激光与光电子学进展, 2011, 48 (10): 101002, 网络出版: 2011-08-22
高光谱的有监督Isomap-SVM苹果粉质化分类
Apple Mealiness Detection Using Supervised Isometric Feature Mapping and Support Vector Machine Based on Hyperspectral Scattering Image
光谱 高光谱散射图像技术 有监督等距映射 支持向量机 非线性降维 BP神经网络 spectroscopy hyper-spectral scattering technology supervised isometric feature mapping support vector machine nonlinear dimension reduction BP neural network model
摘要
苹果粉质化程度是衡量其内部品质的一个重要因素,粉质化造成苹果质量的降低以及商业价值的贬值。高光谱图像技术结合了光谱技术和图像技术的优点,能够无损检测苹果内部品质。提出了有监督等距映射(S-Isomap)和支持向量机(SVM)相结合的用于检测苹果粉质化的新分类方法。S-Isomap-SVM分类方法首先用S-Isomap对高光谱数据作非线性降维,再用SVM对降维后的数据进行分类。对于未知类别的测试样本,采用BP神经网络建模输出的方法,而后结合SVM得到对应的测试精度。这里将S-Isomap-SVM分类方法与SVM以及Isomap-SVM分类方法比较。结果表明,对高光谱数据而言,用S-Isomap-SVM得到的检测精度最高。
Abstract
Apple mealiness is a symptom of internal fruit disorder. Mealiness degrades the quality of apples and reduces their commercial value. Hyperspectral scattering, as a promising technique, combines the advantages of spectroscopy technology and image technology, and can make noninvasive measurement of apple mealiness. A supervised isometric feature mapping (S-Isomap) coupled with support vector machine (SVM) is proposed to detect the mealiness in the apple. S-Isomap is a nonlinear lowering dimension method classifying the dimension reduction of hyperspectral data by SVM. For the unknowned category of the test samples, BP neural network model combined with SVM is used to get the corresponding testing precision. The classification results from S-Isomap-SVM are compared with those obtained using the traditional SVM and Isomap-SVM. The results show that the accuracy of the calibration models obtained with the S-Isomap is higher than that of others.
赵桂林, 朱启兵, 黄敏. 高光谱的有监督Isomap-SVM苹果粉质化分类[J]. 激光与光电子学进展, 2011, 48(10): 101002. Zhao Guilin, Zhu Qibing, Huang Min. Apple Mealiness Detection Using Supervised Isometric Feature Mapping and Support Vector Machine Based on Hyperspectral Scattering Image[J]. Laser & Optoelectronics Progress, 2011, 48(10): 101002.