Skip to main content
ARS Home » Research » Publications at this Location » Publication #167101

Title: SIMPLE REGION OF INTEREST ANALYSIS FOR SYSTEMICALLY DISEASED CHICKEN IDENTIFICATION USING MULTISPECTRAL IMAGING

Author
item YANG, CHUN-CHIEH - SCA, UNIV. KY
item Chao, Kuanglin - Kevin Chao
item Chen, Yud
item Kim, Moon
item EARLY, HOWARD - USDA, FSIS

Submitted to: Transactions of the ASAE
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 12/12/2005
Publication Date: 2/1/2006
Citation: Yang, C.C., Chao, K., Chen, Y.R., dKim, M.S., Early, H.L. 2006. Simple region of interest analysis for systemically diseased chicken identification using multispectral imaging. Transactions of ASAE. 49(1):245-257.

Interpretive Summary: The Food Safety and Inspection Service (FSIS) of the United States Department of Agriculture (USDA) has implemented the Hazard Analysis and Critical Control Point (HACCP) system in poultry plants to improve food safety and prevent food safety hazards in the inspection process (USDA, 1996). FSIS is also testing the proposed HACCP-based Inspection Models Project (HIMP) in a small number of volunteer poultry processing plants to determine if FSIS inspectors and resources can be used more effectively for the poultry inspection program. For poultry plants to meet government food safety regulations and satisfy consumer demand while maintaining their competitiveness, FSIS has recognized the need for new inspection technologies, such as automated computer imaging inspection systems. The overall objective of this research was to develop a simple method for classification of wholesome and systemically diseased chickens based on multispectral images. The wavebands for four filters at 488 nm, 540 nm, 580 nm, and 610 nm were selected using analysis of chicken spectra in the visible region. A total of 660 chicken carcasses were collected in three batches over a period of 6 months in 2003 and 2004, of which 328 were systemically diseased and 332 were wholesome carcasses. Image processing algorithms were developed to extract image features that were then used to develop classification models for the identification of systemically diseased chickens. Independent images were then used to test the classification models. Classification by average intensity (AI) in the region of interest (ROI) area, using the 540 nm and 580 nm wavebands, achieved the best accuracies. The AI540 feature achieved 96.3% and 97.1% accuracies in the ROI area for wholesome and systemically diseased chickens in the third batch, respectively, using thresholds determined from the combined first and second batch images. Similarly, the AI580 feature achieved 96.3% and 98.6% accuracies in the ROI area for wholesome and systemically diseased chickens, respectively, using thresholds determined from the combined first and second batch images. This classification method, using the simple calculation of average intensity, appears well-suited for testing in an automated on-line multispectral inspection system. This information is useful to the Food Safety and Inspection Service (FSIS), and poultry equipment and processing plants.

Technical Abstract: A simple multispectral classification method for the identification of systemically diseased chickens was developed and demonstrated. Color differences between wholesome and systemically diseased chickens were used to select interference filters at 488 nm, 540 nm, 580 nm, and 610 nm for the multispectral imaging system. Over a period of 6 months, 660 chicken images were collected in three batches. An image processing algorithm to locate the region of interest (ROI) was developed in order to define four classification areas on each image: whole carcass (WC), region of interest (ROI), upper region (UR), and lower region (LR). Three feature types, average intensity (AI), average normalization (AN), and average difference normalization (ADN) were defined using several wavebands for a total of 12 classification features. A decision tree algorithm was used to determine threshold values for each of the 12 classification features in each of the 4 classification areas. The AI feature type was found to classify wholesome and systemically diseased chickens better than the AN and ADN features types. Classification by AI in the ROI area, using the 540 nm and 580 nm wavebands, achieved the best accuracies. AI540 achieved 96.3% and 97.1% classification accuracies for wholesome and systemically diseased chickens, respectively. AI580 achieved 96.3% and 98.6% classification accuracies for wholesome and systemically diseased chickens, respectively. This simple classification method shows potential for use in automated on-line applications for chicken inspection.