Non-destructive assessment of apricot fruit quality by portable visible-near infrared spectroscopy
Non-destructive assessment of apricot fruit quality by portable visible-near infrared spectroscopy
复制标题
DOI：
10.1016/j.lwt.2009.01.015
复制
发表时间：
2009-07-01
影响因子：
6
通讯作者：
Christen, D.
中科院分区：
文献类型：
--
作者：
Camps, C.;Christen, D.
研究了便携式近红外光谱仪对杏果实品质的检测能力。采用不同精度的校正模型，对葡萄果实的可溶性固形物含量（SSC）、总酸度（TA）和硬度（Fi）进行了测定。为每个品种建立了模型，并尝试了结合不同品种的全球模型。确定SSC，交叉验证的均方根误差（RMSECV）在0.67和1.1白利糖度之间，R值在0.88和0.96之间。关于Fi，预测的准确性是品种依赖性的。这些预测对品种Kioto和Harostar是正确的，RMSECV值在6.2%和13%之间（R值在0.85和0.92之间），而对Bergarouge不满意（RMSECV = 24%）。TA预测的RMSECV值在0.79和2.61 g 100 ml（-1）之间，R值在0.73和0.97之间。在第二个应用中，近红外光谱被用来分类杏果实根据其品种和颜色强度与正确的效率。本研究获得的结果表明，NIRS技术可适用于杏子质量，并且此类便携式设备可以帮助对果园和收获后的水果进行完整的跟踪。(C)2008爱思唯尔有限公司保留所有权利。
The ability of portable Near Infrared Spectroscopy to determine apricot fruit quality has been studied. Calibration models allowing the determination of soluble solids content (SSC), total acidity (TA) and firmness (Fi) of apricots were carried out with variable precisions. Models were built for each variety and global models combining different varieties were attempted. SSC was determined with a root mean square error of cross-validation (RMSECV) comprised between 0.67 and 1.1 degrees Brix and R-values between 0.88 and 0.96. Concerning Fi, the accuracy of the prediction was variety dependant. These predictions were correct for the varieties Kioto and Harostar with RMSECV-values between 6.2% and 13% (R-values between 0.85 and 0.92) and unsatisfactory for Bergarouge (RMSECV = 24%). TA was predicted with RMSECV-values between 0.79 and 2.61 g 100 ml(-1) and R-values between 0.73 and 0.97. In a second application, near infrared spectra were used to classify apricot fruits according to their variety and colour intensity with correct efficiency. The results obtained in the present study showed that NIRS technology could be applicable to apricot quality and that such portable devices could help to obtain a complete follow-up of the fruits in orchards and during post-harvest. (C) 2008 Elsevier Ltd. All rights reserved.