Skip to main content
ARS Home » Plains Area » Las Cruces, New Mexico » Cotton Ginning Research » Research » Publications at this Location » Publication #74203

Title: GRAVIMETRIC BARK CONTENT USING MACHINE VISION

Author
item Lieberman, Michael
item Bragg, Charles
item BRENNAN, SEAN

Submitted to: Textile Research Journal
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 2/11/1997
Publication Date: N/A
Citation: N/A

Interpretive Summary: A method is needed to accurately and rapidly determine the gravimetric bark content of a cotton sample. Gravimetric bark content represents the bark mass percent throughout the volume of a cotton sample. Measuring gravimetric bark content is a labor-intensive, lengthy process. Using different bark definitions, the best predicted GBC measurements were shown to correlate with manually measured GBC with the best r-squared of 0.76 Machine vision is proposed as a fast, inexpensive method to perform this measurement. In this study, machine vision was used to predict this bulk cotton property. Ten images were acquired of surfaces throughout each sample. Classical digital image processing techniques were used on monochrome video images to isolate foreign-matter regions. Geometric properties (area and perimeter) were used to identify which foreign matter was bark and to predict the gravimetric bark content (GBC) in 48 cotton samples with varying bark and total foreign-matter content.

Technical Abstract: A method is needed to accurately and rapidly determine the gravimetric bark content of a cotton sample. Gravimetric bark content represents the bark mass percent throughout the volume of a cotton sample. Measuring gravimetric bark content is a labor-intensive, lengthy process. Machine vision is proposed as a fast, inexpensive method to perform this measurement. In this study, machine vision was used to predict this bulk cotton property. Ten images were acquired of surfaces throughout each sample. Classical digital image processing techniques were used on monochrome video images to isolate foreign-matter regions. Geometric properties (area and perimeter) were used to identify which foreign matter was bark and to predict the gravimetric bark content (GBC) in 48 cotton samples with varying bark and total foreign-matter content. Using different bark definitions, the best predicted GBC measurements were shown to correlate with manually measured GBC with the best r-squared of 0.76.