Machine Vision-based Tool Location and Wear Monitoring in Micro turning: Optical Setup and Image Processing Approaches

In micro turning, tool placement, wear, and breakage are crucial as they affect precision and surface finish. Flank wear is problematic since it impacts dimensional accuracy, and current methods that rely on CNC encoder inputs and radius compensation are insufficient for micro turning due to the small tool sizes involved. Detecting flank wear in micro turning is difficult as the tools are often too small for visual inspection, making monitoring at micron levels a challenge. This research investigates using machine vision to monitor tool position and flank wear in micro turning. The optical system, comprising a 12X super zoom lens and 3MP CMOS image sensor, is assessed alongside three lighting techniques: direct, indirect, and backlighting. Preliminary image analysis, using Gaussian filters and histogram equalization, indicates lighting's importance. Early results, comparing new and worn tools using exclusive operators, are promising, but further image enhancements are necessary for accurate classification and wear detection.