Application of a Noise-Smoothing Filter Based on Adaptive Windowing to Penumbral Imaging

摘要

Penumbral imaging is a technique which uses the facts that spatialinformation can be recovered from the shadow or penumbra that anunknown source casts through a simple large circular aperture. Thetechnique is based on a linear deconvolu- tion. In this paper, atwo-step method is proposed for decoding penumbral images. First alocal-statistic filter based on adap- tive windowing is applied tosmooth the noise; then, followed by the conventional lineardeconvolution. The simulation results show that the reconstructedimage is dramatically improved in comparison to that without thenoise-smoothing filtering, and the proposed method is also applied toreal experimetal x-ray imaging.