Multi-dimensional wear parameter detection of milling cutters using microscopic image sequence reconstruction

In precision machining, the geometry of the milling cutter plays a critical role in determining the quality of the milled surface. Detecting cutter wear is essential for maintaining machining quality. However, traditional measurement methods, primarily relying on signals and images, are inadequate for direct assessment of the wear area owing to the cutter's complex structure. This study introduces a 3D reconstruction method based on a sequence of microscopic images. To improve the measurement of milling tool morphology, a multi-partition Laplace focus evaluation operator is proposed to determine the optimal focus position of the images. Furthermore, a Gaussian fitting method with hierarchical window multi-scale fusion is implemented to efficiently identify the optimal focus position. Compared to measurements obtained using a Sensofar microscope, the standard deviation of the reconstructed wear area morphology is less than 1.65 %, and that of the milling cutter diameter is less than 0.010 mm. These findings confirm the accuracy of the proposed method in evaluating milling cutter wear.