基于自制便携式近红外光谱仪的枇杷果实可溶性固形物无损检测及年度重复验证  

Non-destructive detecting and annual duplicate verification of loquat fruit’s total soluble solids based on self-developed portable near-infrared spectrometer

作　　者：魏雨晴 王毓宁[2] 李绍佳 孙崇德 吴迪 WEI Yuqing;WANG Yuning;LI Shaojia;SUN Chongde;WU Di(Institute of Fruit Science,College of Agriculture and Biotechnology,Zhejiang University,Hangzhou 310058,China;Jiangsu Taihu Area Institute of Agricultural Sciences,Suzhou 215106,Jiangsu,China)

机构地区：[1]浙江大学农业与生物技术学院果树科学研究所,杭州310058 [2]江苏太湖地区农业科学研究所,江苏苏州215106

出　　处：《浙江大学学报（农业与生命科学版）》2020年第1期119-125,共7页Journal of Zhejiang University:Agriculture and Life Sciences

基　　金：“十三五”国家重点研发计划(2017YFD0401302);江苏省农业科技自主创新资金项目[CX(17)3029];浙江省科协育才工程(2018YCGC006)。

摘　　要：采用自主研发的便携式枇杷果实品质无损检测仪,分别采集2018和2019年的枇杷光谱数据,然后采用化学计量学分析方法构建枇杷果实中的重要品质指标——可溶性固形物(total soluble solids,TSS)的检测模型,并开展独立样本年度验证。结果表明:建立的竞争性自适应重加权算法-最小二乘支持向量机(competitive adaptive reweighted sampling-least squares support vector machines,CARS-LS-SVM)模型的预测集相关系数和预测集均方根误差分别为0.818和1.453,说明枇杷果实品质无损检测仪可以实现‘白玉’品种枇杷可溶性固形物的快速无损检测,而且经过分级模型筛选后,一类好果率和二类好果率分别从50.00%和73.33%提升到了86.67%和85.71%。总之,本研究建立了更加实用的枇杷果实分级模型,实现了江苏省苏州地区‘白玉’品种采后枇杷果实的快速无损检测,为进一步提升苏州地区枇杷果实销售的经济价值和市场规范提供了强有力的技术支持。A self-developed portable non-destructive quality detector was used to collect spectral data of‘Baiyu’loquat fruit harvested in 2018 and 2019,respectively,and then chemometric analysis methods were used to build detection model for total soluble solids(TSS)of loquat fruit.Finally,annual verification of independent samples was conducted.The results showed that correlation coefficient of prediction(Rp)and root mean square error of prediction(RMSEP)of the established competitive adaptive reweighted sampling-least squares support vector machines(CARS-LS-SVM)model were 0.818 and 1.453,respectively,which indicated that the self-developed non-destructive detector could achieve rapid non-destructive measurement on TSS of‘Baiyu’loquat.Moreover,the grading and screening model improved the first-class fruit rate and the second-class fruit rate from 50.00%to 86.67%,73.33%to 85.71%,respectively.In conclusion,this study establishes a more practical loquat fruit grading model,which achieves rapid non-destructive measurement of postharvest‘Baiyu’loquat fruit in Suzhou area of Jiangsu Province and provides technological support for further improving economic value of loquat fruit and market regulation in Suzhou.

关 键 词：枇杷 可溶性固形物 无损检测 近红外光谱技术 年度重复验证 

分 类 号：O657.33[理学—分析化学] S667.3[理学—化学]