Classifying of Nellore cattle beef on Normal and DFD applying a non conventional technique

Abstract

The aim of this study was to evaluate the accuracy in the Normal and DFD classification in Nellore beef using a bench-top hyperspectral imaging system. A hyperspectral imaging system (λ=928–2524nm) was used to collect hyperspectral images of the Longissimus thoracis et lumborum (n=78) of Nellore cattle. The images were processed, being selected region of interest and extracted spectra image and were selected the wavelengths considered most important for the treats evaluated. Six linear discriminant models were developed to classify beef samples on Normal and DFD. The model using all wavelengths associated with the reflectance and absorbance spectrum transformed with the pretreatment 2nd derivative resulted in an overall accuracy of 93.6% for both pretreatments. In this configuration, the model was able to classify correctly 73 samples from a total of 78 samples. The results demonstrate that the hyperspectral imaging system may be considered a viable technology for beef classification on Normal and DFD.