Inspection method and apparatus for the inspection of either random or repeating patterns

• US 6,665,432 B1
• Filed: 02/01/2000
• Issued: 12/16/2003
• Est. Priority Date: 09/30/1993
• Status: Expired due to Fees

• Current Assignee: Bin-Ming Ben Tsai, David M.W. Evans, Jason Z. Lin
• Original Assignee: Bin-Ming Ben Tsai, David M.W. Evans, Jason Z. Lin
• Inventors: Lin, Jason Z., Tsai, Bin-Ming Ben, Evans, David M. W.
• Primary Examiner(s): Tran, Phuoc
• Application Number:
  US09/496,013
• Time in Patent Office: 1,414 Days
• Field of Search: 382/141, 382/149, 382/168, 382/170, 382/172, 382/280, 382/278, 382/282, 382/294, 382/270, 382/275, 382/151
• US Class Current: 382/141
• CPC Class Codes: G03F 7/7065 Defects, e.g. optical inspection of patterned layer for defects
• CPC Class Codes: G06F 18/00 Pattern recognition
• CPC Class Codes: G06T 2207/20056 Discrete and fast Fourier transform, [DFT, FFT]
• CPC Class Codes: G06T 2207/30148 Semiconductor; IC; Wafer
• CPC Class Codes: G06T 7/0004 Industrial image inspection
• CPC Class Codes: G06T 7/37 using transform domain methods
• CPC Class Codes: G06V 10/431 Frequency domain transformation; Autocorrelation
• CPC Class Codes: G06V 10/751 Comparing pixel values or logical combinations thereof, or feature values having positional relevance, e.g. template matching

Abstract

The present invention is a hybrid technique for finding defects on digitized device images using a combination of spatial domain and frequency domain techniques. The two dimensional spectra of two images are found using Fourier like transforms. Any strong harmonics in the spectra are removed, using the same spectral filter on both spectra. The images are then aligned, transformed back to the spatial domain, and subtracted. The resulting spectrally-filtered difference image is thresholded and analyzed for defects. Use of the hybrid technique of the present invention to process digitized images results in the highest-performance and most flexible defect detection system. It is the best performer on both array and random devices, and it can cope with problems such as shading variations and the dark-bright problem that no other technique can address. The hybrid technique of the present invention also uses frequency domain techniques to align the images with fewer errors than spatial domain techniques of similar or lesser complexity. Further, the relative offsets of the pairs of images are determined by frequency domain techniques—and this method may be the most accurate and the least expensive.

Claims

1. A method of inspecting a patterned substrate, comprising:

a) obtaining a first image of a first portion of said substrate;

b) obtaining a second image of a second portion of said substrate;

c) converting said first image and said second image into a frequency domain;

d) aligning said first image with said second image in said frequency domain;

e) converting said first image and said second image to a spatial domain; and

f) comparing said first image to said second image in said spatial domain.

2. The method of claim 1, wherein said first and second images are compared in said spatial domain to determine whether defects are present in said substrate.

3. The method of claim 1, wherein said conversion of said first and second images into said frequency domain is performed by use of a mathematical operation.

4. The method of claim 3, wherein said mathematical operation is a sinusoidal transform.

5. The method of claim 3, wherein said first image and said second image are each divided into a plurality of sub-images, and said mathematical operation is performed independently and simultaneously on each of said sub-images.

6. A method of inspecting a patterned substrate, comprising:

a) obtaining a first image of a first portion of said substrate;

b) obtaining a second image of a second portion of said substrate;

c) mathematically converting at least one of said first image and said second image into a frequency domain;

d) analyzing said converted image to detect repeating elements within said pattern; and

e) comparing said first image to said second image in a spatial domain.

7. The method of claim 6, wherein said first and second images are compared in said spatial domain to determine whether defects are present in said substrate.

8. The method of claim 7, wherein said mathematical conversion comprises a sinusoidal transform.

9. The method of claim 7, wherein at least one of said first image and said second image is divided into a plurality of sub-images, and said mathematical operation is performed independently and simultaneously on each of said sub-images.