Page Banner

United States Department of Agriculture

Agricultural Research Service

Research Project: SORTING AGRICULTURAL MATERIALS FOR DEFECTS USING IMAGING AND PHYSICAL METHODS Title: X-Ray Detection of and Sorting of Olives Damaged by Fruit Fly

Authors
item Jackson, Eric
item Haff, Ronald

Submitted to: Transactions of the ASABE
Publication Type: Proceedings
Publication Acceptance Date: May 16, 2006
Publication Date: July 31, 2006
Citation: Jackson, E.S., Haff, R.P. 2006. X-ray detection of and sorting of olives damaged by fruit fly. Transactions of the ASABE. ASABE Paper #06-6062.

Interpretive Summary: A computer program was created to look at x-ray images and detect the ones infested with fruit flies. The data set consisted of 249 olives with various degrees of infestation and 161 non-infested olives. Each olive was x-rayed on film and digital images were acquired with a film scanner at a resolution of 59 pixels per cm. Information from the images was fed into the computer program to see how well it worked. The information taken from the images involved the pixel brightness at different points in the image and how quickly the brightness changed. Internal damage to the olive was a factor in detection, with slight damage correctly identified 50% of the time and severe damage correctly identified 86% of the time. Non-infested olives were correctly identified with 90% accuracy.

Technical Abstract: An algorithm using a Bayesian classifier was developed to automatically detect olive fruit fly infestations in x-ray images of olives. The data set consisted of 249 olives with various degrees of infestation and 161 non-infested olives. Each olive was x-rayed on film and digital images were acquired with a film scanner at a resolution of 59 pixels per cm. Features extracted from the images were submitted to the classification algorithm and error rates for detection of the infestations obtained. Feature selection involved pixel intensity values and pixel derivative values at each pixel location in the image. The ability of the algorithm to differentiate infested and non-infested olives was tested. Internal damage to the olive was a factor in detection, with slight damage correctly identified 50% of the time and severe damage correctly identified 86% of the time. Non-infested olives were correctly identified with 90% accuracy.

Last Modified: 10/30/2014
Footer Content Back to Top of Page