In order to combine feature extraction operations with specific hyperspectral remote sensing information processing objectives,two aspects of feature extraction were explored. Based on clustering and decision tree alg...In order to combine feature extraction operations with specific hyperspectral remote sensing information processing objectives,two aspects of feature extraction were explored. Based on clustering and decision tree algorithm,spectral absorption index (SAI),continuum-removal and derivative spectral analysis were employed to discover characterized spectral features of different targets,and decision trees for identifying a specific class and discriminating different classes were generated. By combining support vector machine (SVM) classifier with different feature extraction strategies including principal component analysis (PCA),minimum noise fraction (MNF),grouping PCA,and derivate spectral analysis,the performance of feature extraction approaches in classification was evaluated. The results show that feature extraction by PCA and derivate spectral analysis are effective to OMIS (operational modular imaging spectrometer) image classification using SVM,and SVM outperforms traditional SAM and MLC classifiers for OMIS data.展开更多
In order to differentiate regions, varieties, and parts of tobacco leaves, two pattern recognition methods through pattern classification modeling were developed based on the comprehensive information of ultraviolet-v...In order to differentiate regions, varieties, and parts of tobacco leaves, two pattern recognition methods through pattern classification modeling were developed based on the comprehensive information of ultraviolet-visible spectroscopy （UV-VIS） by employing one-way analysis of variance （ANOVA1） and wave range random combination （WRRC） technology from MATLAB. This proposed classification method has never been reported previously and the instrument and operation for this method is much more convenient and efficient than previous reported classification methods. The result of this paper demonstrated that the spectral features extracted by ANOVAI and WRRC methods could be used to differentiate tobacco leaves with different patterns. The ANOVAI method had a training recognition rate range of 75.00-87.50%,4 and a validation recognition rate range of 57.14-100%. The WRRC method had a training recognition rate range of 75.00-94.12% and a validation recognition rate range of 66.67-100%. The ANOVAI method is more convenient and efficient in model developing, while the WRRC method utilizes fewer model variables and is more robust.展开更多
基金Projects 40401038 and 40871195 supported by the National Natural Science Foundation of ChinaNCET-06-0476 by the Program for New Century Excellent Talents in University20070290516 by the Specialized Research Fund for the Doctoral Program of Higher Education
文摘In order to combine feature extraction operations with specific hyperspectral remote sensing information processing objectives,two aspects of feature extraction were explored. Based on clustering and decision tree algorithm,spectral absorption index (SAI),continuum-removal and derivative spectral analysis were employed to discover characterized spectral features of different targets,and decision trees for identifying a specific class and discriminating different classes were generated. By combining support vector machine (SVM) classifier with different feature extraction strategies including principal component analysis (PCA),minimum noise fraction (MNF),grouping PCA,and derivate spectral analysis,the performance of feature extraction approaches in classification was evaluated. The results show that feature extraction by PCA and derivate spectral analysis are effective to OMIS (operational modular imaging spectrometer) image classification using SVM,and SVM outperforms traditional SAM and MLC classifiers for OMIS data.
文摘In order to differentiate regions, varieties, and parts of tobacco leaves, two pattern recognition methods through pattern classification modeling were developed based on the comprehensive information of ultraviolet-visible spectroscopy （UV-VIS） by employing one-way analysis of variance （ANOVA1） and wave range random combination （WRRC） technology from MATLAB. This proposed classification method has never been reported previously and the instrument and operation for this method is much more convenient and efficient than previous reported classification methods. The result of this paper demonstrated that the spectral features extracted by ANOVAI and WRRC methods could be used to differentiate tobacco leaves with different patterns. The ANOVAI method had a training recognition rate range of 75.00-87.50%,4 and a validation recognition rate range of 57.14-100%. The WRRC method had a training recognition rate range of 75.00-94.12% and a validation recognition rate range of 66.67-100%. The ANOVAI method is more convenient and efficient in model developing, while the WRRC method utilizes fewer model variables and is more robust.
