Enhanced upconversion photoluminescence of LiYF4:Yb3+/Tb3+ crystals by introducing Mn2+ ions for efficient anti-counterfeiting recognition based on convolution neural network

A series of LiYF4:Yb3+/Tb3+ upconversion micro-particles (UCMPs) with varying Mn2+ doping concentration were prepared and examined. The composition, morphology and luminescent properties of the synthesized micro-particles was analysed by XRD, SEM, EDS and other characterization methods. Notably the significant increase of UC photoluminescence UCMPs by introducing Mn2+ ions was obtained by 980 nm laser excitation and reached the highest value as the Mn2+ ions doping concentration was 5 mol%. Furthermore, the produced UCMPs used as screen-printing materials of anti-counterfeiting ink were used for anti-counterfeiting recognition labels reliably. Moreover, based on Convolutional Neural Network (CNN), an anti-counterfeiting image recognition system was designed. The relevant research results indicate that the accuracy of the designed anti-counterfeiting image recognition system has reached 94 %, which also provides important reference significance for the application of CNN in the field of anti-counterfeiting recognition.