Date of Award

8-1-2011

Degree Name

Master of Science

Department

Computer Science

First Advisor

Mogharreban, Namdar

Abstract

Traditional fruit categorizations by humans in agricultural settings are inefficient, labor intensive and prone to errors. Automated grading systems not only speed up the time of the process but also minimize error. In this work, we propose and implement methodologies and algorithms that utilize digital image processing, content predicated analysis, and statistical analysis to detect, fuzzify, and classify the fruit dates. Our main contribution is design and development of an efficient algorithm for detecting and sorting dates. The system was accurate 85% of the times when compared to a human expert sorting. A larger sample size could help with tweaking the fuzzy system and improve the success rate. The proposed system is flexible and with minor changes can be adapted to other produce such as apples.

Share

COinS
 

Access

This thesis is only available for download to the SIUC community. Others should
contact the interlibrary loan department of your local library.