The multi-resolution capability of Tchebichef moments and its applications to the analysis of fluorescence excitation-emission spectra

Abstract

Fluorescence spectroscopy with an excitation-emission matrix (EEM) is a fast and inexpensive technique and has been applied to the detection of a very wide range of analytes. However, serious scattering and overlapping signals hinder the applications of EEM spectra. In this contribution, the multi-resolution capability of Tchebichef moments was investigated in depth and applied to the analysis of two EEM data sets (data set 1 consisted of valine-tyrosine-valine, tryptophan-glycine and phenylalanine, and data set 2 included vitamin B1, vitamin B2 and vitamin B6) for the first time. By means of the Tchebichef moments with different orders, the different information in the EEM spectra can be represented. It is owing to this multi-resolution capability that the overlapping problem was solved, and the information of chemicals and scatterings were separated. The obtained results demonstrated that the Tchebichef moment method is very effective, which provides a promising tool for the analysis of EEM spectra. It is expected that the applications of Tchebichef moment method could be developed and extended in complex systems such as biological fluids, food, environment and others to deal with the practical problems (overlapped peaks, unknown interferences, baseline drifts, and so on) with other spectra.

New Constructed EEM Spectra Combined with N-PLS Analysis Approach as an Effective Way to Determine Multiple Target Compounds in Complex Samples

“…The attractive merit of the “second-order advantage” of these methods makes sure that multiple target compounds can be accurately quantified even in the presence of unknown interferences [ 14 , 15 ]. Moreover, image moments such as the Zernike moment [ 16 ], Wavelet moment [ 17 ], Krawtchouk moment [ 18 ], and Tchebichef moment [ 19 ] have also been successfully applied in EEM spectra analysis. Owing to the multiple advantages of the image moments, such as inherent invariance property, image description capability, and multi-resolution capability, the overlapping, scatterings, and other interference signals did not affect the analytical results.…”

The activity prediction of indole inhibitors against HCV NS5B polymerase

“…Several image moment approaches have successfully contributed in the interdisciplinary fields for quantitative analysis, such as 3D spectra of fluorescence probes (Chen et al, ; Li et al, ), complex samples of NIR spectra (Liu et al, ), and multi‐set 1 H NMR (Luca et al, ). Discrete orthogonal moments such as Hahn moments, Krawtchouk moments, and Tchebichef image moments (TMs) have powerful feature encoding ability with no numerical approximation during computation due to their property of orthogonality in the image co‐ordinate space (See et al, ).…”

Morphological Grayscale Reconstruction and ATLD for Recognition of Organic Pollutants in Drinking Water Based on Fluorescence Spectroscopy

“…The steps of detecting organic pollutants in drinking water generally include preprocessing of 3D fluorescence spectrum, feature extraction, and the classification and identification of organic pollutants [16]. Li et al [17] proved that the Tchebichef moment method can independently extract quality information of the target compound by fully using its multi-resolution capability and obtained accurate quantitative analysis results even if severely overlapping signals existed. Nicolas et al [18] investigated the predictability of disinfection by-products (DBPs) with low concentration in drinking water using fluorescence data coupled with neural networks and revealed that this method can improve prediction accuracy.…”