Aiming at the problem that the image feature information utilized in the existing single-laser visual microdisplacement measurement methods is not rich, which leads to inaccurate measurement results, a dual-laser microdisplacement measurement method combining direct and oblique shooting, together with BP (Back Propagation) neural network is proposed to achieve high-precision measurement of displacement. In this paper, the principles of lens imaging and small-aperture imaging are adopted to theoretically analyze the dual-laser model, and ZEMAX is used to numerically simulate the ranging model to verify the theoretical feasibility and superiority of the proposed method. Secondly, according to the numerical simulation results, the experimental platform is designed and constructed for image acquisition experiments, a series of image features are extracted as the inputs to the BP network, and the displacement prediction with the displacement parameters are used as the outputs to construct the displacement prediction model. The experimental results show that the dual-laser displacement model proposed in this paper has higher measurement accuracy compared with the single-laser model, and the measurement accuracy reaches more than 99% after the introduction of BP neural network. This paper provides a new method and new ideas for the high-precision measurement of tiny displacement.