Automatic visual inspection methods for product quality checking are spreading more and more in the present 4.0 manufacturing industry. This paper addresses the automatic inspection of edge integrity in accurate holes obtained by direct drilling of cemented carbide with innovative diamond coated tools. These revolutionary cutting tools, recently appeared on the market, can process extremely hard carbide in the sintered state, with massive increase of productivity with respect to standard methodologies like electrical discharge machining (EDM). However, due to the brittleness of the materials, the mechanical cutting process becomes critical and sensitive to tool breakage and workpiece defects generation. In particular, chipping of the hole edges represents one of the most important issues to monitor and take under control. A software procedure, that analyses high-resolution images taken from optical microscopes, was then developed for that aim. Image processing algorithms were designed and applied to enable the automatic extraction of the holes profile, thus permitting the identification and quantification of the leading edge damage in the radial direction. The proposed approach is fully automatic and is based on a profile segmentation that exploits an edge detection algorithm followed by a contour extraction method based on the solution of a partial differential equation. Dedicated metrics were specifically developed to evaluate the extracted profiles. The approach was validated with a factorial plane involving 1.6 mm diameter holes generated with different cutting parameters and tools on tungsten-carbide (WC) material. The technique resulted suitable for the aim, enabling the automatic characterisation of the defects generation phenomenon throughout the entire tools life. This moves a step toward the implementation of both in-line hole inspection procedures and advanced drilling process control.
|Number of pages||11|
|Early online date||7 Sep 2019|
|Publication status||Published - 1 Nov 2019|