Fluorescence hyper-spectral imaging to detecting faecal contamination on fresh tomatoes

Submitted: 6 July 2015
Accepted: 24 January 2016
Published: 8 March 2016
Abstract Views: 1728
PDF: 665
HTML: 822
Publisher's note
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.

Authors

Faecal contamination of fresh fruits represents a severe danger for human health. Thus some techniques based on microbiological testing were developed to individuate faecal contaminants but those tests do not results efficient because their non-applicability on overall vegetable unity. In this work a methodology based on hyper-spectral fluorescence imaging was developed and tested to detecting faecal contamination on fresh tomatoes. Two image-processing methods were performed to maximise the contrast between the faecal contaminant and tomatoes skin: principal component analysis and band image ratio (BRI). The BRI method allows classifying correctly 70% of contaminated area, with no false-positives in all examined cases. Thus, the developed methodology can be employed for a fast and effective detection of faecal contamination on fresh tomatoes.

Dimensions

Altmetric

PlumX Metrics

Downloads

Download data is not yet available.

Citations

How to Cite

Romaniello, R., Peri, G. and Leone, A. (2016) “Fluorescence hyper-spectral imaging to detecting faecal contamination on fresh tomatoes”, Journal of Agricultural Engineering, 47(1), pp. 7–11. doi: 10.4081/jae.2016.491.