Skip to main content
ARS Home » Pacific West Area » Albany, California » Western Regional Research Center » Foodborne Toxin Detection and Prevention Research » Research » Publications at this Location » Publication #240163

Title: Detection of pits in fresh cherries

Author
item Haff, Ronald - Ron
item TSUTA, MIZUKI - National Food Research Institute - Japan

Submitted to: UJNR Food & Agricultural Panel Proceedings
Publication Type: Proceedings
Publication Acceptance Date: 6/30/2008
Publication Date: 8/24/2009
Citation: Haff, R.P., Tsuta, M. 2009. Detection of pits in fresh cherries. UJNR Food & Agricultural Panel Proceedings.

Interpretive Summary: X-ray imaging techniques that could be implemented for the detection of pits in cherries include linescan and real-time imaging using an image intensifier and CCD camera. This equipment is both expensive and bulky, and implementation on the processing line would be cumbersome. For this report, a simple one dimensional x-ray detector is tested for the ability to detect pits in fresh cherries. Acoustic impact analysis, a technique based on the difference in sounds generated as samples of different classes strike a solid surface, is also tested for the ability to detect pits in cherries. For the one dimensional x-ray detector, the results indicated almost perfect detection of pits with an average false positive rate of around 3%. For impact acoustics, a discriminant analysis routine was able to separate pitted from unpitted cherries in the validation set with 97% accuracy.

Technical Abstract: There are a number of x-ray imaging techniques that could be implemented for the detection of pits in cherries, including linescan and real-time imaging using an image intensifier and CCD camera. However, x-ray imaging equipment is both expensive and bulky, and implementation on the processing line would be cumbersome. Here, a simple one dimensional x-ray detector is tested for the ability to differentiate between pitted and unpitted fresh cherries. Acoustic impact analysis, a technique based on the difference in sounds generated as samples of different classes strike a solid surface, is also tested for the ability to detect pits in cherries. For the one dimensional x-ray detector, the results indicated almost perfect detection of pits with an average false positive rate of around 3%. For impact acoustics, a discriminant analysis routine was able to separate pitted from unpitted cherries in the validation set with 97% accuracy.