Convolutional neural network for self-mixing interferometric displacement sensing

Self-mixing interferometry is a well established interferometric measurement technique. In spite of the robustness and simplicity concept, interpreting self-mixing signal often complicated in practice, which detrimental to availability. Here we discuss use convolutional neural network reconstruct displacement target from semiconductor laser. The network, once trained on periodic patterns, can arbitrarily complex different alignment conditions setups. approach validated here amenable generalization modulated schemes or even totally sensing tasks.