Method of edge detection in optical images using neural network classifier

• US 5,311,600 A
• Filed: 09/29/1992
• Issued: 05/10/1994
• Est. Priority Date: 09/29/1992
• Status: Expired due to Fees

• Current Assignee: Board of Trustees of the Leland Stanford Junior University (Stanford University)
• Original Assignee: Board of Trustees of the Leland Stanford Junior University (Stanford University)
• Inventors: Aghajan, Hamid K., Kailath, Thomas
• Primary Examiner(s): Boudreau, Leo H.
• Application Number:
  US07/953,392
• Time in Patent Office: 588 Days
• Field of Search: 382/22, 382/14, 382/15, 395/21, 395/23
• US Class Current: 382/156
• CPC Class Codes: G06T 2207/20084 Artificial neural networks [ANN]
• CPC Class Codes: G06T 7/13 Edge detection
• CPC Class Codes: G06V 10/44 Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components

Abstract

An image processor employing a camera, frame grabber and a new algorithm for detecting straight edges in optical images is disclosed. The algorithm is based on using a self-organizing unsupervised neural network learning to classify pixels on a digitized image and then extract the corresponding line parameters. The image processor is demonstrated on the specific application of edge detection for linewidth measurement in semiconductor lithography. The results are compared to results obtained by a standard straight edge detector based on the Radon transform; good consistency is observed; however, superior speed is achieved for the proposed image processor. The results obtained by the proposed approach are also shown to be in agreement with Scanning Electron Microscope (SEM) measurements, which is known to have excellent accuracy but is an invasive measurement instrument. The method can thus be used for on-line measurement and control of microlithography processes and for alignment tasks as well.

Claims

1. A method of measurement of linewidths in integrated circuit fabrication comprising the steps of

a) obtaining a digitized pixel image of a surface of said integrated circuit including conductive lines,

b) training a neural network classifier using a series of random sequences of adjacent pixels until classifier weights are converged to a required tolerance for said measurement,

c) applying said weights to all pixels to obtain a binary image,

d) subtracting a one-pixel shifted version of said binary image from said binary image to obtain an edge map, and

e) determining minimum distance between edges in said edge map as a measure of linewidths of said conductive lines.