Several methods have been proposed for guiding pedestrians by generating guide sentences from navigation routes.
These methods use landmarks to explain the next point to which the users should go.
However, in these cases specifying the landmarks is difficult because they use only names as guide annotations without any exterior form of identification. Hattori proposed a utility that uses
exterior identification and a method of collecting them using
web mining of documents on the Web and in blogs. The
proposed system collects information about landmarks and
their exterior via user-based structure information. Fukui
[13] proposed a navigation system for pedestrians that utilizes
a portable telephone and scene images. Persons who
know the areas well make specific navigation from the
nearest station to the destination using guide sentences and
scene images. Further, they can reuse same information
others used to make the specific navigation. Hence, it is
possible to furnish appropriate guides. Further, the results
of experiments conducted prove that the system is effective.
However, the proposed system appears to be only able to
use certain specific routes. In contrast, our proposed system
makes it possible to navigate any route because users do not
register information about specific routes, but about points.