Detection of Cutting Tool Breakages in CNC Machining Centers Using Image Processing Method

In this research, a novel method has been developed to identify cutting tool breakages, utilizing image processing and establishing the required electronic infrastructure rather than relying on sensor or switch-based systems. So, drills and taps of various sizes in the tool magazine can be determined without any limitation of sensors/switches whether they are broken or not. The results of the classification performance on the dataset comprising 6.8 mm and 8.5 mm diameter drills, as well as Metric 8-sized taps, are highly promising. For the 6.8 mm and 8.5 mm drills, the models achieved perfect sensitivity, specificity, accuracy, and F1 score, indicating their exceptional capability to accurately distinguish between intact and broken tools. Similarly, for the Metric 8-sized tap, the models demonstrated outstanding performance with a sensitivity of 99.14%, specificity of 100%, accuracy of 99.74%, and F1 score of 99.57%. These results underscore the effectiveness of the developed models in tool condition monitoring, showcasing their potential for practical implementation in real-world machining scenarios.