http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
The Use of Hyperspectral Imaging to Detect the Quality Parameters of Packaged Apples
( Umuhoza Aline ),( Byoung-kwan Cho ) 한국농업기계학회 2023 한국농업기계학회 학술발표논문집 Vol.28 No.2
The research on packaged food products has perceived substantial growth, mainly driven by the demand of consumers and processing industries to ascertain the actual quality level of the product inside the packaging. This study proposes a novel prediction of apples' moisture content, solid soluble content, and pH under polyethylene packaging material. A total of 50 apples were imaged using hyperspectral imaging systems in the visible near-infrared (VIS-NIR 400-1000 nm) and short-wave infrared (SWIR 1000-2500 nm) regions. The spectral data were collected from the region of interest (ROI) of the images. Subsequently, the Partial least squares regression (PLSR) model was developed based on the calculated variables using variable selection algorithms in combination with different pre-processing techniques, and their performance was evaluated using correlation coefficients and root mean square prediction error. The findings showed the essential role of HSI as a valuable tool to effectively predict packaged apples' moisture content, solid soluble content, and pH. This approach could contribute to easy recognition and line processing of food products, benefiting consumers and packaging industries for quality control and safety purposes.
( Tanjima Akter ),( Mohammad Akbar Faqeerzada ),( Umuhoza Aline ),( Ye-na Kim ),( Byoung-kwan Cho ) 한국농업기계학회 2023 한국농업기계학회 학술발표논문집 Vol.28 No.2
Apples and pears are among the most prominent deciduous fruits globally, with their external quality recognized as a pivotal factor in enhancing consumer perceptions of overall fruit quality. These fruits occasionally presented flaws, a problem for keeping food processing quality standards, such as irregular shape, bruises, discoloration, diseases, scars, and deformities in apples and pears. Hence, there are growing demands for non-destructive inspection of the external quality of fruits. Therefore, this study employed visible near-infrared (Vis-NIR: 400-1000 nm) and short-wave infrared (SWIR: 1000-2500 nm) hyperspectral imaging for quality inspection of pears and apples. In total, 2000 (5x2x200) spectral data points were acquired from the region of interest (ROI) within the imaged samples. Principal Component Analysis (PCA) was employed to identify the optimal wavelengths by loading principal components (PCs). Moreover, a Partial Least Squares Discriminant Analysis (PLS-DA) model was constructed based on the calculated variables, employing the Variable Importance in Projection (VIP) band selection algorithms. Both models demonstrated robust performance in calibration and prediction when applied to hyperspectral imaging systems. The results underscore the potential of integrating multivariate analysis techniques as a rapid and comprehensive method for effectively measuring the external quality attributes of apples and pears.