Author
Grover, Henry
Other Contributors
Stewart, Charles V.; Goldschmidt, David E.; Zaki, Mohammed J., 1971-;
Date Issued
2020-05
Subject
Computer science
Degree
MS;
Terms of Use
This electronic version is a licensed copy owned by Rensselaer Polytechnic Institute, Troy, NY. Copyright of original work retained by author.;
Abstract
This thesis addresses the issue of animal orientation estimation and its effect to increase accuracy of identification algorithms. Sets of animal images that have similarly oriented subjects have the potential to result in a higher accuracy with identification algorithms. The problem of identification is to decide if the animal of interest is a previously seen individual or a new animal not seen before in previous images. The main focus is to discuss the most effective ways in which to estimate the canonical rotation of an image based on the area of interest, as well as to analyze the effect of consistent animal orientation within the identification problem. This thesis introduces and analyzes three algorithms to predict the orientation of animal images then compares the effect of oriented and non-oriented image sets on identification. Finally it introduces simple animal fiducial marker location estimation for future use in identification.;
Description
May 2020; School of Science
Department
Dept. of Computer Science;
Publisher
Rensselaer Polytechnic Institute, Troy, NY
Relationships
Rensselaer Theses and Dissertations Online Collection;
Access
Restricted to current Rensselaer faculty, staff and students. Access inquiries may be directed to the Rensselaer Libraries.;