Tokyo Electron Limited

University of California

Ultratech, Inc.

Shin-Yee Lu

Ivan Maleev

A method for detecting defects on a sample based on a defect inspection apparatus is provided. In the method, an image data set that includes defect data and non-defect data is organized. A convolutional neural network (CNN) model is defined. The CNN model is trained based on the image data set. The defects on the sample are detected based on inspection data of the defect inspection apparatus and the CNN model. The sample includes uniformly repeating structures, and the inspection data of the defect inspection apparatus is generated by filtering out signals of the uniformly repeating structures of the sample.

Method and apparatus for inspecting pattern collapse defects

A method of measuring machine alignment offset of an optical machine having an alignment system, so that subsequent processing of substrates on set of optical machines can be performed in a machine-independent manner. The optical machine forms overlayed images of first and second patterns formed on either one or two reticles onto a substrate at respective first and second levels. The method of the invention includes forming a virtual zero-offset alignment pattern and a virtual zero-offset metrology pattern and imaging first and second metrology patterns on the substrate at the first and second levels, respectively. The second metrology pattern is aligned to the first metrology pattern using the zero-offset alignment pattern so that the exposures are performed in an overlayed manner. The first and second metrology patterns are based on the virtual zero-offset metrology pattern. An image of the overlayed first and second metrology patterns formed on the substrate is obtained using the alignment system of the optical machine. The virtual zero-offset metrology pattern is compared to corresponding portions of the image of the overlayed metrology patterns to deduce an offset from an idea alignment of the first and second metrology patterns. Zero-offset alignment patterns for one or more jobs may also be created so that the jobs can be run without an extra step of determining the job-dependent offset for each job.

Machine-independent alignment system and method

A three-dimensional motion camera system comprises a light projector placed between two synchronous video cameras all focused on an object-of-interest. The light projector shines a sharp pattern of vertical lines (Ronchi ruling) on the object-of-interest that appear to be bent differently to each camera by virtue of the surface shape of the object-of-interest and the relative geometry of the cameras, light projector and object-of-interest Each video frame is captured in a computer memory and analyzed. Since the relative geometry is known and the system pre-calibrated, the unknown three-dimensional shape of the object-of-interest can be solved for by matching the intersections of the projected light lines with orthogonal epipolar lines corresponding to horizontal rows in the video camera frames. A surface reconstruction is made and displayed on a monitor screen. For 360.degree. all around coverage of theobject-of-interest, two additional sets of light projectors and corresponding cameras are distributed about 120.degree. apart from one another.

Growing community of inventors

Pleasanton, CA, United States of America

Shin-Yee Lu