Skip to main content
ARS Home » Southeast Area » Raleigh, North Carolina » Market Quality and Handling Research » Research » Publications at this Location » Publication #117496
Title: COMPUTER ASSISTED COLOR CLASSIFICATION OF PEANUT PODS

Author
item BOLDOR, DORIN
item Sanders, Timothy
item SWARTZEL, KENNETH
item SIMUNOVIC, JOSIP

Submitted to: Peanut Science
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 11/22/2000
Publication Date: 1/25/2003
Citation: BOLDOR, D., SANDERS, T.H., SWARTZEL, K.R., SIMUNOVIC, J. COMPUTER ASSISTED COLOR CLASSIFICATION OF PEANUT PODS. PEANUT SCIENCE. 2003. V. 29. P. 41-46.

Interpretive Summary: Current methodology for peanut harvest date determination is subjective, time-consuming, and costly because of the personnel required. Upgraded methods need to be developed. In cooperation with the Department of Food Science, North Carolina State University, a computer-assisted color vision system for peanut hull scrape colors was developed. Use of the system will improve peanut harvest date prediction in a manner which is rapid, less expensive, objective and more consistent.

Technical Abstract: Mesocarp color classification in the Hull Scrape Maturity method is the most important step in determining peanut maturity and optimum harvest date. This research involved the development of an image acquisition system and a software procedure for color classification. Images of peanut pods which had been manually sorted into color classes and subclasses were used in computer training. After training with assorted color classes, the computer-assisted procedure correctly classified and identified the peanut pods with a 99% precision for the same sample in different alignment, and for a 95% precision for different size samples taken from the same population. Pod size measurement with a resolution of 0.1 mm was also performed using image processing techniques.