An approach for microscopy image restoration
Veska M. Georgieva, Plamen P. Petrov, Rumen Mironov, Antonia Mihaylova · 2019
The blind deconvolution algorithms are widely used in microscopy image restoration. However in more of the biological and biomedical experiments due to high level of noise, it is difficult to obtain good results. The blind deconvolution algorithm can be effectively applied when no information about the blurring and noise is given. In addition, a modified homomorphic filter based on wavelet decomposition is a frequently used instrument for noise reduction. In this paper, we propose an integrated approach that combines properties of blind deconvolution and modified homomorphic filter based on adaptive wavelet packet decomposition for noise reduction. As next, for contrast enhancement gamma correction is applied. We have made a quantitative analysis of the quality achieved by the proposed approach over deconvolution schemes, based on classical Richardson-Lucy algorithm and wavelet discrete transformation by experiments with real microscopy images. To prove our approach and theoretical statements results of laboratory experiments are suggested.