Potable NIR spectroscopy predicting soluble solids content of pears based on LEDs
A portable near-infrared (NIR) instrument was developed for predicting soluble solids content (SSC) of pears equipped with light emitting diodes (LEDs). NIR spectra were collected on the calibration and prediction sets (145:45). Relationships between spectra and SSC were developed by multivariate linear regression (MLR), partial least squares (PLS) and artificial neural networks (ANNs) in the calibration set. The 45 unknown pears were applied to evaluate the performance of them in terms of root mean square errors of prediction (RMSEP) and correlation coefficients (r). The best result was obtained by PLS with RMSEP of 0.62°Brix and r of 0.82. The results showed that the SSC of pears could be predicted by the portable NIR instrument.
Near-infrared spectroscopy Artificial neural network Soluble solids content Partial least squares Portable LEDs
Yande Liu Wei Liu Xudong Sun Rongjie Gao Yuanyuan Pan Aiguo Ouyang
School of Mechatronics Engineering, East China Jiaotong University, Changbei Open and Developing District, Nanchang, 330013, China
国际会议
武汉
英文
1-11
2010-11-02（万方平台首次上网日期，不代表论文的发表时间）