Skip to main content
ARS Home » Research » Publications at this Location » Publication #131453


item Lawrence, Kurt
item Windham, William
item Smith, Douglas

Submitted to: International Society for Optical Engineering
Publication Type: Abstract Only
Publication Acceptance Date: 3/26/2002
Publication Date: 6/10/2002
Citation: Park, B., Lawrence, K.C., Windham, W.R., Smith, D.P. 2002. Hyperspectral imaging for food processing automation [abstract]. International Society for Optical Engineering. 4816:208-316.

Interpretive Summary:

Technical Abstract: A hyperspectral imaging system (HIS) was developed for food processing automation, specifically poultry safety inspection for the identification of fecal surface contamination on poultry carcasses. Both spectral and spatial image data between 400 and 900 nm with approximately 1.0 nm spectral resolution were acquired from fecal and ingesta contaminated poultry carcasses. Calibration model of hyperspectral imaging system was obtained from calibration lighting sources (HgAr and Kr) and reflectance calibration panel for creating spectral images. Optimum wavelength selection from hyperspectral images to identify spatial and spectral characterization of fecal and ingesta materials were discussed. Principal Component and Minimum Noise Fraction transformations were examined to analyze performance of the hyperspectral imaging system for detecting fecal material on carcasses to improve signal to noise ratio of each spectral image. Hyperspectral image processing algorithms, specifically band ratio of dual-wavelength images and further processing such as histogram stretching, were useful to segregate contamination of fecal materials on carcasses. This algorithm can be further applied for real-time identification of fecal contamination. This paper presents the research results of hyperspectral image data collection techniques, hyperspectral image processing algorithms, batch processing for near real-time hyperspectral image analysis, and GCP calibration techniques. HIS can be used effectively for detecting feces (from duodenum, ceca, and colon) and ingesta on poultry carcasses and demonstrates potential application for on-line processing of poultry for safety inspection.