ARS | Publication request: ON LINE DETECTION OF FECAL CONTAMINANTS USING MULTISPECTRAL IMAGING SYSTEM

Submitted to: Meeting Abstract
Publication Type: Proceedings
Publication Acceptance Date: February 2, 2004
Publication Date: March 7, 2004
Citation: Park, B., Lawrence, K.C., Windham, W.R., Smith, D.P. 2003. On line detection of fecal contaminants using multispectral imaging system. [abstract] Proceedings 9th International Conference Engineering and Food. March 7-11, 2003. Montpellier, France.

Interpretive Summary: On-line detection of poultry carcasses contaminated by feces and/or ingesta is necessary to protect consumers from a potential source of food poisoning for food safety, in particular, the control of food borne pathogens. Development of high speed and reliable science-based inspection systems to ensure safe production of poultry processing is very important for Hazard Analysis, Critical Control Point (HACCP) compliance. ARS developed a real-time multispectral imaging system to detect fecal and ingesta contaminants on the surface of poultry carcasses. The system includes an industrial-scale prototype hardware and real-time image processing software for fecal detection. The system enables poultry producers to separate fecal and ingesta contaminated carcasses from non-contaminated carcasses at poultry processing plants. Multispectral imaging system that retains individual carcass inspection for on-line detection of fecal and ingesta contamination provides a science-based process control and possibly decrease water usage by cleaning contaminated carcasses only for the FSIS zero tolerance standards for fecal and/or ingesta on poultry carcass prior to the immersion chiller.

Technical Abstract: A common aperture multispectral imaging system with three optical trim filters (515.4, 566.4 and 631 nm), which were selected by visible/NIR spectroscopy and validated by a hyperspectral imaging system, was developed for a real-time, on-line poultry safety inspection application. In conjunction with an image ratio (566.4/515.4), a multispectral imaging system was an effective technique for the application of real-time detection of fecal and ingesta contaminants on the surface of broiler carcasses. The overall accuracies of contaminant detection were 96.8%, 94.4%, and 91.3% for corn, milo, and wheat with soybean mixed fed broiler carcasses, respectively. This system can be effectively used for detecting feces (from duodenum, ceca, and colon) and ingesta on the surface of poultry carcasses with a current industry processing speed at a rate of 140 birds per minute.