization problem","Performance","Pixel","Quadratic function","Simulation","Statistics - Machine Learning"],"metadata":{"authorlinks":{}}},"search_terms":["astronomical","image","reconstruction","convolutional","neural","networks","flamary"],"keywords":["algorithmic efficiency","artificial neural network","astrophysics - instrumentation and methods for astrophysics","computer science - computer vision and pattern recognition","constrained optimization","constraint (mathematics)","convolution","convolutional neural network","experiment","image resolution","iterative reconstruction","mathematical optimization","numerical analysis","optimization problem","performance","pixel","quadratic function","simulation","statistics - machine learning"],"authorIDs":[],"dataSources":["ECdoAcmvZ8tYirN8u","7gvjSdWrEu7z5vjjj"]}