基于信息结构分析的图象重建理论的研究

In this research, image is expressed into the combination of information gene related image construction characteristic according to information construction hypothesis in information theories, established the image's construction model, then removed the construction relativity of image pixel values, and reduce the complete demand of projection data in projection imaging. So, an accurate image can be reconstructed by extracting the complete information gene and gaining model parameter from incompletely projection data. We have found out the sufficient and essential condition in reconstructing image from incompletely projection and solved the artifacts problem in imaging from incomplete projection. This method can raise the CT image's resolution response in time and space and extend the CT application in industry and earthquake checking.

本研究根据信息论中信息结构理论,以图象结构特征相关的信息基因组合表达图象,建立图象的结构模型,获得结构模型参数方法准确重建图象.这一研究的成功,必将开创CT成象的新局?有重大学术价值和应用前景.