The Approach of Nanoscale Vision-Based Measurement Via Diamond-Machined Surface Topography

Micro-vision positioning measurement combines optical microscopy and computer vision to achieve noncontact, high-precision positioning measurement. Ultra-precision machined microstructures as targets can achieve low-cost, high-resolution measurement. However, there is an unknown mechanism by which the microscopic topography of microstructure surface affects the measurement resolution, limiting the further improvement of the measurement resolution of this method. From the perspective of physical optical imaging and a template matching algorithm, this paper theoretically analyzes the mechanism of the pixel intensity distribution of different surface topography and the effect of pixel intensity distribution on the measurement resolution, reveals the mechanism of the influence of surface topography on the measurement resolution, and then deduces the most suitable topography for precision measurement, referred to as anchor microstructure. The experimental results demonstrate that the anchor microstructures obtained by ultra-precision cutting are ideal targets for vision positioning measurements, which are theoretically able to achieve a resolution of better than 2 nm. Finally, the positioning performance verification experiments prove that the method is a feasible high-resolution positioning measurement method.