1. Key Laboratory of Environmental Optics & Technology, Anhui Institute of Optics and Fine Mechanics, Chinese Academy of Sciences, Hefei 230031, China 2. Departments of Mathematics, Hefei Normal University, Hefei 230061, China
Abstract：The analysis of multi-component three-dimensional fluorescence overlapping spectra is always very difficult. In view of the advantage of differential spectra and based on the calculation principle of two-dimensional differential spectra, the three-dimensional fluorescence spectra with both excitation and emission spectra is fully utilized. Firstly, the excitation differential spectra and emission differential spectra are respectively computed after unfolding the three-dimensional fluorescence spectra. Then the excitation differential spectra and emission differential spectra of the single component are obtained by analyzing the multi-component differential spectra using independent component analysis. In this process, the use of cubic spline increases the data points of excitation spectra, and the roughness penalty smoothing reduces the noise of emission spectra which is beneficial for the computation of differential spectra. The similarity indices between the standard spectra and recovered spectra show that independent component analysis based on differential spectra is more suitable for the component recognition of three-dimensional fluorescence overlapping spectra.
[1] Harshman R A. UCLA Working Papers in Phonetics, 1970, 16: 1. [2] Carroll J D, Chang J J. Psychometrika, 1970, 35: 283. [3] Bro R, Ph. D. Dissertation D(University of Amsterdam), 1998. [4] CHEN Guo-qing, WU Ya-min, WEI Bo-lin, et al(陈国庆, 吴亚敏, 魏柏林, 等). Acta Physica Sinica(物理学报), 2010, 59: 5100. [5] YAN Yuan, WANG Le-tian, XU Jin-gou, et al(鄢 远, 王乐天, 许金钩, 等). Acta Chimica Sinica(化学学报), 1996, 54: 772. [6] Common P. Signal Processing, 1994, 36(3): 287. [7] Bell A J, Sejnowski T J. Neural Computation, 1995, 7: 1129. [8] Hyvarinen A. Neural Computation, 1997, 9(7): 1483. [9] Makeig S, Bell A J, Jung T P, et al. Advances in Neural Information Processing Systems, 1996, 8: 145. [10] Vigario R, Jousmaki V, Hamalainenet M, et al. Advances in Neural Information Processing Systems, 1998, 10: 229. [11] McKeown M J, Makeig S, Brown G G, et al. Humman Brain Mapping, 1998, 6: 160. [12] Bartlett M S, Movellan J R, Sejnowski T J. IEEE Transactions on Neural Networks, 2002, 13(6): 1450. [13] Back A D, Weigend A S. International Journal of Neural Systems, 1997, 8: 473. [14] Green P J, Silverman B W. 1994 Nonparametric Regression and Generalized Linear Models. A Roughness Penalty Approach(1st Ed.) (New York: Chapman and Hall).
