Modelling rain patterns: towards automatic interpretation of radar images

作者:

Highlights:

摘要

A method for the formalization, development and validation of image interpretation procedures through the design, implementation and use of suitable software tools is proposed and illustrated for the case of radar-derived images of rain. The lack of a well defined image model in this and many other experiments forces the human interpreter to rely on heuristic reasoning. The design of corresponding automatic procedures has, in the first place, to face the formalization of rules which govern such interpretation activity. To this end, the proposed method combines the use of attributed structural tools to derive rich image description, with multivalued logic tools to map the set of hints collected in the descriptions into plausible judgements about their physical meaning. These tools express the rules which control the Automatic Assistant a software system which serves to develop and validate the proposed interpretation procedure.

论文关键词:interpretation procedure,Automatic Assistant,attributed description,multivalued logic

论文评审过程:Available online 10 June 2003.

论文官网地址:https://doi.org/10.1016/0262-8856(90)90066-E