1a.Objectives (from AD-416)
To develop algorithms for detection of individual diseases and defects on poultry carcasses.

1b.Approach (from AD-416)
1. Develop real-time image acquisition and disease and defect detection software modules using LabView. 2. Integrate the disease and defect detection modules with ISL pilot-scale poultry inspection system. 3. Test processing speed and robustness of the algorithms at the ISL pilot-scale poultry processing line.

3.Progress Report
This report documents research conducted under a Specific Cooperative Agreement between ARS and the University of Kentucky. Additional details of the research can be found in the report for the parent project 1265-42000-013-00D "Development of New and Improved Systems to Enhance Food Safety Inspection and Sanitation of Food Processing." Work during the past year has resulted in a hyperspectral/multispectral line-scan imaging system for differentiation of wholesome and systemically diseased chickens. In-plant testing was conducted for chickens on a commercial evisceration line moving at a speed of 70 birds per minute. Hyperspectral image data was acquired for a calibration data set of 543 wholesome and 64 systemically diseased birds, and for a testing data set of 381 wholesome and 100 systemically diseased birds. The calibration data set was used to develop the imaging system’s parameters for conducting multispectral inspection based on fuzzy logic detection algorithms using selected key wavelengths. Multispectral classification using a threshold of 0.4 was able to achieve 90.6 percent accuracy for wholesome birds and 93.8 percent accuracy for systemically diseased birds in the calibration data set, and 97.6 percent accuracy for wholesome birds and 91.4 percent accuracy for systemically diseased birds in the testing data set. By adjusting the classification threshold, 100 percent accuracy was achieved for systemically diseased birds with a decrease in accuracy for wholesome birds to 88.7 percent. This adjustment shows that the system can be feasibly adapted as needed for implementation for specific purposes, such as paw harvesting operations or pre-screening for food safety inspection. This line-scan imaging system is ideal for directly implementing multispectral classification developed from hyperspectral image analysis.

Monitoring activities included several conference calls, and also a meeting during the 2007 ASABE Annual International Meeting, June 17-20, with the cooperator from the University of Kentucky.