Alexander Shyrokov

MS Thesis

LIBRARY FOR ORGANIZATION OF IMAGE RECOGNITION SYSTEMS
by
Alexander Shyrokov
under supervision of Dr. Messner
University of New Hampshire, September, 2002

Abstract

In many recognition problems it is possible to divide the recognition task into a collection of separate processes. This thesis develops a method that simplifies the creation of automatic recognition systems. Formal separation of different algorithm steps was performed and an interface for communicating between each step was developed. Inorder to evaluate the robustness of the proposed method, the algorithms developed by Mr. Tony Pawlak (M.S. UNH ECE 1998) in his thesis work on automatic human brain cell recognition were implemented. The original human brain cell recognition research was done using Matlab software. After the original recognition process was ported into the proposed standard, improvements were made to demonstrate how the designed system could be extended and modified. To demonstrate how a "new" recognition task could be implemented a different recognition problem was chosen: the counting of bacteria in a pure culture. Documented examples show that the proposed method and standards simplify the organization, execution, and maintenance of the recognition procedures. Developed in this thesis is a ready to use library for various algorithms with examples and documentation as well as a full discussion on how researchers can develop their own modules for inclusion.Complete documentation on standards and interfacing is included in appendices.

Disclaimer

Copyright (c) 2002 Alexander Shyrokov

Permission to use, copy, modify and distribute the contests of this web-site for non-commercial purpose is hereby granted without fee, provided that (i) the above copyright notices and this permission notice appear inall copies of the software and related documentation, and (ii) the name of Alexander Shyrokov may not be used in any advertising or publicity relating to the software without the specific, prior written permission of Alexander Shyrokov.

For commercial use of the source code or parts of the source code for this thesis,written permission of Alexander Shyrokov is required!

THE SOFTWARE IS PROVIDED "AS-IS" AND WITHOUT WARRANTY OF ANY KIND, EXPRESS, IMPLIED OR OTHERWISE, INCLUDING WITHOUT LIMITATION, ANY WARRANTY OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE.

IN NO EVENT SHALL ALEXANDER SHYROKOV BE LIABLE FOR ANY SPECIAL, INCIDENTAL, INDIRECT OR CONSEQUENTIAL DAMAGES OF ANY KIND,OR ANY DAMAGES WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS,WHETHER OR NOT ADVISED OF THE POSSIBILITY OF DAMAGE, AND ON ANY THEORY OF LIABILITY, ARISING OUT OF OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.

Downloads

Thesis pdf file (2Mb) Mirror
Presentation pdf file (1.5Mb) Mirror
Binaries (Win32, 6Mb) Mirror
Source code (STLport, TIFF lib and samples are not included) (Win32, 1Mb) Mirror
Samples (2.5Mb) Mirror
STLport 4.5 sources (1Mb) Mirror
TIFF lib sources (1Mb) Mirror

Screen shot

Other links

University of New Hampshire
The Synthetic Vision and Pattern Analysis Laboratory
Department of Computer and Electrical Engineering