In this paper, we propose a method for inspecting machining chatter marks on automotive valve seats using image processing technique. Chatter or self-excited tool-workpiece relative vibration is unacceptable in the machining processes of valve seats, because it causes poor surface finish and can create internal leaks in the engine. Robust identification of chatter marks on machined surfaces is one of the main challenges in the realization of an automatic in-line inspection system. We suggest several methods for calculating the chatter marks severity, using circle fitting and gray level co-occurrence matrix technique, which was for the first time implemented on circular images.