Title :
Non-contact measurement of droplets concentration on the leaf based on Vis/NIR spectroscopy
Author :
Qiu Baijing ; Yin Lei
Author_Institution :
Key Lab. of Modern Agric. Equip. & Technol., Jiangsu Univ., Zhenjiang, China
Abstract :
A rapid, real-time, non-contact and nondestructive way to measure droplets on the leaf was put forward based on Vis/NIR spectroscopy, and the relationship between nondestructive Vis/NIRS measurement and droplets concentration was developed. The multivariable analysis concluding multiple linear regression, stepwise regression, principal component analysis and partial least squares(PLS) was applied to build the regression models, by comparision the accuracy of prediction models, selected the best models, The calibration model was estimated through the correlation coefficient (r) and the root mean square error of cross validation (RMSECV). Finally, partial least squares (PLS) was applied to establish the optimal model, the correlation coefficient of calibration set was 0.998, the root mean square error of cross-validation of calibration set was 0.102.
Keywords :
drops; infrared spectra; infrared spectroscopy; least squares approximations; mean square error methods; organic compounds; principal component analysis; regression analysis; visible spectra; visible spectroscopy; PLS; RMSECV; Vis-NIR spectroscopy; correlation coefficient; cross validation; droplets concentration; multiple linear regression; multivariable analysis; noncontact measurement; nondestructive way; partial least squares; principal component analysis; root mean square error; stepwise regression; Analytical models; Calibration; Correlation; Predictive models; Root mean square; Smoothing methods; Spectroscopy; Near infrared spectroscopy; Partial least squares (PLS); droplets;
Conference_Titel :
Fuzzy Systems and Knowledge Discovery (FSKD), 2011 Eighth International Conference on
Conference_Location :
Shanghai
Print_ISBN :
978-1-61284-180-9
DOI :
10.1109/FSKD.2011.6020019