Page Banner

United States Department of Agriculture

Agricultural Research Service

Title: Contaminant Classification of Poultry Hyperspectral Imagery Using Spectral Angle Mapper Algorithm

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

Submitted to: Biosystems Engineering
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: November 30, 2006
Publication Date: March 15, 2007
Citation: Park, B., Windham, W.R., Lawrence, K.C., Smith, D.P. 2007. Contaminant classification of poultry hyperspectral imagery using spectral angle mapper algorithm. Biosystems Engineering. 96(3): 323-333

Interpretive Summary: Food safety is important for public health and reductions in the potential health risks to consumers from human pathogens in food are crucial for food safety. While a number of factors can influence bacterial contamination of chicken carcasses, fecal contamination at the poultry processing plants is one of the issues be considered. Improper handling of carcasses and equipment that is not good working condition can often contribute to the problem. The scientific inspection methods play an important role for the detection and prevention of various types of contamination, and produce safe food. Since hyperspectral imaging technique has been demonstrated to be a potential tool for poultry safety inspection, particularly fecal contamination, a hyperspectral image classification method was developed for identifying the type and source of fecal contaminants. This new imaging technology can improve the FSIS poultry safety inspection program by incorporating scientific testing and systematic detection of fecal contamination in poultry processing plants.

Technical Abstract: Spectral angle mapper (SAM) supervised classification method for hyperspectral poultry imagery was performed well for classifying fecal and ingesta contaminants on the surface of broiler carcasses. Spatially averaged spectra of three different feces from the duodenum, ceca, colon, and ingesta of corn/soybean diet were used for classification data. SAM classifier using reflectance of hyperspectral data with 512 narrow bands from 400 to 900 nm was able to classify three different feces and ingesta on the surface of poultry carcasses. Based on the comparison with ground truth region of interest, both classification accuracy and kappa coefficient increased when spectral angle increased. The overall mean accuracy and corresponding mean kappa coefficient to classify fecal and ingesta contaminants were 90.2% (standard deviation = 5.4%) and 0.884 (standard deviation = 0.063) when a spectral angle of 0.3 radians was used as a threshold.

Last Modified: 9/22/2014