Two-dimensional linear discriminant analysis for classification of three-way chemical data

Quality evaluation and traceability of Bletilla striata by fluorescence fingerprint coupled with multiway chemometrics analysis

“…U‐PLS‐DA works similarly to what was described for PLS‐DA, but unlike this it allows one to operate with data with orders higher than one . U‐PLS‐DA does not work with trilinear data, but it unfolded to a bidimensional array.…”

“…U-PLS-DA works similarly to what was described for PLS-DA, but unlike this it allows one to operate with data with orders higher than one. 27 U-PLS-DA does not work with trilinear data, but it unfolded to a bidimensional array. The unfolding operation employed in U-PLS-DA consists of concatenating the rows of each data matrix X (m × n) into a row vector (1 × nm).…”

A three-dimensional discriminant analysis approach for hyperspectral images

“…However, for "between-image" analysis, when multiple images/samples are compared, the unfolding process might affect the variance structure of the data, once the relationship between neighboring pixels is lost. Some strategies to deal with 2D and 3D arrays without unfolding have been reported, e.g., da Silva et al 22…”

A Novel Strategy for Identifying Oil Pollutants Based on Excitation-Emission Matrix Fluorescence Spectroscopy and Zernike Moments

“…Recently, image processing algorithms was widely employed to extract features from EEMF for the classification [14]- [16], and have proven to be advantageous in three-way data analysis [17]. Image processing algorithms such as two-dimensional linear discriminant analysis (2D-LDA) [14] and two-dimensional principal component analysis (2D-PCA) [17] directly perform discriminant feature analysis on image matrices rather than vectors. Consequently, low computation costs are achieved and the dimension of data matrices can be effectively reduced by using image processing algorithms [15].…”