Automatic detection of alignment or registration marks
OwnerShih-Jong J. Lee, Seho Oh +1inventors
InventorsShih-Jong J. Lee, Seho Oh, Larry A. Nelson
Life of this patent
Estimating the term…
Abstract
Mark detection and position determination are improved by use of directional elongated filters, symmetry, gray scale image processing, structural constraints, and learning. Directional elongated filters are used to pre-process images of registration marks to create masks and enhanced images. Working sequentially, portions of the mark are detected and classified. The input gray scale image of the mark is processed using its structural constraints in conjunction with a mask for the detected mark. A cost function estimation determines mark position and orientation with sub-pixel accuracy. Learning is used to improve specific application performance.
Claims21
A method for detecting the mark in an image comprising:
- a. obtaining an image of at least one mark;
- b. locating the center of each mark based on symmetry
- c. processing the mark image using at least one directional elongated filter;
- d. rejecting artifacts based on symmetry
A method of locating a detected mark's position in an image comprising:
- a. creating a gray scale image of at least one mark;
- b. masking portions of the image based upon detected mark elements;
- c. estimating mark position using a structure guided estimation process.
A method of measuring orientation of a mark in an image comprising:
- a. obtaining an image of at least one mark;
- b. locating the center of the each mark along each axis of symmetry;
- c. measuring the mark orientation using a structure-guided estimation process.
A method of learning specific mark structure comprising;
- a. obtaining an image of at least one mark;
- b. filtering the image using at least one directional elongated filter;
- c. determining at least one parameter of the mark through a learning process.
Dependencies are inferred from the claim text.