Chemistry of Street\nArt: Neural Network for the Spectral\nAnalysis of Berlin Wall Colors

This research starts with the analysis of some fragments\nof the\nBerlin Wall street art for the characterization of the painting materials.\nThe spectroscopic results provide a general description of the paint\nexecutive technique but more importantly open the way to a new advantage\nof Raman application to the analytic analysis of acrylic colors. The\nstudy highlights the correlation between peak intensity and compound\npercentage and explores the powerful application of deep learning\nfor the quantification of a pigment mixture in the acrylic commercial\nproducts from Raman spectra acquired with hand-held equipment (BRAVO\nby Bruker). The study reveals the ability of the convolutional neural\nnetwork (CNN) algorithm to analyze the spectra and predict the ratio\nbetween the coloring compounds. The reference materials for calibration\nand training were obtained by the dilution of commercial acrylic colors\ncommonly practiced by street artists, using Schmincke brand paints.\nFor the first time, Raman investigation provides valuable insights\ninto calibrations for determining dye dilution in mixtures of commercial\nproducts, offering a new opportunity for analytical quantification\nwith Raman hand-held spectrometers and contributing to a comprehensive\nunderstanding of artists’ techniques and materials in street\nart.