基于荧光高光谱图像的柑桔糖度无损检测  

Hyperspectral Laser-induced Fluorescence Imaging for Nondestructive Assessing Soluble Solids Content of Orange

作　　者：喻晓强[1] 刘木华[2] 郭恩有[2] 杨勇[2] 

机构地区：[1]江西农业大学农学院,江西南昌330045 [2]江西农业大学工学院,江西南昌330045

出　　处：《安徽农业科学》2007年第36期11807-11808,共2页Journal of Anhui Agricultural Sciences

基　　金：国家自然科学基金项目(30460059)

摘　　要：采用632nm的连续波激光作为激发光,应用激光诱导荧光高光谱成像技术对柑桔的糖度值进行无损测量。先将该激光照射到南丰蜜桔和脐橙样品上,后用高光谱图像采集系统收集诱导出的荧光散射图像;接下来对荧光散射图像进行分析,选取100×50像素的荧光区域作为感兴趣区域(ROIs);再提取感兴趣区域在波长700～1000nm的光谱值作为荧光高光谱图像数据;最后用线性回归方法建立荧光高光谱图像数据预测果实糖度值的预测模型。结果表明,该模型预测柑桔糖度值的相关系数分别为南丰蜜桔的R=0.970,脐橙的R=0.960。因此可以看出,应用激光诱导荧光高光谱成像对柑桔糖度值进行无损检测是一种很有效的方法。Laser-induced fluorescence imaging is a promising technique for assessing quality of fruit.In this paper a hyperspectral laser-induced fluorescence imaging technique for measurement of laser-induced fluorescence from orange for predicting soluble solids content（SSC） of fruit was reported.A continuous wave laser（632 nm） was used as an excitation source for inducing fluorescence in oranges.Fluorescence scattering images were acquired from ＇Nanfeng＇ orange and navel orange by a hyperspectral imaging system at the instance of laser illumination.Subsequent analysis of Fluorescence scattering images consisted in selecting regions of interest（ROIs） of pixels,and ROIs were segment around the laser illumination point from Fluorescence scattering images.The hyperspectral fluorescence image data in the wavelength range of 700～1 000 nm were represented by mean grey value of the ROIs.The fruit soluble solids content were measured using hand-held refractometer.A line regressing method was used for developing prediction models to predict fruit soluble solids content.Excellent predictions were obtained for soluble solids content with the correlation coefficient of prediction of R=0.970（Nanfeng orange） and R=0.960（navel orange）.The results show that hyperspectral laser-induced fluorescence imaging is a very good method for nondestructive assessing soluble solids content of orange.

关 键 词：高光谱图像 激光诱导荧光 无损检测 柑桔 糖度 

分 类 号：S666[农业科学—果树学]