The Global Patent Exchange

Web: |  Phone: (650)292-4849 |  Email:

Patent for License:

Defect detection using gray level signatures    

A method and apparatus for inspecting isolated features on a reticle wherein the isolated features are analyzed for defects globally as individual, whole features, rather than locally, pixel-by-pixel.


In one embodiment, the reticle is imaged to produce pixels having a gray level, and an isolated feature is identified. An energy value for the isolated feature to be inspected is calculated by summing the gray levels of its pixels. A plurality of scatter values for the feature in different directions in the feature is calculated based on the equilibrium center of the feature to produce a unique "signature" for the inspected feature. The energy and scatter values of the inspected features signature are then compared with energy and scatter values associated with a corresponding reference feature to determine whether a defect exists in the inspected feature.
Summary of Technology
An advantage of the technology is the ability to reliably detect defects in isolated features or otherwise delimited areas without increasing inspection time.
Other advantages are achieved in part by a method of inspecting a target feature formed on a surface, the method comprising imaging the target feature to produce one or more target data elements representative of the target feature, each target data element having a gray level and associated with a respective location on the surface. An energy value for the target feature is calculated by summing the gray levels of the target data elements corresponding to the target feature, an equilibrium center of the target feature is determined, and then a plurality of scatter values, each in a different predetermined direction in the target feature, are calculated.
The target feature is thereafter identified as corresponding to a target reference feature, and the energy and scatter values of the target feature are compared with energy and scatter values in a historical database of previously inspected features associated with the target reference feature to determine whether a defect exists in the target feature.
Another aspect of the technology is an inspection tool for carrying out the steps of the above method.

Patent Summary

U.S. Patent Classes & Classifications Covered in this Patent:

Class 382: Image Analysis

This is the generic class for apparatus and corresponding methods for the automated analysis of an image or recognition of a pattern. Included herein are systems that transform an image for the purpose of (a) enhancing its visual quality prior to recognition, (b) locating and registering the image relative to a sensor or stored prototype, or reducing the amount of image data by discarding irrelevant data, and (c) measuring significant characteristics of the image.

Subclass 144: Mask inspection (e.g., semiconductor photomask)
Subclass 149: Fault or defect detection

View this listing on the Tynax website: