Skip to main content
ARS Home » Research » Publications at this Location » Publication #172118

Title: SYSTEMICALLY DISEASED CHICKEN IDENTIFICATION USING MULTISPECTRAL IMAGES AND REGION OF INTEREST ANALYSIS

Author
item YANG, CHUN-CHIEH - UNIV OF KENTUCKY
item Chao, Kuanglin - Kevin Chao
item Chen, Yud
item EARLY, HOWARD - USDA, FSIS

Submitted to: Computers and Electronics in Agriculture
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 5/5/2005
Publication Date: 6/5/2005
Citation: Yang, C.C., Chao, K., Chen, Y.R., Early, H.L. 2005. Systemically diseased chicken identification using multispectral images and region of interest analysis. Computer and Electronics in Agriculture. 49:255-271.

Interpretive Summary: To improve food safety and prevent food safety hazards in the inspection process, it is important for poultry plants to meet government food safety regulations and satisfy consumer demand while maintaining their competitiveness. The need for new inspection technologies, such as automated computer imaging inspection systems, has been recognized. Multispectral imaging is a non-destructive method and can obtain high classification accuracies for chicken conditions. It has great potential for use in high-speed on-line processing plant operations. In this study, a simple multispectral classification method for the identification of systemically diseased chickens was developed and tested between two different imaging systems. An image processing algorithm was developed to define and locate the region of interest as classification areas on the image. The average intensity was calculated for each classification area of the chicken image. A decision tree algorithm was used to determine threshold values for each classification areas. The wavelength of 540 nm was used for image differentiation purpose. There were 164 wholesome and 176 systemically diseased chicken images collected using the first imaging system, and 332 wholesome and 318 systemically diseased chicken images taken by the second imaging system. The differentiation thresholds, generated by the decision tree method, based on the images from the first imaging system were applied to the images from the second imaging system, and vice versa. The accuracy from evaluation was 95.7% for wholesome and 97.7% of systemically diseased chickens for the first image batch, and 99.7% for wholesome and 93.5% for systemically diseased chickens for the second image batch. The result showed that using single wavelength and threshold, this simple classification method can be used in automated on-line applications for chicken inspection.

Technical Abstract: A simple multispectral classification method for the identification of systemically diseased chickens was developed and cross-system validated using two different imaging systems. An image processing algorithm was developed to define and locate the region of interest (ROI) as the classification area on the image. The average relative reflectance intensity of the ROI was calculated for each chicken image. The Classification and Regression Trees (C&RT) decision tree algorithm was used to determine the threshold value to differentiate systemically diseased chickens from wholesome ones. The wavelength of 540 nm, selected as the key wavelength, was used in two different multispectral imaging systems for image differentiation. For image masking, the 700 nm wavelength was implemented in the first imaging system and the 610 nm wavelength in the second imaging system. The first image batch, containing 164 wholesome and 176 systemically diseased chicken images, was collected using the first imaging system. The second image batch, containing 332 wholesome and 318 systemically diseased chicken images, was taken by the second imaging system. The first differentiation threshold, based on the first image batch and generated by the decision tree method, was applied to the second image batch for cross-system validation, and vice versa. The accuracy from validation was 95.7% for wholesome and 97.7% of systemically diseased chickens for the first image batch, and 99.7% for wholesome and 93.5% for systemically diseased chickens for the second image batch. The threshold values, each generated using only one of the two image batches, were similar. The results showed that using a single key wavelength and a threshold, this simple image processing and classification method could be used in automated on-line applications for chicken inspection.