Detection of fracture in ceramic cutting tools from workpiece profile signature using image processing and fast Fourier transform

• An effective method of detecting fracture in ceramic tools from workpiece profile signature is proposed.
• FFT analysis on the sub-pixel edge profile of workpiece surface was performed.
• From the frequency spectrum the amplitude of fundamental frequency was found to fluctuate rapidly after tool fracture.
• The random fluctuations in the amplitude can be easily distinguished from the gradual increase observed during gradual wear.

A non-contact vision-based method to detect the occurrence of fracture in ceramic cutting tool inserts using the workpiece profile signature is proposed. Machining experiments were carried out to turn stainless steel using ceramic cutting tool inserts. The images of the workpiece profile were captured after each turning pass using a high-resolution DSLR camera. The edge profiles of the workpiece were extracted to sub-pixel accuracy using the invariant moment method. The extracted profiles were transformed from the spatial domain to the frequency domain using Fast Fourier Transform (FFT). From the Fourier spectrum the amplitude of the fundamental feed frequency was observed to increase steadily with the cutting duration during gradual wear of the tool edge. However, significant fluctuations in the amplitude of the fundamental feed frequency were observed after the onset of chipping and fracture due to the continuous deterioration of the cutting edge. This was caused by the irregular peak-to-valley heights in the workpiece surface profile resulting from the fractured tool at the end of the cutting time of 84.8 s.