作者: Mike Sissons , Brian Osborne , Shaylene Sissons
DOI: 10.1255/JNIRS.582
关键词: Whole grains 、 Near infrared reflectance spectroscopy 、 Moisture 、 Agronomy 、 Environmental science 、 Cereal grain 、 Plant breeding 、 Near-infrared spectroscopy 、 Black point 、 Calibration
摘要: The breeding of new cereal grain varieties requires methods that are rapid and preferably non-destructive, to assess the quality in early generations. Near infrared (NIR) spectroscopy has great potential meet these requirements for durum wheat breeder. In this study, sample sets representing wheats from an Australian programme were used develop several NIR calibrations. Useful predictive calibrations scanning whole samples developed hectolitre weight, semolina yield, b* hardness, addition highly accurate protein moisture. It was not possible adequately predict dough properties black point percentage scanning. A useful calibration measure browning reaction, applicable generation screening, also developed. These have been extensively validated using large independent sets.