NIR sensitive wavelength selection based on different methods

Author

Liu, Xuemei ; Zhang, Hailiang ; Sun, Xudong

Author_Institution

Sch. of Civil Eng. & Archit., East China Jiao-Tong Univ., Nanchang, China

fYear

2010

fDate

26-28 June 2010

Firstpage

2483

Lastpage

2487

Abstract

Near infrared (NIR) sensitive wavelengths of soluble solids content (SSC) were selected based on genetic algorithms (GAs) and interval partial least square (iPLS) in Nanfeng mandarin fruits, the better result was obtained with sensitive wavelengths of 642-787 nm, 861-1152 nm and 1225-1368 nm by iPLS. Then 30 unknown samples were used to evaluate performance of the model built with sensitive wavelengths. The correlation coefficient (R) and root mean square error of prediction (RMSEP) were 0.9500 and 0.5354, respectively. It was demonstrated NIR spectroscopy combined with sensitive wavelengths selection algorithm was a powerful tool to rapidly determine SSC of Nanfeng mandarin fruits nondestructively.

Keywords

agricultural products; food products; genetic algorithms; infrared spectra; least mean squares methods; NIR sensitive wavelength selection; NIR spectroscopy; Nanfeng mandarin fruits; correlation coefficient; genetic algorithms; interval partial least square; near infrared sensitive wavelengths; root mean square prediction error; soluble solid content; wavelength 1225 nm to 1368 nm; wavelength 642 nm to 787 nm; wavelength 861 nm to 1152 nm; Calibration; Infrared spectra; Least squares methods; Performance evaluation; Predictive models; Root mean square; Solids; Spectroscopy; Sun; Wavelength measurement; GAs; NIR; Nanfeng mandarin fruits; SSC; iPLS;

fLanguage

English

Publisher

ieee

Conference_Titel

Mechanic Automation and Control Engineering (MACE), 2010 International Conference on

Conference_Location

Wuhan

Print_ISBN

978-1-4244-7737-1

Type

conf

DOI

10.1109/MACE.2010.5535575

Filename

5535575