Nagao and Mastuyama have developed a knowledge based
system which performs a structural analysis of complex aerial
photographs using a technique called segmentation-byrecognition.
The flow of the system starts from segmentation
of the air-photo image to get regions, then generate property
tables for regions, and finally use knowledge based rules to
identify each region. The system uses knowledge about the
0-7803-7536-X/$17.00 (C) 2002 IEEE 1340
intrinsic properties of objects such as size, shape, location,
color, and texture. For example, the rule to detect forest and
grassland is as follows: If a region is high contrast texture
area and large vegetation area and large homogeneous region
and non water region, then the area belongs to the forest or
grassland category. This rule will work only when a small
number of land cover types are considered which are easily
distinguished. Gold berg et.al have designed a hierarchical
expert system for updating forest maps in British Columbia,
Canada. The problem addressed is the automatic estimation
forest depletion by logging. The top level of the hierarchy is
an expert with broad knowledge about updating forest maps