Submitted to: Sensing and Instrumentation for Food Quality and Safety
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: May 12, 2008
Publication Date: June 18, 2008
Citation: Park, B., Kise, M., Windham, W.R., Lawrence, K.C., Yoon, S.C. 2008. Textural Analysis of Hyperspectral Images for Improving Contaminant Detection Accuracy. Sensing and Instrumentation for Food Quality and Safety. 2(3):208-214. Interpretive Summary: Detecting fecal and ingesta contaminants during poultry processing is crucial for food safety to protect consumers from foodborne diseases. Food Safety Inspection Service (FSIS) is pursuing a science-based strategy to improve the safety of poultry and poultry products to better protect public health from potential illness due to fecal contaminants, because of high correlation between feces and pathogenic bacteria. The ARS has developed real-time imaging system for fecal and ingesta contaminant detection on broiler carcasses for harsh environmental poultry processing plant. Particularly, real-time image processing software was developed for increasing detection accuracy. The test results of industrial-scale real-time system demonstrated that the multispectral imaging technique was able to detect fecal contaminants with a commercial processing speed (currently 140 birds per minute). This industrial-scale imaging system can improve the FSIS poultry safety inspection program by incorporating scientific testing and efficacy of online fecal detection during poultry processing.
Technical Abstract: Previous studies demonstrated a hyperspectral imaging system has a potential for poultry fecal contaminant detection by measuring reflectance intensity. The simple image ratio at 565 and 517-nm images with optimal thresholding was able to detect fecal contaminants on broiler carcasses with high accuracy. However, differentiating false positives from real contaminants was always challenging, especially cecum. Further image processing such as textural analysis in the spatial domain was able to reduce errors. In this study, textural analysis of hyperspectral images was conducted to improve detection accuracy by reducing false positives. Specifically, textural analysis with co-occurrence matrix of hyperspectral images performed well to identify “true” contamination regardless of fecal types. In addition, co-occurrence matrix textural features including average, variance, entropy, contrast, correlation, moment of poultry hyperspectral images were investigated for selecting optimal features to represent contamination. Image processing with co-occurrence textural analysis was able to improve fecal detection accuracy without additional optical filters.