AUTHOR=Wang Jiarong , Wang Wenxiu , Xu Wenya , An Huanjiong , Ma Qianyun , Sun Jianfeng , Wang Jie 

TITLE=Fusing hyperspectral imaging and electronic nose data to predict moisture content in Penaeus vannamei during solar drying

JOURNAL=Frontiers in Nutrition

VOLUME=Volume 11 - 2024

YEAR=2024

URL=https://www.frontiersin.org/journals/nutrition/articles/10.3389/fnut.2024.1220131

DOI=10.3389/fnut.2024.1220131

ISSN=2296-861X

ABSTRACT=Proper control of moisture content (MC) is crucial when drying shrimp because of its direct effect on quality and shelf life. The aim of this study was to establish an accurate method for determining shrimp MC by combining hyperspectral imaging (HSI) with an electronic nose (E-nose). We employed pixel-, feature-, and decision-fusion methods to establish partial least squares regression (PLSR) models. Decision fusion yielded the best predictions, with determination coefficients of 0.9595 and 0.9448 for calibration and validation sets, respectively. Decision fusion also resulted in the lowest root-meansquare errors for the calibration (0.0370) and validation sets (0.0443), as well as the highest relative percent deviation (3.94). Our results demonstrate that a PLSR model based on decision-fusion analysis of combined HSI and E-nose data enabled the accurate and efficient evaluation of shrimp MC. This method will benefit quality assessment and shrimp market monitoring.