|Chao, Kuanglin - Kevin Chao|
Submitted to: Proceedings of SPIE
Publication Type: Proceedings
Publication Acceptance Date: 10/29/2003
Publication Date: 12/1/2003
Citation: Chao, K., Chen, Y.R. 2003. High-speed poultry inspection using visible/near-infrared spectrophotometer. SPIE Proceedings entitled: Monitoring Food, Safety, Agriculture, and Plant Health. 5271:51-61.
Interpretive Summary: Development of high speed and reliable inspection systems to ensure safe production of poultry during post-harvest processing has become an important issue, as the public is demanding assurance of better and safer food. The Instrumentation and Sensing Laboratory, ARS, USDA, has developed a method and a visible/near-infrared (Vis/NIR) spectroscopic system to automate the poultry inspection process. The inspection system, which detects wholesome and diseased birds with different chemical compositions of tissues and skin color, consisted of a CCD spectrophotometer, a prism-grating-prism spectrograph, light source, associated laser triggering sensors, and an industrial computer. On-site trials of the Vis/NIR chicken carcass inspection system were conducted in a poultry processing plant in Athens, Georgia. Spectra of 450 wholesome and 426 unwholesome chicken carcasses were measured. The instrument measured the spectra of veterinarian-selected carcasses on a processing line running at speeds of 140 and 180 birds per minute (bpm). The automatic inspection system was able to identify wholesome broiler carcasses at a 140 bpm speed with a 95 percent accuracy rate and unwholesome birds with a 92 percent accuracy rate. At 180 bmp, the accuracy rates for wholesome and unwholesome birds were 94 and 92 percent respectively. This information is useful to the FSIS, and poultry equipment manufacturers and processing plants.
Technical Abstract: A new Vis/NIR spectroscopic system was developed using modularized software components and was able to collect real-time spectral measurements for chickens running on high speed processing lines. Data analysis and modeling demonstrated that the system can successfully differentiate between wholesome and unwholesome birds. At 140 bpm, the system was able to correctly classify 95% and 92% of wholesome and unwholesome birds, respectively. At 180 bpm, the system was able to correctly classify 94% and 92% of wholesome and unwholesome birds, respectively. The Automated Poultry Inspector program also has a module for real-time prediction and the capacity to accommodate other functions as needed for real-time operation. The results of this study show this automated poultry inspection system based on Vis/NIR spectroscopy is ready for implementation on commercial high speed poultry processing lines for real-time operation. Using such an automated inspection system would greatly improve overall production efficiency of processing plants.