This paper proposes a novel wavelet-based face recognition method using thermal infrared (IR) and visible-light face images. The method applies the combination of Gabor and the Fisherfaces method to the reconstructed ...This paper proposes a novel wavelet-based face recognition method using thermal infrared (IR) and visible-light face images. The method applies the combination of Gabor and the Fisherfaces method to the reconstructed IR and visible images derived from wavelet frequency subbands. Our objective is to search for the subbands that are insensitive to the variation in expression and in illumination. The classification performance is improved by combining the multispectal information coming from the subbands that attain individually low equal error rate. Experimental results on Notre Dame face database show that the proposed wavelet-based algorithm outperforms previous multispectral images fusion method as well as monospectral method.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
文摘This paper proposes a novel wavelet-based face recognition method using thermal infrared (IR) and visible-light face images. The method applies the combination of Gabor and the Fisherfaces method to the reconstructed IR and visible images derived from wavelet frequency subbands. Our objective is to search for the subbands that are insensitive to the variation in expression and in illumination. The classification performance is improved by combining the multispectal information coming from the subbands that attain individually low equal error rate. Experimental results on Notre Dame face database show that the proposed wavelet-based algorithm outperforms previous multispectral images fusion method as well as monospectral method.
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.
