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
Feature reduction is a key process in pattern recognition. This paper deals with the feature reduction methods for a time-shift invariant feature, power spectrum, in Radar Automatic Target Recognition (RATR) using Hig...Feature reduction is a key process in pattern recognition. This paper deals with the feature reduction methods for a time-shift invariant feature, power spectrum, in Radar Automatic Target Recognition (RATR) using High-Resolution Range Profiles (HRRPs). Several existing feature reduction methods in pattern recognition are analyzed, and a weighted feature reduction method based on Fisher's Discriminant Ratio (FDR) is proposed in this paper. According to the characteristics of radar HRRP target recognition, this proposed method searches the optimal weight vector for power spectra of HRRPs by means of an iterative algorithm, and thus reduces feature dimensionality. Compared with the method of using raw power spectra and some existing feature reduction methods, the weighted feature reduction method can not only reduce feature dimensionality, but also improve recognition performance with low computation complexity. In the recognition experiments based on measured data, the proposed method is robust to different test data and achieves good recognition results.展开更多
By means of the reaction between a DOTA-NHS-ester bifunctional reagent and N-terminal peptides of proteins, and then che- lation of lanthanide metal ions as tags, we established a novel method for the identification o...By means of the reaction between a DOTA-NHS-ester bifunctional reagent and N-terminal peptides of proteins, and then che- lation of lanthanide metal ions as tags, we established a novel method for the identification of N-terminal peptides of proteins and their relative quantification using metal-element-chelated tags coupled with mass spectrometry. The experimental results indicate that metal elements are able to completely label N-terminal peptides at the protein level. The N-terminal peptides are enriched as the peptides digested with trypsin are selectively eliminated by isothiocyanate-coupled silica beads. We success- fully identified the N-terminal peptides of 158 proteins of Thermoanaerobacter tengcongensis incubated at 55 and 75 ℃, among which N-terminal peptides of 24 proteins are partially acetylated. Moreover, metal-element tags with high molecule weights make it convenient for N-terminal peptides consisting of less than 6 amino acids to be identified; these make up 55 percent of the identified proteins. Finally, we developed a general approach for the relative quantification of proteins based on N-terminal peptides. We adopted lysozyme and ribonuclease B as model proteins; the correlation coefficients （R2） of the standard curves for the quantitative method were 0.9994 and 0.9997, respectively, with each concentration ratio ranging from 0.1 to 10 and both relative standard derivations （RSD） measured at less than 5%. In T. tengcongensis at two incubation tem- peratures, 80 proteins possess quantitative information. In addition, compared with the proteins of T. tengcongensis incubated at 55 ℃, in T. tengcongensis incubated at 75 ℃, 7 proteins upregulate whereas 16 proteins downregulate, and most differential proteins are related to protein synthesis.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.
基金Partially supported by the National Natural Science Foundation of China (No.60302009)the National Defense Advanced Research Foundation of China (No.413070501).
文摘Feature reduction is a key process in pattern recognition. This paper deals with the feature reduction methods for a time-shift invariant feature, power spectrum, in Radar Automatic Target Recognition (RATR) using High-Resolution Range Profiles (HRRPs). Several existing feature reduction methods in pattern recognition are analyzed, and a weighted feature reduction method based on Fisher's Discriminant Ratio (FDR) is proposed in this paper. According to the characteristics of radar HRRP target recognition, this proposed method searches the optimal weight vector for power spectra of HRRPs by means of an iterative algorithm, and thus reduces feature dimensionality. Compared with the method of using raw power spectra and some existing feature reduction methods, the weighted feature reduction method can not only reduce feature dimensionality, but also improve recognition performance with low computation complexity. In the recognition experiments based on measured data, the proposed method is robust to different test data and achieves good recognition results.
基金supported by the National Basic Research Program of China(2012CB910603,2010CB912701)the National Key Scientific Instrument Development Program of China(2011YQ09000504,2012YQ-12004407,2011YQ06008408)+1 种基金the High-Technology Research and Development Programme of China(2012AA020202)the National Natural Science Foundation of China(20735005,21275159,31100591)
文摘By means of the reaction between a DOTA-NHS-ester bifunctional reagent and N-terminal peptides of proteins, and then che- lation of lanthanide metal ions as tags, we established a novel method for the identification of N-terminal peptides of proteins and their relative quantification using metal-element-chelated tags coupled with mass spectrometry. The experimental results indicate that metal elements are able to completely label N-terminal peptides at the protein level. The N-terminal peptides are enriched as the peptides digested with trypsin are selectively eliminated by isothiocyanate-coupled silica beads. We success- fully identified the N-terminal peptides of 158 proteins of Thermoanaerobacter tengcongensis incubated at 55 and 75 ℃, among which N-terminal peptides of 24 proteins are partially acetylated. Moreover, metal-element tags with high molecule weights make it convenient for N-terminal peptides consisting of less than 6 amino acids to be identified; these make up 55 percent of the identified proteins. Finally, we developed a general approach for the relative quantification of proteins based on N-terminal peptides. We adopted lysozyme and ribonuclease B as model proteins; the correlation coefficients （R2） of the standard curves for the quantitative method were 0.9994 and 0.9997, respectively, with each concentration ratio ranging from 0.1 to 10 and both relative standard derivations （RSD） measured at less than 5%. In T. tengcongensis at two incubation tem- peratures, 80 proteins possess quantitative information. In addition, compared with the proteins of T. tengcongensis incubated at 55 ℃, in T. tengcongensis incubated at 75 ℃, 7 proteins upregulate whereas 16 proteins downregulate, and most differential proteins are related to protein synthesis.
