Location: Quality & Safety Assessment ResearchTitle: Prediction of quality traits and grades of intact chicken breast fillets by hyperspectral imaging
|YANG, YI - China Agricultural University|
|WANG, WEI - China Agricultural University|
|JIANG, HONGZHE - China Agricultural University|
Submitted to: British Poultry Science
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 7/8/2020
Publication Date: 9/29/2020
Citation: Yang, Y., Wang, W., Zhuang, H., Yoon, S.C., Jiang, H. 2020. Prediction of quality traits and grades of intact chicken breast fillets by hyperspectral imaging. British Poultry Science. https://doi.org/10.1080/00071668.2020.1817326.
Interpretive Summary: Assessment of chicken meat quality traits such as meat color and pH typically requires different sensors and techniques that are sometimes destructive to test samples and time-consuming. The color, especially lightness, and pH of fresh chicken breast fillets are important quality traits because they can be used to determine the quality grade of chicken meat according to three different categories: 'normal', 'dark, firm and dry', and 'pale, soft and exudative'. This study was conducted to determine if a non-destructive and non-contact imaging technology, called hyperspectral imaging, is feasible to predict the quality grade of individual chicken breast fillets. Hyperspectral imaging is a technology that can extract and process the unique spectral fingerprints of agricultural and food products to analyze at each image pixel. Hyperspectral images of fresh 104 chicken breast fillets were measured in the full visible and near-infrared spectral range from 400 to 2,500 nm with two hyperspectral image sensors. This study found that multivariate data analysis (partial least squares regression) was effective to predict the grades with the accuracy of about 82%. The results of our study indicate that hyperspectral imaging has the potential for quality prediction of fresh chicken meat.
Technical Abstract: In this study, hyperspectral imaging was evaluated for its usefulness to predict quality traits and grading of intact chicken breast fillets. Lightness of colour (L*) and pH of the fillets were measured as quality traits, and samples were then selected and graded to three different quality categories i.e., dark, firm and dry (DFD), normal (NORM), and pale, soft and exudative (PSE) based on these two quality traits. Based on the prediction performance of full wavelength partial least square regression (PLSR) models, the spectral range of visible and near infrared (Vis-NIR) were more suitable for the evaluation of quality traits and grading than the range of near infrared (NIR). Key wavelengths of each quality trait and grade value were selected by regression coefficient (RC) method. The new key wavelength PLSR models showed good predictive performances (Rp=0.85 and RMSEp=2.18 for L*, Rp=0.84, and RMSEp=0.13 for pH, and Rp=0.80 and RMSEp=0.44 for quality grading). The classification accuracy for grades were 85.71% (calibration set) and 81.82% (prediction set), respectively. Finally, distribution maps showed that quality traits and grades of samples were able to be visualised. These results suggested that hyperspectral imaging has the potential for quality prediction of fresh chicken meat.