Assessment of Intramuscular Fat Content of Pork using NIR Hyperspectral Images of Rib End
Authors: Huang, H., L. Liu, M. Ngadi
Description: The purpose of this study was to investigate the effective method for rapid determination of intramuscular fat (IMF) content of pork at the last 6 ribs using NIR images of round rib end. A near infrared (NIR) hyperspectral imaging system (900-1700 nm) was used for NIR image acquisition of rib ends. Pattern analysis techniques were applied to process hyperspectral images. Spectral features and image features were extracted and selected from hyperspectral data. Multiple linear regression (MLR) models were established at 6 different rib locations to predict IMF content using the selected features at the rib end. Leave-one-out cross validation was used to test the robustness of established models. All the 6 MLR models showed good performance with correlation coefficients of calibration (Rc) ? 0.90, correlation coefficients of cross validation (Rcv) ? 0.87, correlation coefficients of prediction (Rp) ? 0.81. The model for IMF content at the 2nd/ 3rd last rib provided the best prediction result with Rc of 0.96, Rcv of 0.95, and Rp of 0.83. These results demonstrated the potential of replacing the conventional destructive determination of IMF content of pork at different ribs along one Longissimus dorsi with non-destructive assessment using hyperspectral images of rib ends.
Keywords: pork, intramuscular fat content, NIR, hyperspectral imaging, rib end.
Technical field: Information Systems Engineering
Conference name: CSBE/SCGAB 2015 Annual Conference, Edmonton, AB, 5-8 July 2015.
Session name: Food Engineering - Instrumentation and Modelling
Citation: Huang, H., L. Liu, M. Ngadi. 2015. Assessment of Intramuscular Fat Content of Pork using NIR Hyperspectral Images of Rib End. CSBE/SCGAB 2015 Annual Conference, Edmonton, AB, 5-8 July 2015.
Publisher: Canadian Society for Bioengineering
Language 1: en
Rights: Canadian Society for Bioengineering