Displacement measurement of micro-nano flexible mechanism based on vision

In this paper, for the problem that micro-visual measurement is more easily interfered with by light and noise, and measurement is easily limited by camera resolution and cost, a measurement algorithm combining polynomial interpolation+Gaussian interpolation edge subpixel algorithm and adaptive improved SIFT algorithm is proposed, which combines the advantages of polynomial interpolation+Gaussian interpolation edge subpixel algorithm with the high robustness and strong anti-interference of feature points and the advantages of adaptive improved SIFT algorithm with high measurement accuracy. This algorithm ensures the precision and speed of measurement while reducing light noise and light influence. At the same time, a BP neural network correction algorithm is designed to correct the results further to compensate for specific effects caused by lens distortion, to make the accuracy higher. Finally, relevant experiments are designed to prove that this method can be applied to micro-displacement visual displacement measurement.