A multispectral imaging system using solar illumination to distinguish faecal matter on leafy greens and soils. (July 2018)

• Record Type: Journal Article
• Authors: Everard, Colm D.; Kim, Moon S.; Siemens, Mark C.; Cho, Hyunjeong; Lefcourt, Alan; O'Donnell, Colm P.
• Is Part Of: Biosystems engineering. Volume 171(2018)
• Page Start: 258
• Page End: 264
• Publication Date: 2018-07
• Subjects: Multispectral imaging -- Faecal contamination -- Leafy greens -- Soil -- In-field -- Food safety; Bioengineering -- Periodicals; Agricultural engineering -- Periodicals; Biological systems -- Periodicals; Génie rural -- Périodiques; Systèmes biologiques -- Périodiques; 631
• DOI: 10.1016/j.biosystemseng.2018.05.001
• Languages: English
• ISSNs: 1537-5110

Abstract

Abstract : Faecal contaminated fruits and vegetables have been linked to several outbreaks of foodbourne diseases. In-field detection of faecal matter would allow the producer to take action to reduce faecal contaminated produce entering the post-harvest processing line and the human food supply. No viable systems to accomplish this task have been developed to date. To address this, a prototype imaging system was developed to detect faecal matter on leafy greens. The system principally comprised of two monochrome cameras that were used to simultaneously capture images of the same target at two separate wavelengths, 690 and 710 nm, by utilising a beam-splitter and bandpass filters. The 710 nm and 690 nm waveband images were used for pixel-by-pixel calculation of a ratio image of the target, to which a thresholding technique was applied to classify pixels as either faecal matter or non-faecal matter. The system was tested on spinach leaves ( Spinacia olerace ) and three types of soils in an outdoor environment. On all samples evaluated, the imaging system, coupled with this waveband ratio normalisation method, successfully distinguished faecal matter from spinach leaves and soil under varying atmospheric conditions. These findings are very encouraging and further study is needed to determine if such a technique would reliably detect faecal material in an agricultural field environment where leaf orientation and contamination concentration levels are highly variable. Highlights: Prototype imaging system developed to detect faecal matter on leafy greens and soil. The system captured images at 690 and 710 nm using monochromatic cameras. Developed image algorithm created a single image from the ratio of 710:690 nm. Developed image algorithm successfully identified faecal contaminated locations.