'Seeing' Poultry Carcass Contamination

Agricultural Research Service scientists have modified a hyperspectral imaging system used to scan the surface of poultry carcasses so it can detect very low levels of faecal contamination.

12 October 2009

3 minute read

Agricultural Research Service scientists Kurt Lawrence, Bosoon Park, Bob Windham, and Seung-Chul Yoon – all in the Quality and Safety Assessment Research Unit in Athens, Georgia – have made two improvements to a hyperspectral imaging system used to scan the surface of poultry carcasses for contaminants. They have refined the system so that it can detect even tiny amounts of fecal contamination, which can vary significantly depending on where in the digestive tract it originated. They have also developed and implemented a new image-processing method to identify and remove false-positive readings.

To facilitate the transfer of their system, a prototype on-line multispectral imaging system was installed and tested in a commercial poultry plant to detect fecal-contaminated carcasses. The system was developed through a research agreement with Stork Gamco, a manufacturer of poultry processing equipment based in Gainesville, Georgia. Carcasses were imaged after evisceration but before washing, at a rate of 150 birds per minute.

“The system ran for several days with no hardware or software problems, and it demonstrated the feasibility of accurately detecting fecal-contaminated carcasses,” says Mr Lawrence.

The Athens team is collaborating with agricultural engineer Kevin Chao and biophysical scientist Moon Kim at the Beltsville Agricultural Research Center’s Environmental Microbial and Food Safety Laboratory, who developed an on-line imaging system to differentiate systemically diseased poultry carcasses from wholesome ones.

The ARS groups and their industry partner are now merging the fecal-detection and diseased-carcass-detection systems onto a common platform that includes a line-scan hyperspectral imaging camera, lighting, and operating and detection software. Merging the two systems will aid in commercialization by creating one interchangeable imaging system that can be installed in different locations of the processing line to solve two separate and significant problems in the poultry processing industry. This will allow processors to more easily integrate such a system into their operations.

“We are currently modifying our system to work on the camera system used by the Beltsville group,” says Mr Lawrence. “Our goal is to have the new prototype tested by the end of 2009.”