Defect Extraction in Photomask Image Using High Order Moment

This paper describes an automated technique to extract defects in photomask images. Conventional approaches to photomask inspection rely on difference based techniques which are susceptible to distortion from image `noise´. We propose a robust method based on high order moment (HOM) between reference and test images. In comparison to difference based methods, HOM reveals a wide distribution range of pixel values. This result implies that we can specify pixel value thresholds, and use these thresholds as a basis for distinguishing defects from the background. Our proposed algorithm guarantees the accurate extraction of defects. In this paper, we compare and contrast the transmitted reference image with the test image and reflected image pairs.