{"references": ["C. Martins, J. Carvalho, and M. Ferreira, \"CH and C2 radicals characterization\nin natural gas turbulent diffusion flames,\" Journal of the Brazilian\nSociety of Mechanical Sciences and Engineering, vol. 27, no. 2, pp. 110-\n118, 2005.", "N. Docquier, S. Belhalfaoui, F. Lacas, N. Darabiha, and C. Rolon, \"Experimental\nand numerical study of chemiluminescence in methane/air\nhigh-pressure flames for active control applications,\" Proceeding of the\ncombustion institute, vol. 28, pp. 1765-1774, 2000.", "G. Zizak, \"Flame emission spectroscopy: Fundamentals and applications,\"\nInstituto per la Tecnologia dei Materiali e dei Processi Energitici,\nTech. Rep., 2000, iCS training course on laser diagnostics of combustion\nprocesses, NILES, University of Cairo, Egypt.", "L. Arias, S. Torres, D. Sbarbaro, and P. Ngendakumana, \"On the\nspectral band measurements for combustion monitoring,\" Combustion\nand Flame, vol. 158, pp. 423-433, 2011.", "G. Schulze, A. Jirasek, M. Yu, A. Lim, R. Turner, and M. Blades,\n\"Investigation of selected baseline removal techniques as candidates for\nautomated implementation,\" Appl. Spectrosc., vol. 59, no. 5, pp. 545-\n574, 2005.", "A. G. Gaydon and H. G. Wolfhard, The spectroscopy of flames, 1st ed.\nChapman and Hall LTD., 1957.", "P. Ngendakumana, B. Zuo, and E. Winandy, \"A spectroscopic study of\nflames for a pollutant formation regulation in a real oil boiler,\" Proc. 2th\nInt-l Conf. on Tech. and Combustion for a Clean Environment, 1993.", "Y. Kojima, Y. Ikeda, and T. Nakajima, \"Basic aspect of oh(a), ch(a) and\nc2(d) chemiluminiscence in the reaction zone of laminar methane-air\npremixed flames,\" Combustion and Flame, vol. 140, pp. 34-45, 2004.", "Y. Hardalupas, M. Orain, C. Panoutsos, A. Taylor, J. Olofsson,\nH. Seyfried, M. Richter, J. Hult, M. Alden, F. Hermann, and J. Klingmann,\n\"Chemiluminescence sensor for local equivalence ratio of reacting\nmixtures of fuel and air (flameseek),\" Applied Thermal Engineering,\nvol. 24, pp. 1619-1632, 2004.\n[10] L. Shao and P. Griffiths, \"Automatic baseline correction by wavelet transform\nfor quantitative open-path fourier transform infrared spectroscopy,\"\nEnviron. Sci. Technol., vol. 41, no. 20, pp. 7054-7059, 2007.\n[11] M. Lopez, J. Hern'andez, E. Valero, and J. Romero, \"Selecting algorithms,\nsensors, and linear bases for optimum spectral recovery of\nskylight,\" J. Opt. Soc. Am., vol. 24, no. 4, pp. 942-956, 2007.\n[12] P. Courrieu, \"Fast computation of moore-penrose inverse matrices,\"\nNeural Information Processing, vol. 8, no. 2, pp. 25-29, 2005.\n[13] F. Imai, M. Rosen, and R. Berns, \"Comparative study of metrics for\nspectral match quality,\" CGIV First European Conference on Colour\nGraphics, Imaging, and Vision, pp. 492-496, 2002.\n[14] M. Lopez, J. Hern'andez, E. Valero, and J. Nieves, \"Colorimetric and\nspectral combined metric for the optimization of multispectral systems,\"\nProceeding of the 10th Congress of the International Colour Association\n(AIC05), pp. 1685-1688, 2005.\n[15] J. L. Nieves, E. M. Valero, J. Hernandez, and J. Romero, \"Recovering\nfluorescent spectra with an rgb digital camera and color filters using\ndifferent matrix factorizations,\" Appl. Opt., vol. 46, no. 19, pp. 4144-\n4154, 2007."]}

In this paper, an automated algorithm to estimate and remove the continuous baseline from measured spectra containing both continuous and discontinuous bands is proposed. The algorithm uses previous information contained in a Continuous Database Spectra (CDBS) to obtain a linear basis, with minimum number of sampled vectors, capable of representing a continuous baseline. The proposed algorithm was tested by using a CDBS of flame spectra where Principal Components Analysis and Non-negative Matrix Factorization were used to obtain linear bases. Thus, the radical emissions of natural gas, oil and bio-oil flames spectra at different combustion conditions were obtained. In order to validate the performance in the baseline estimation process, the Goodness-of-fit Coefficient and the Root Mean-squared Error quality metrics were evaluated between the estimated and the real spectra in absence of discontinuous emission. The achieved results make the proposed method a key element in the development of automatic monitoring processes strategies involving discontinuous spectral bands.