<p>Industry 4.0 efforts consistently aim to create and utilize valuable data to automate manufacturing processes, reduce downtime, minimize subjectivity, and enhance overall efficiency. In this study a fully automated method for Tool Condition Monitoring (TCM) is introduced where through a novel approach a tool’s 2D image profile, captured by a camera, is transformed into a unique 1D signal that represents its projected area as a function of the angle of rotation. Through signal processing methods and leveraging the observation that most tools display distinct geometric periodicity (e.g., repeating two, three, four, six or eight times), this 1D signal is segmented into distinct parts exhibiting these geometric periodic patterns. Through statistical comparison, if these segments violate a dynamically determined threshold, the algorithm is able to autonomously classify the tool’s condition. The method’s principles were validated on a variety of tools, including end mills, tappers, drills, and chamfers, under intact and fractured conditions. The method’s capabilities and limitations are also discussed. This approach contributes to the advancement of smart manufacturing by introducing a new data representation for TCM that enables fully automated in-cycle monitoring (e.g. during a tool change or after a part is completed) and decision-making without the need for human intervention.</p>