Abstract
We use Dataflow Engines (DFE) to construct an efficient Wiener filter of noisy and incomplete image data, and to quickly draw probabilistic samples of the compatible true underlying images from the Wiener posterior. Dataflow computing is a powerful approach using reconfigurable hardware, which can be deeply pipelined and is intrinsically parallel. The unique Wiener-filtered image is the minimum-variance linear estimate of the true image (if the signal and noise covariances are known) and the most probable true image (if the signal and noise are Gaussian distributed). However, many images are compatible with the data with different probabilities, given by the analytic posterior probability distribution referred to as the Wiener posterior. The DFE code also draws large numbers of samples of true images from this posterior, which allows for further statistical analysis. Naive computation of the Wiener-filtered image is impractical for large datasets, as it scales as $n^3$, where $n$ is the number of pixels. We use a messenger field algorithm, which is well suited to a DFE implementation, to draw samples from the Wiener posterior, that is, with the correct probability we draw samples of noiseless images that are compatible with the observed noisy image. The Wiener-filtered image can be obtained by a trivial modification of the algorithm. We demonstrate a lower bound on the speed-up, from drawing 10$^5$ samples of a 128$^2$ image, of 11.3 ${\pm}$ 0.8 with 8 DFEs in a 1U MPC-X box when compared with a 1U server presenting 32 CPU threads. We also discuss a potential application in astronomy, to provide better dark matter maps and improved determination of the parameters of the Universe.
Highlights
Dataflow computing has recently aided the significant acceleration of many computationally-intensive and data-intensive problems
This paper discusses the use of Dataflow Engines (DFEs) for sampling realisations of noise-free images from the Wiener posterior distribution given noisy and incomplete data, with particular applicability to astronomy and cosmology
Jeffrey et al / Astronomy and Computing 25 (2018) 230–237 repeated operations inherent in drawing samples from the Wiener posterior lend themselves to efficient computation on DFEs, and we demonstrate that by a comparison with an implementation on multiple Central Processing Unit (CPU)
Summary
Dataflow computing has recently aided the significant acceleration of many computationally-intensive and data-intensive problems. This paper discusses the use of Dataflow Engines (DFEs) for sampling realisations of noise-free images from the Wiener posterior distribution given noisy and incomplete data, with particular applicability to astronomy and cosmology. The Wiener filter (Wiener, 1949) is a useful statistical tool in many image analyses, as it is a minimum variance linear filter,. If the covariance matrices of the noise and signal are known, the Wiener filtered image has the smallest variance of any linear-filtered image. It is straightforward to write down the expression for the Wiener-filtered image, and the covariance of compatible images, but evaluation is problematic as it involves the inversion of large matrices that are in general nondiagonal. Jeffrey et al / Astronomy and Computing 25 (2018) 230–237 repeated operations inherent in drawing samples from the Wiener posterior lend themselves to efficient computation on DFEs, and we demonstrate that by a comparison with an implementation on multiple CPUs
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.