The invention discloses a multi-layer heterogeneous
coating laser layering dynamic removal method and
system based on spectral characteristics and a medium, and belongs to the technical field of
laser cleaning,
spectrograph data of different coatings in
laser paint removal are collected, and a cross-
coating data set is obtained through division; based on the
spectrograph data of the different coatings, fitting and determining
Gaussian fitting curves of characteristic elements of the different coatings so as to identify the information of the currently processed
coating; training a
deep learning model based on the cross-coating
data set to obtain a cross-coating prediction model so as to predict early warning information about whether the currently processed coating is about to cross the coating; based on the coating information,
laser processing parameters of the corresponding number of coating
layers are configured; and if the early warning information that the coating is about to cross the coating is predicted, the
laser processing parameters of the next layer are dynamically configured based on the early warning information, so that accurate layered removal is realized, and the damage to the substrate or the lower coating is avoided. According to the invention, real-time self-
adaptive control of laser process parameters is efficiently realized, and the method has good practicability.