KMS Chongqing Institute of Green and Intelligent Technology, CAS
Distinguishing different cultivar kiwifruits by near-infrared spectroscopy Based on pattern recognition methods | |
Liu, Bin1,2; Tang, Mingjie2,3; Lu, Liming1; Lin, Yaling1; Cai, Jianrong2; Ma, Teng1 | |
2015 | |
摘要 | An approach that used to distinguish four kinds of different cultivar kiwifruits by near-infrared spectroscopy was proposed in this paper. Original spectra of kiwifruit in wavelength range of 10000-4000/cm were acquired and pre-processed by standard normal vitiate transformation (SNV), and principle component analysis (PCA) was employed to extract useful information. To develop a robust discrimination model, three pattern recognition algorithms (i.e. K-nearest neighbours, artificial neural network, and support vector machine) were attempted comparatively in this work. Some parameters of the model were optimized by cross-validation in building model. Experimental results showed that the performance of support vector machine model is the best in contrast to K-nearest neighbours and artificial neural network models. The optimal support vector machine model was obtained with identification rates of 98.5% in calibration set and 97.5% in prediction set respectively. Based on the results, it was concluded that the near-infrared spectroscopy combined with supervised pattern recognition has significant potential in kiwifruit species detection. |
发表期刊 | International Agricultural Engineering Journal |
ISSN | 08582114 |
卷号 | 24期号:3页码:61-67 |
通讯作者 | Cai, Jianrong |
收录类别 | EI |
语种 | 英语 |