Skip to main content
ARS Home » Research » Publications at this Location » Publication #110719
Title: ON-LINE INSPECTION OF POULTRY CARCASSES BY DUAL-CAMERA SYSTEM

Author
item Chao, Kuanglin - Kevin Chao
item Chen, Yud
item Hruschka, William
item GWOZDZ, FRANK - USDA,FSIS,WASHINGTON,DC

Submitted to: Journal of Food Engineering
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 1/29/2001
Publication Date: N/A
Citation: N/A

Interpretive Summary: The Instrumentation and Sensing Laboratory (ISL) of the USDA Agricultural Research Service (ARS) in Beltsville, Maryland has developed a multi-spectral imaging system for poultry carcass inspection. The system uses one pair of black and white cameras with two different color spectrum filters (540 nm and 700 nm) to measure the intensity of diffusely reflected dlight from poultry carcasses. The information collected is used to verify that wholesome carcasses have no visible abnormalities. On-line testing of this system has been conducted in a poultry processing plant. Poultry carcasses of 13,132 normal and 1,459 abnormal were tested by the system. The inspection system gave accuracies of 94% and 87% for normal and abnormal carcasses, respectively. The inspection system shows promise for separation of abnormal chicken carcasses from normal carcasses in the poultry processing line. This information is useful to the Food Safety and Inspection Service (FSIS) and the poultry processing industry.

Technical Abstract: The Instrumentation and Sensing Laboratory (ISL) has developed a multi-spectral imaging system for on-line inspection of poultry carcasses. The ISL design is based on the principle that normal and diseased birds have different chemical compositions of tissues and may have different skin color. On-line trials of the multi-spectral chicken carcass sinspection system were conducted during a 14-day period in a poultry-processing plant, where spectral images of 13,132 normal and 1,459 abnormal chicken carcasses were measured. For off-line model development, the accuracies for classification of normal and abnormal carcasses were 95% and 88%. On-line testing of the neural network classification models with combination of the filter information was performed. The inspection system gave accuracies of 94% and 87% for normal and abnormal carcasses, respectively. This accuracy was consistent with the results obtained previously on laboratory studies. Thus, the inspection system shows promise for separation of abnormal chicken carcasses from normal carcasses in the poultry processing lines.