Date of Award

5-1986

Degree Type

Report

Degree Name

Master of Landscape Architecture (MLA)

Department

Landscape Architecture and Environmental Planning

Committee Chair(s)

John Nicholson

Committee

John Nicholson

Committee

Larry Wegkamp

Committee

Larre Egbert

Abstract

The purpose of this investigation was to explore the depiction of trees in three dimensions on a microcomputer. While the use of computer-aided design in landscape architecture is increasing, imagery for plant materials is found to be at a more or less symbolic level. The literature concerning previous inquiries into the mechanisms of tree growth and differentiation provide a good deal of information ranging from physiological basics to sophisticated structural and mathematical growth models. This forms the basis from which programming work proceeded.

In this context, the body of work reported here emphasizes the development of a programming methodology for achieving better tree images, rather than the sophistication of the images them-selves. A major goal in this effort was simplicity in the resulting algorithms. This is significant in both minimizing use of computer memory, and in aiding the transfer of the algorithms to other devices and uses. Discussed are the developmental steps taken from an initial tree model requiring a digitizing tablet and the internal storage of coordinates, to a tree model in which machine memory and algorithm complexity are minimized.

The methodology deemed most useful is that of storing the tr:es as a general set of rules for image generation, rather than a lengthy data file for each tree. The operational value of this process is intrinsic to future applications; whether six discrete tree types are to be used or sixty types, the computer is working vii with the same amount of "data" -- the tree generation algorithm. Further applications of this approach could offer savings in both storage requirements and data input for a variety of complex graphic images.

Share

COinS