Forest management on public and private land is under intense public
scrutiny. Public concern for the aesthetic qualities of forests has
yielded a growing demand for conserving natural forest environments.
Legislation to maintain the scenic beauty of the forests is even
part of United States Department of Agriculture, Forest Service
(USFS) land management planning. Traditional research methods to
determine scenic quality involve time-consuming judgments by a large
group of people. Our goal is to develop intelligent algorithms to
automate these processes. We are using digitized images of a range
of forest stand treatments from color slide photographs. Ongoing
research extracts digital metrics such as color, line edge detection,
image segmentation, and object recognition. Modeling involves
calibration of these measures with estimates of scenic beauty and
other psychological judgments.
|