PARAMETRIC BLIND DECONVOLUTION OF MICROSCOPIC IMAGES

描述：（摘自摘要）计算光学切片显微镜

DESCRIPTION: (From the Abstract) Computational optical sectioning microscopy (COSM) is gaining acceptance as an alternative as well as a complement to confocal scanning microscopes (CSM) and two-photo fluorescence excitation microscopes (TPFEM). COSM algorithms remove out-of-focus light from wide-field microscope images and can also improve the resolution in CSMs and TPFEMs. The algorithms require knowledge of the process of image formation usually represented by the point-spread function (PSF), the image of a point source of light. In modern scanning microscopes, the measurement of computation of the PSF is difficult at best. In these cases, it is necessary to estimate the PSF and the specimen fluorescence distribution simultaneously from the image; an approach called blind deconvolution (BD). We developed a method for blind deconvolution in which the PSF is constrained to follow a mathematical model that depends on a small number of unknown parameters. The parametric BD (PBD) method determines these parameters together with the specimen function. We have found encouraging results but the PBD is still computationally intensive. The purpose of this research is to9 make this method practical for routine use and to extend its application to CSMs and TOFEMS. The method will be optimized and its capabilities and limitations thoroughly assessed to be able to provide guidelines for its use. In addition, the method will be incorporated into a deconvolution package that our group developed.