Skip to main content
ARS Home » Research » Publications at this Location » Publication #159176
Title: AUTOMATED ON-LINE INSPECTION OF POULTRY

Author
item Park, Bosoon
item Lawrence, Kurt
item Windham, William
item Smith, Douglas

Submitted to: Feedinfo News
Publication Type: Trade Journal
Publication Acceptance Date: 12/9/2003
Publication Date: 12/10/2003
Citation: Park, B., Lawrence, K.C., Windham, W.R., Smith, D.P. 2003. Automated on-line inspection of poultry. Feedinformation News.

Interpretive Summary: Development of science-based inspection systems to ensure safe production of poultry is very important for Hazard Analysis, Critical Control Point (HACCP) compliance. ARS developed a pilot-scale multispectral imaging system for an automated on-line inspection to detect fecal and ingesta contaminants on the surface of poultry carcasses. The system enables poultry producers to separate fecal and ingesta contaminated carcasses from non-contaminated carcasses, and further prevent carcasses with visible fecal contaminants from entering the chlorinated ice-water tank (chiller) for preventing cross-contamination with other carcasses at poultry processing plants. The system also can be used for washing only those carcasses detected as contaminated, resulting in reduction of water usage in plants and make the system economically feasible for the poultry industry.

Technical Abstract: A multispectral imaging system was developed for an automated on-line fecal detection of poultry carcasses at a processing speed of 140 birds per minute. A common aperture real-time multispectral imaging system with selected trim filters showed a potential to detect fecal and ingesta contaminants on poultry carcasses. In conjunction with an image ratio (566.4-nm image/515.4-nm image) and threshold algorithms, the multispectral imaging system was effective for the real-time detection of fecal and ingesta contaminants on poultry carcasses with a current industrial processing speed. The real-time image processing software developed using Visual C++ programming language, performed well for detecting contaminants (duodenum, ceca, colon, and ingesta) on the skin of poultry carcasses fed by corn, milo, and wheat with soybean meals. The industrial-scale imaging system is currently being tested in our pilot-scale poultry processing facility.