Submitted to: Trends in Food Science and Technology
Publication Type: Peer reviewed journal
Publication Acceptance Date: 8/6/2009
Publication Date: 3/5/2010
Publication URL: hdl.handle.net/10113/40856
Citation: Chao, K., Yang, C., Kim, M.S. 2010. Spectral line-scan imaging system for high-speed nondestructive wholesomeness inspection of broilers. Trends in Food Science and Technology. 21(3):129-137. Interpretive Summary: USDA-FSIS inspectors currently remove birds that exhibit signs of injury or contamination from the processing lines during bird-by-bird inspections conducted at a maximum speed of 35 birds per minute for each individual inspector. Subject to human variability, the inspection process makes inspectors prone to fatigue and repetitive injuries, and the inspectors’ speed also limits the maximum possible output for the processing plants. The need to increase production throughput to satisfy increasing chicken consumption and demand places additional pressure on both chicken producers/processors and the U.S. food safety inspection program. A spectral line-scan imaging system was developed for real-time online imaging of chickens on commercial processing lines. The system is particularly well suited for pre-sorting poultry carcasses on high-speed processing lines by removing systemically diseased birds prior to the inspection stations. This system can increase efficiency and reduce cross-contamination risks by minimizing the presence and unnecessary processing of unwholesome birds on the processing line. Real-time detection and prompt removal of unwholesome poultry carcasses can enhance the production throughput of processing plants and help to ensure the safety of poultry products. Commercialization of the system use will provide food processors with a tool to help improve operations and increase production to meet consumer demand.
Technical Abstract: A spectral line-scan imaging system was developed for automated online wholesomeness inspection of broilers and evaluated in a commercial chicken processing plant. Real-time online hyperspectral images acquired by the system on a 140 bird-per-minute processing line were analyzed to optimize Region of Interest (ROI) size and location and to determine key wavebands by which to implement online high-speed multispectral inspection. Multispectral imaging algorithms were implemented to automatically recognize individual carcasses entering and exiting the field of view, to locate the ROI on the bird, and to determine the condition for each carcass as being wholesome or unwholesome. Online multispectral inspection of over 100,000 chickens at 140 birds per minute during two eight-hour shifts demonstrated over 99 percent accuracy in identifying unwholesome birds for removal. The high accuracy obtained from the in-plant evaluation results showed that the system can effectively perform food safety inspection tasks on high-speed processing lines. The system is being adapted for commercial use in pre-sorting chicken during initial processing operations, to help poultry processors improve production efficiency and satisfy increasing consumer demand for poultry products.