Share Email Print
cover

Proceedings Paper

Nondestructive technique for detecting diseased poultry carcasses
Author(s): Yud-Ren Chen
Format Member Price Non-Member Price
PDF $14.40 $18.00

Paper Abstract

In response to the need of the U.S. Food Safety and Inspection Service, the Agriculture Research Service has undertaken a project to develop an accurate, reliable, and nondestructive sensor for detecting poultry diseased carcasses on-line at poultry processing plants. This paper presents some results of a study on the development of a nondestructive technique for the detection of abnormal poultry carcasses based on the spectroscopy of the carcasses. A diode array spectrophotometer equipped with a fiber optic probe was used to obtain optical spectra of the breasts of normal, septicemic, and cadaver poultry carcasses in visible and near-infrared regions (500 - 1100 nm). Optimal wavelengths of reflectance and interactance in the range of 500 to 850 nm were obtained for classifying the carcasses into normal and abnormal (septicemic and cadaver) classes. A back-propagation neural network model was used to develop classifiers for the classification of poultry carcasses into normal, septicemic, and cadaver classes.

Paper Details

Date Published: 30 April 1993
PDF: 12 pages
Proc. SPIE 1796, Chemical, Biochemical, and Environmental Fiber Sensors IV, (30 April 1993); doi: 10.1117/12.143542
Show Author Affiliations
Yud-Ren Chen, U.S. Dept. of Agriculture (United States)


Published in SPIE Proceedings Vol. 1796:
Chemical, Biochemical, and Environmental Fiber Sensors IV
Robert A. Lieberman, Editor(s)

© SPIE. Terms of Use
Back to Top