|Yang, Chun Chieh|
|KANG, SUKWON - Rural Development Administration - Korea|
|TAO, TAO - University Of Maryland|
|Chao, Kuanglin - Kevin Chao|
Submitted to: Sensing and Instrumentation for Food Quality and Safety
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 11/24/2010
Publication Date: 12/14/2010
Citation: Yang, C., Kim, M.S., Kang, S., Tao, T., Chao, K., Lefcourt, A.M., Chan, D.E. 2010. Development of a Simple Multispectral Algorithm Using a Hyperspectral Line-Scan Imaging System for Detection of Fecal Contamination on Apples. Sensing and Instrumentation for Food Quality and Safety. 5:10-18.
Interpretive Summary: To help address growing food safety concerns about contamination of fresh food products by pathogens associated with animal fecal contamination, EMFSL researchers are developing image-based machine vision technologies suitable for monitoring and inspection of fruits and vegetables on high speed processing lines. The EMFSL hyperspectral imaging system can scan the surface of foods at speeds sufficient to match high-speed automated processing lines typically used by commercial processors. Broad-spectrum hyperspectral images of feces-contaminated apples were analyzed to select a few optimal wavebands from among all the available wavebands in the hyperspectral data. The goal of this research is to develop a simple multispectral algorithm for detection of fecal contamination on apple surfaces, derived from hyperspectral image analysis of fluorescence emissions produced by violet LED excitation. A multispectral algorithm based on the selected essential wavebands was developed that can be implemented by an online machine vision system for rapid real-time inspection of fresh produce. In the results of this study, the multispectral-based imaging algorithm demonstrated an average of 98% success rate for contamination detection for a range of fecal dilutions applied to the apple surfaces. The developed multispectral detection algorithm can be used as a food safety inspection tool to be used by the food processing industry and regulatory agencies to minimize food safety risks for the general public, and the methodology used in this work provides research helpful to food process scientists and engineers .
Technical Abstract: Foodborne diseases are of serious concern for public health. It is necessary to develop fast and reliable non-destructive detection methods to improve food product monitoring for the food industry. This research was conducted to investigate hyperspectral fluorescence imaging using violet/blue LED excitation to develop a multispectral algorithm for detection of fecal contamination on Golden Delicious apples. From the hyperspectral image data, four wavebands, 680 nm, 684 nm, 720 nm, and 780 nm, were selected for potential use in a multispectral detection algorithm. The algorithm could detect 96% to 100% of different dilutions of feces on apples. The highly successfully detection of feces showed that a simple multispectral fluorescence imaging algorithm based on violet/blue LED excitation can be used to detect fecal contamination on apple processing lines.