Human-experts rules modeling for linear planimetric features extraction in a remotely sensed images data fusion context

This study deals with the use of knowledge engineering techniques applied to the design of linear planimetric features (LPF) extraction and classification tools. These features include the most important cartographic elements like roads, energy lines, railroads, etc. Since human knowledge can be classified into different categories such as declarative, procedural and meta-knowledge, the research work presented in this study is related to a part of the procedural knowledge known as rules. These rules presented in the case of the LPF detection are not only essential in the development of semi-automatic general cartographic systems but they also put highlights on inexact and fuzzy reasoning which are powerful tools used in intelligent systems development.