Maximum-likelihood estimation of an astronomical image from a sequence at low photon levels

摘要: We consider the problem of estimating one nonblurred and cleaned image from a sequence P randomly translated images corrupted with Poisson noise. develop new algorithm based on maximum-likelihood (ML) estimation for two unknown parameters: reconstructed itself set translations low-light-level images. demonstrate that ML is proportional to sum after correcting movement its entropy minimal. The are matched together by means an iterative minimum-entropy algorithm, where systematic search under displacements performed. fast version this we present results simulated experimental data. probability good matching low-level estimated numerically when light level in decreases, corresponding small numbers photons detected (down 20) each sequence. compare these those obtained known reference, i.e., linear correlation method, optimal one, noise has distribution. This approach applied astronomical acquired photocounting balloon-borne ultraviolet imaging telescope.