脐橙糖度的高光谱图像无损检测技术  

Nondestructive Detection of Sugar Content on Navel Orange with Hyperspectral Imaging

作　　者：郭恩有[1] 刘木华[1] 赵杰文[2] 陈全胜[2] 

机构地区：[1]江西农业大学工学院,南昌市330045 [2]江苏大学生物与环境工程学院,镇江市212013

出　　处：《农业机械学报》2008年第5期91-93,103,共4页Transactions of the Chinese Society for Agricultural Machinery

基　　金：国家自然科学基金资助项目(项目编号:30460059)

摘　　要：提出了利用高光谱图像系统来检测脐橙糖度的方法。由脐橙反射光谱图像获取反映脐橙糖度的光谱特征波长;应用人工神经网络系统建立了脐橙糖度的预测模型。结果表明,脐橙糖度预测模型相关系数R为0.831,采用高光谱图像无损检测脐橙糖度是可行的。Sugar content is an important quality attribute. The feasibility of using hyperspectral imaging for nondestructive detection of sugar content of navel orange was investigated. The spectrum features wavelength for predicting the navel orange＇s sugar content of the navel orange was obtained via the scatting spectral imaging. Subsequently, artificial neural network system was used for developing a prediction model to predict the sugar content of navel orange. Finally, the prediction results were obtained for sugar content with the correlation coefficient of prediction of R = 0. 831. The results showed that hyperspectral imaging is an effective method for nondestructive assessing the sugar content of navel orange.

关 键 词：脐橙 糖度 高光谱图像 无损检测 

分 类 号：S-03[农业科学] O434.13[机械工程—光学工程]