Abstract: As a fundamental aspect of bone fracture treatment, fracture reduction plays a crucial role in restoring the structural integrity and function of bone. At present, fracture reduction techniques mostly rely on semi-automatic interaction methods or healthy side bone templates, which have many limitations in clinical practice. In order to enhance the efficiency and treatment accuracy, an automatic fracture reduction method is proposed. This method utilizes the similarity of fracture cross-sections for registration, thereby reducing the workload of physicians and eliminating the need for the healthy side template. The method utilizes multi-scale techniques to extract potential edge points, maps them onto a surface roughness heat value map, achieves precise extraction of closed edges, and obtains fracture cross-sections based on this approach. During the registration phase, the iterative closest point (ICP) algorithm is highly sensitive to distance. Therefore, the geometric features of point clouds are incorporated into the objective function of the registration algorithm to mitigate the influence of noise. Finally, the algorithm was tested and compared on 180 simulated datasets and 16 publicly available datasets. The results show that the proposed algorithm significantly enhances the registration accuracy, and the registration error of clinical bone fracture cases is controlled within 1.7 mm.