Using Spectral Imaging to Classify Pork Muscle31 October 2014
Using visible and near infrared hyperspectral imaging can help improve the classification of fresh and thawed frozen pork meat.
An Chinese Irish study looked at the classification of fresh and frozen-thawed pork muscles using visible and near infrared hyperspectral imaging and textural analysis.
Spectral features were identified for classifying fresh and frozen-thawed meats and textural features of hyperspectral imaging were extracted.
The research team from the College of Light Industry and Food Sciences, South China University of Technology and the Food Refrigeration and Computerised Food Technology (FRCFT) department at the School of Biosystems Engineering at University College Dublin, developed PNN classifiers with spectral and textural features.
Differences in fresh and frozen-thawed meats were visualised.
The research team of Hongbin Pu, Da-Wen Sun, Ji Ma and Jun-Hu Cheng investigated the potential of visible and near infrared hyperspectral imaging as a rapid and nondestructive technique for classifying fresh and frozen-thawed meats by integrating critical spectral and image features extracted from hyperspectral images in the region of 400–1000 nm.
Six feature wavelengths (400, 446, 477, 516, 592 and 686 nm) were identified using uninformative variable elimination and successive projections algorithm.
Image textural features of the principal component images from hyperspectral images were obtained using histogram statistics (HS), gray level co-occurrence matrix (GLCM) and gray level-gradient co-occurrence matrix (GLGCM).
By these spectral and textural features, probabilistic neural network (PNN) models for classification of fresh and frozen-thawed pork meats were established.
Compared with the models using the optimum wavelengths only, optimum wavelengths with HS image features, and optimum wavelengths with GLCM image features, the model integrating optimum wavelengths with GLGCM gave the highest classification rate of 93.14 per cent and 90.91 per cent for calibration and validation sets, respectively.
The results indicated that the classification accuracy can be improved by combining spectral features with textural features and the fusion of critical spectral and textural features had better potential than single spectral extraction in classifying fresh and frozen-thawed pork meat.
The research has been published in the January 2015 issue of Meat Science.
You can view the full report by clicking here.