Submitted to: Applied Engineering in Agriculture
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: May 7, 2001
Publication Date: January 1, 2002
Citation: CHAO, K., MEHL, P.M., CHEN, Y.R. USE OF HYPER- AND MULTI-SPECTRAL IMAGING FOR DETECTION OF CHICKEN SKIN TUMORS. APPLIED ENGINEERING IN AGRICULTURE. 2002. Interpretive Summary: Poultry and poultry products have increased in popularity with U.S. consumers in recent years. Due to massive production of poultry and the inherent variability and complexity in individual birds, there are great challenges for improvement of the existing organoleptic inspection methods. 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. This paper reports the results of applying hyperspectral and multispectral imaging to inspect chicken skin tumors. Chicken carcasses (68 tumorous and 20 normal) images were examined by the systems. Statistical analysis of tumorous and normal chickens was performed to select wavelength for a multispectral imaging system. Fuzzy logic based classifiers were developed to identify tumorous areas from chicken carcasses. The classifier rgave success rates of 93% and 91% for normal and tumorous skin tissue, respectively. This information is useful to the Food Safety and Inspection Service (FSIS) and poultry processing plants.
Technical Abstract: Hyperspectral and multispectral imaging techniques were used to detect chicken skin tumors. Hyperspectral images of eight tumorous chickens were taken in the spectral range of 420-850 nm. Principle component analysis was applied to select useful wavelength bands (465, 575, 705 nm) from the tumorous chicken images. A multispectral imaging system capable of simultaneously capturing three registered images was used to image 60 tumorous and 20 normal chickens. Multispectral image analysis was performed to generate ratioed images and image features including mean, standard deviation, skewness, and kurtosis for use as inputs to fuzzy classifiers. With respective feature selections, the fuzzy classifiers were able to separate normal from tumorous skin areas with accuracies in the range of 74 to 95%. The fuzzy classifier with combinations of three-input features correctly classified 93 and 91% for normal and tumorous tissue, respectively.