Nondestructive Detection for forecasting the level of acidity and sweetness of apple based on NIR spectroscopy

摘要

Apple plays an important role in modern agricultural economic. With the international trade competition and focus of apple consumption has been transferred from appearance quality to internal quality. Currently, people pay more attention to the potential of near infrared spectroscopy in nondestructive detection of fruits. In this paper, we measure the spectrum, sweetness and acidity of apples, use ISOMAP algorithm, BP neural network and GRNN network to conduct sweetness and acidity prediction of apples. Through data analysis, we could know that it is possible to predict the value of sweetness and acidity of apples through fast and nondestructive detection, and the models perform much better in prediction of acidity than that of in sweetness prediction.