Doctoral Dissertations

Author

Luo Cheng

Date of Award

12-1997

Degree Type

Dissertation

Degree Name

Doctor of Philosophy

Major

Human Ecology

Major Professor

Kermit E. Duckett

Committee Members

Randall R. Bresee, Gajanan Bhat, Mongi A. Abidi

Abstract

Wool is an important textile fiber. The price of the wool is largely depending on fiber diameter, with length, vegetable matter and dark fiber content next. Numerous of individual instruments were developed to measure wool fiber properties but until now, there is no integrated fiber information system to measure major fiber properties. Therefore, it is desirable to develop an integrated system to measure wool fiber diameter, length and defects as HVI or AFIS used in cotton textile industry. The feasibility of this approach is discussed in this research project.

Using the latest technologies in digital signal and image processing, a research prototype instrument was developed. The instrument has an innovational design of a fiber individualizer based on drafting mechanism. Neural networks, fuzzy logic and genetic algorithms were applied to the waveform recognizer and defects classifier of the instrument and the results are discussed. Evaluated using a special set of wool samples, the instrument show promising results comparing with conventional individual instruments and human manual testing/inspection.

Files over 3MB may be slow to open. For best results, right-click and select "save as..."

Share

COinS