English：

Machine vision displacement monitoring technology often requires the pre installation of artificial markers, and the installation of artificial markers is not only limited by multiple factors such as site environment, construction cost, and structural surface characteristics, but may even be difficult to achieve under certain harsh or special working conditions. Therefore, this study conducted research on markerfree machine vision displacement monitoring methods. Specifically, a lowillumination environment markerfree machine vision displacement monitoring method combining homomorphic filtering image enhancement algorithm and RGDIC is proposed to address issues such as missing image details, insufficient brightness, and difficulty in stably extracting and accurately tracking natural texture features of target structure surfaces in lowillumination environments. To evaluate the effectiveness of the proposed method, lowillumination environment displacement monitoring experiments were conducted, and active markers were set as benchmarks for verification. The results showed that the root mean square errors of horizontal and vertical displacement of natural markers and active markers were 033pixel and 035pixel, respectively, and the displacement of the two showed a highly consistent dynamic change pattern, with correlation coefficients higher than 075. In summary, the markerfree machine vision displacement monitoring method can maintain stable recognition accuracy and monitoring reliability without artificial markers, and can effectively overcome the adverse effects of lowillumination environments, providing a feasible solution for expanding the application of marker free machine vision monitoring technology in low light conditions.