Authors: HONGYUAN TANG, GUIPING LIAO
Abstract: The inconsistency of chlorophyll content in rapeseed directly affects the quality of seed, so it is necessary to establish a rapid and nondestructive detection technology of chlorophyll in mature rapeseed. In this paper, the content of chlorophyll in rapeseed samples was determined by ultraviolet-visible spectrophotometry, and the near-infrared spectrum data were collected. The partial least squares regression model of chlorophyll content in rapeseed was established based on hyperspectral technology through first derivative and standard normal variable transformation pretreatment. The results of the cross-test showed that the determination coefficient of the validation set of the prediction model of chlorophyll content in rapeseed was greater than 0.85, and the root mean square error (RMSE) of the cross-test was less than 0.3 mg/kg. The results showed that the method could accurately predict the chlorophyll content in rapeseed and provide important technical support for rapid monitoring of seed quality.
Keywords: Hyperspectral technology, rapeseed leaf, chlorophyll, rapid detection
Full Text: PDF