Expert Knowledge-based Segmentation Algorithm for IC Layout SEM Images

This paper will introduce a novel image processing algorithm based on expert knowledge of SEM (Scanning Electron Microscope) image segmentation of integrated circuit (IC) layouts. Our approach incorporates chip design phase-derived features and SEM microscope image characteristics, enabling high-accuracy and efficient image segmentation. To demonstrate the effectiveness of our approach, we present a comprehensive evaluation of our approach through a series of real analysis scenarios. We conduct a comparative analysis of our proposed algorithm against existing state-of-the-art (SOTA) methodologies for this task. By doing so, we will highlight the advantages of our approach and demonstrate its potential to enhance the accuracy of IC layout segmentation, streamline the verification process, and enhance confidence in the outcomes.