The main focus of this project is to estimate the scenic beauty of forest images using independent component analysis and support vector machines, two of the most valuable tools for multigroup classification and data reduction. This project originated from a need of the United States Forest Service (USFS) to determine the scenic beauty of forests to preserve recreation and aesthetic resources in forest management. The results of this project will be useful to determine a predefined pattern to cut the trees so as to retain the scenic beauty even after cutting the forest for timber. The algorithms will be initially developed and tested in Matlab and then they will be developed in C++. The software developed will be tested on 637 images available in a standard evaluation database used to benchmark progress on this problem. Every image will be classified into one of the three classes: high scenic beauty class, medium scenic beauty class and the low scenic beauty class. Results obtained will be compared with the Scenic Beauty Estimation using human subjective ratings.

Vijay Ramani, Xinping Zhang, Zhiling Long, and Yu Zeng
Department of Electrical and Computer Engineering
Mississippi State University
email: {ramani, zhang, long, zeng}@cavs.msstate.edu
URL: http://www.cavs.msstate.edu/resources/courses/ece_4773/research/isip/projects/1998/group_image/presentation/