Click on “Download PDF” for the PDF version or on the title for the HTML version.

If you are not an ASABE member or if your employer has not arranged for access to the full-text, Click here for options.

Non-destructive determination of fat and moisture in pork using near-infrared hyperspectral imaging

Published by the American Society of Agricultural and Biological Engineers, St. Joseph, Michigan

Citation:  2022 ASABE Annual International Meeting  2200172.(doi:10.13031/aim.202200172)
Authors:   Jiewen Zuo, Yankun Peng, Yongyu Li, Qibin Zhuang, Yang Li, Renhong Zhao
Keywords:   fat, hyperspectral imaging, moisture, near-infrared, pork

Abstract. Rapid and non-destructive detection of nutrients in fresh pork is of great importance, in this study, a near-infrared hyperspectral imaging technique was investigated for non-destructive determination of fat content and moisture content of fresh pork. Hyperspectral images (1000-1700 nm) were acquired for fresh pork samples, and the average spectra of the samples were obtained by the threshold segmentation method, the fat content and moisture content of the samples were determined by the Soxhlet extractor method and the ovine drying method. The effects of the models built based on two modeling methods, four preprocessing methods and four feature wavelength extraction algorithms were compared, and the best model building method was preferred, the fat and moisture contents in pork were visualized based on the best prediction model. The results showed that the PLSR model built by SNV preprocessing and using the feature wavelength extracted by sCARS was the best for fat and moisture content prediction in pork, and the Rp, RMSEp, RPD of the fat and moisture prediction models were 0.900 and 0.916, 0.780 and 0.593, 2.247 and 2.510, respectively. using the fat and moisture distribution pseudo-color graphs, a visual representation of the nutrient composition in fresh pork was achieved.

(Download PDF)    (Export to EndNotes)