Click on “Download PDF” for the PDF version or on the title for the HTML version.
If you are not an ASABE member or if your employer has not arranged for access to the full-text, Click here for options.
Evaluation of White Striping in Broiler Meat Using Structured Illumination Reflectance Imaging
Published by the American Society of Agricultural and Biological Engineers, St. Joseph, Michigan www.asabe.org
Citation: 2022 ASABE Annual International Meeting 2200707.(doi:10.13031/aim.202200707)
Authors: Ebenezer Obaloluwa Olaniyi, Yuzhen Lu, Jiaxu Cai, Anuraj Theradiyil Sukumaran, Tessa Jarvis
Keywords: Grading, Image Processing, Myopathy, Machine Learning, Poultry, Structured illumination
Abstract. Muscle myopathies or defects such as white striping (WS), woody breasts (WB), and spaghetti meat, downgrade the quality of broiler meat, and cause loss of hundreds of millions of dollars annually for the United States poultry industry. Visual inspection is currently practiced for assessing myopathies in broiler meat. This method is however prone to human evaluation error, labor-intensive, and expensive. Imaging techniques under uniform or diffuse illumination have been investigated as alternatives to visual inspection for assessing broiler meat quality, but their performance is not always satisfactory, especially for detecting subtle defects with few visual symptoms. This study proposes using emerging structured-illumination reflectance imaging (SIRI) technique for quality assessment of poultry products. Unlike existing imaging techniques, SIRI uses spatially modulated, patterned light for imaging poultry samples. The study made the first proof-of-concept evaluation of the utility of SIRI for detecting WS conditions in broiler breast meat. Chicken breast fillets with varying degrees of WS severity were imaged by an in-house assembled broadband SIRI system under sinusoidal illumination at eight different spatial frequencies (0.05-0.40 cycles/mm). The acquired pattern images were demodulated into direct components (DC) and amplitude components (AC) at each spatial frequency. Texture features were extracted from DC and AC images for building classification models using regularized linear discriminant analysis, which were to classify meat samples into two (normal vs defect), three (normal, moderate vs severe) and four (normal, moderate, severe vs extreme) classes according to WS conditions. Visual inspection of demodulated images revealed the improved power of AC images over DC in resolving WS characteristics, depending on the spatial frequency of illumination patterns. Modeling results show that, compared to DC images, AC images yielded the maximum accuracy improvements of 8.33%, 7.93%, and 12.3% in 2-, 3- and 4-class classification, respectively. Models using ranked features yielded similar improvements of up to 9.53% by AC over DC. The SIRI technique is promising for enhancing the detection of WS conditions of broiler meat.
(Download PDF) (Export to EndNotes)