Search

Wooden Breast (WB) anomaly on poultry meat causes changes in appearance, reduction of technological and nutritional quality, and consumer acceptance. The objective of this study was to identify and classify chicken with WB using a Computer Vision System (CVS) and spectral information from the Near Infrared (NIR) region by linear and nonlinear algorithms. Moreover, it was characterized the physicochemical and technological parameters, which supported a decision tree modeling. Pectoralis major muscle (n = 80) were collected from a poultry slaughterhouse, spectral information was obtained by NIR and CVS, and WB of chicken was characterized. Combining image analyses with a Support Vector Machine (SVM) classification model, 91.8% of chicken breasts were correctly classified as WB or Normal (N). NIR spectral information showed 97.5% of accuracy. WB showed significant increases in moisture and lipid contents and value of a*, decreases of protein and ash contents, and water holding capacity. The shear force of raw WB was 49.51% hardness, and after cooking was 31.79% softer than N breast. CVS and NIR spectroscopy can be applied as rapid and non-destructive methods for identifying and classifying WB in slaughterhouses.

Customary methods for assessing poultry meat quality are invasive, generally time-consuming and require specialised analysts. Near-infrared (NIR) spectroscopy represents a powerful alternative with none of these drawbacks. Combining the absorbance for different wavelengths in NIR range with analytical information from reference methods, it is possible to build models for meat quality prediction. These prediction models can be developed independent of each property or based on the statistical dependency of desired extrinsic properties, namely Single-target (ST) or Multi-target (MT), respectively. A new MT method, designated Multi-target Augmented Stacking (MTAS), is compared to the performance of ST and other three MT methods (Stacked Single Target, Ensemble of Regressor Chains and Deep Regressor Stacking) to predict twelve poultry meat characteristics. Different learning algorithms were selected to compose each prediction method: Support Vector Machine (SVM), Random Forest (RF) and Classification and Regression Tree (CART). Results demonstrated that the coefficient of determination was greater than, or equal to, 0.5 for nine out of twelve targets. In addition, the prediction errors were comparable to the error obtained by traditional analysis. Furthermore, MT methods were statistically superior to ST method. In particular, SVM and RF outperformed CART, providing a new tool for identification of poultry meat attributes. (C) 2018 IAgrE. Published by Elsevier Ltd. All rights reserved.