|Heitschmidt, Gerald - Jerry|
Submitted to: Transactions of the ASABE
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 12/16/2006
Publication Date: 8/1/2007
Citation: Heitschmidt, G.W., Park, B., Lawrence, K.C., Windham, W.R., Smith, D.P. 2007. Improved hyperspectral imaging system for fecal detection on poultry carcasses. Transactions of the ASABE. 50(4):1427-1432.
Interpretive Summary: The Agricultural Research Service (ARS) has developed a hyperspectral imaging system to detect fecal contaminants on poultry carcasses. The system operates from about 400 to 1000 nm and has been used as a research tool to identify select wavelengths for detecting contaminants. Selected wavelengths are to be used in a real-time multispectral system for contaminant detection. ARS has reported that the ratio of reflectance images at 565 nm and 517 nm was able to identify fecal contaminants. However, this ratio alone also misclassified numerous non-fecal carcass features (false positives). Recent modifications to the system, including improved lighting, new camera, new spectrograph, and a new algorithm with an additional wavelength, have increased fecal detection accuracy while reducing the number of false positives. The algorithms and new reflectance-value wavelengths are now being tested in a real-time multispectral imaging system for in-plant trials.
Technical Abstract: The Agricultural Research Service (ARS) has developed imaging technology to detect fecal contaminants on poultry carcasses. The hyperspectral imaging system operates from about 400 to 1000 nm, but only a few wavelengths are used in a real-time multispectral system. Recently, the upgraded system, including improved lighting, a new hyperspectral camera, and a new algorithm with an additional wavelength, has increased fecal detection accuracy while reducing the number of false positives. Additional lighting, targeted at critical areas of the carcass, has resulted in fewer shadows and less glare. New components in the hyperspectral imaging camera have effectively removed misregistration inherent in earlier models and has simplified wavelength and reflectance calibration. The addition of a third wavelength, coupled with a decision tree algorithm, has significantly reduced false positives associated with the 565/517-nm ratio. The new system was used to collect hyperspectral data on 56 stationary poultry carcasses. Carcasses were contaminated with both large and small spots of feces from the duodenum, ceca, colon, and ingesta. There were a total of 1030 contaminants applied to the carcasses. The algorithm correctly identified over 99% of the contaminants with only 25 false positives. About a quarter of the carcasses had at least one false positive pixel.