The purpose of this project is to
develop digital image processing algorithms that would assist conservation
biologists who study the location and behavior of snow leopards. Cameras placed
by researchers in remote areas inhabited by snow leopards take photographs when
a source of heat (i.e. snow leopard, fox, goat, or a large bird) passes in
front of it. The researchers then spend many hours recognizing specific cats
and their habitat. Because each snow leopard has a completely unique coat, snow
leopards are identified based on the characteristics of their spot patterns
such as their size, shape, orientation, and coloration. In this research, Dr.
Miguel will use techniques from image processing, pattern recognition, and
machine learning to aid researchers in their work with camera pictures. First,
she will develop algorithms to sort the images obtained from one camera. Once
the images are sorted, they will be analyzed by a program that searches for
matches among the many different spot patterns. Close match between two images
will indicate the same individual. The goal of the program will be to classify
each image as representing a particular snow leopard.