College of Mechanical and Electrical Engineering, Shandong Intelligent Engineering Laboratory of Agricultural Equipment, Shandong Agricultural University, Tai'an 271018, China; Nanjing Institute of Agricultural Mechanization, Ministry of Agriculture and Rural Affairs, Nanjing 210014, China.
College of Mechanical and Electrical Engineering, Shandong Intelligent Engineering Laboratory of Agricultural Equipment, Shandong Agricultural University, Tai'an 271018, China.
Food Chem. 2022 Aug 30;386:132864. doi: 10.1016/j.foodchem.2022.132864. Epub 2022 Mar 31.
The quality of tomatoes is usually predicted by measuring a single index, rather than a comprehensive index. To find a comprehensive index, visible and near infrared (Vis-NIR) hyperspectral imaging was used for capturing the images of three varieties of tomatoes, and twelve quality indexes were measured as the reference standards. The changing trends and correlations of different indexes were analyzed, and comprehensive quality index (CQI) was proposed through factor analysis. The characteristic wavelengths were selected by successive projection algorithm (SPA) based on the hyperspectral data, which was used to establish three regression models for CQI prediction. The result indicated that MLR achieved good performance withR = 0.87, RMSEV = 1.33 and RPD = 2.58. After that, spatial distribution map was generated to visualize the CQI in tomato fruit. This study indicated that the comprehensive quality of tomatoes can be predicted non-destructively based on hyperspectral imaging and chemometrics, determining the optimal harvesting period.
番茄的品质通常通过测量单一指标来预测，而不是综合指标。为了找到一个综合指标，利用可见近红外（Vis-NIR）高光谱成像技术获取了三种番茄的图像，并将十二个质量指标作为参考标准进行测量。分析了不同指标的变化趋势和相关性，并通过因子分析提出了综合质量指数（CQI）。基于高光谱数据，采用连续投影算法（SPA）选择特征波长，建立了用于 CQI 预测的三个回归模型。结果表明，MLR 具有较好的性能，R=0.87，RMSEV=1.33，RPD=2.58。之后，生成了空间分布图，直观地显示了番茄果实中的 CQI。本研究表明，基于高光谱成像和化学计量学可以无损地预测番茄的综合品质，确定最佳收获期。
