期刊名称：ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences
印刷版ISSN：2194-9042
电子版ISSN：2194-9050
出版年度：1992
卷号：XXIX Part B3
页码：488-496
出版社：Copernicus Publications
摘要：Image understanding is the enterprise of automatingand integrating a wide range of processes and representationsused for vision perception. It includes techniquesnot only for geometric modeling but also forinference and reasoning. In this paper, we look at theissue of inductive inference in digital image understanding.Many goals in both low-level and high-levelimage analysis can be formulated generally as a problemof inferring object-properties from image data,having assistance of a priori knowledge. This processof information processing would be called image inversion,as the desired information about the objectsis derived from image dat.a. Based on the inverse problemtheory, we provide a sound theoretical basis fordetermination of generalizations, descriptions, rules,and laws, from a set of raw data, observations, featuresor facts. To demonstrate this approach, we presentits application in the limited domain of surface reconstructionfrom multiple images.
