System and method of optically inspecting structures on an object

• US 6,487,307 B1
• Filed: 03/04/1999
• Issued: 11/26/2002
• Est. Priority Date: 11/30/1994
• Status: Expired due to Term

• Current Assignee: Rudolph Technologies Incorporated (Onto Innovation Inc.)
• Original Assignee: ISOA, Inc. (Onto Innovation Inc.)
• Inventors: Hennessey, Kathleen, Lin, Youling
• Primary Examiner(s): ROGERS, SCOTT A
• Application Number:
  US09/262,603
• Time in Patent Office: 1,363 Days
• Field of Search: 382/145, 382/149, 382/151, 382/141, 382/199, 382/203-204, 382/224, 356/237.2, 356/237.3, 356/237.4, 356/237.5, 356/394, 356/448
• US Class Current: 382/149
• CPC Class Codes: G06F 18/24765 Rule-based classification
• CPC Class Codes: G06T 2207/30148 Semiconductor; IC; Wafer
• CPC Class Codes: G06T 7/001 using an image reference approach

Abstract

A system and method of optically inspecting a structure on a surface of a production object which is supported on a moving inspection platform. The method reliably traces structure edges and stores structure and structure island characteristics. The system uses a camera and a light or energy source to sharply delineate the structure edges. A sequence of images of the object and the structure are captured, and the structure is detected in each image. The structure is then symbolically decomposed into primitives, and a histogram is produced for each image identifying the slope and length of each edge of the structure. The histograms are compared in each image, and are aligned to eliminate differences due to wobble of the inspection platform or differences in magnification. A production structure grammar is then produced from the aligned images. The production structure grammar is compared to a reference structure grammar generated from a defect-free object. Differences in the two grammars, if any, are used to identify missing, misaligned, or misoriented structures on the production object, and to detect foreign objects and other defects.

Claims

1. A system for inspecting structures on a surface of an object, said system comprising:

a support which supports the object, said object having a structure on the surface thereof;

a device which emits energy onto the surface of the object for illuminating the surface and the structure;

an image capturing device mounted adjacent to the object which captures an image of the illuminated surface of the object and the structure; and

a computer which generates a structure grammar for the entire structure from the image.

2. The system of claim 1 wherein the device which emits energy is selected from the group consisting of a visible light source, an infrared light source, an ultraviolet light source, an X-ray source, a focused ion beam source, and an electron source.

3. The system of claim 1 wherein the image capturing device is selected from the group consisting of a line-scan camera and an area-scan camera.

4. The system of claim 1 wherein the computer includes:

means for comparing the structure grammar generated from the image to a reference structure grammar for a defect-free object; and

means for recording any differences between the structure grammar generated from the image and the reference structure grammar.

5. The system of claim 4 wherein the reference structure grammar for the defect-free object is generated from computer aided drafting.

6. The system of claim 4 wherein the reference structure grammar for the defect-free object is generated from an image of a defect-free object.

7. The system of claim 1 further comprising a knowledge base accessible by the computer for receiving and storing reference structure grammar.

8. The system of claim 4 wherein the computer performs Automatic Defect Classification (ADC) for classifying any differences between the structure grammar generated from the image and the reference structure grammar.

9. A method of inspecting a photoresist structure on a top surface of a semiconductor wafer die, said photoresist structure having a plurality of edges, the method comprising the steps of:

shining a light on the wafer having the die thereon, said light illuminating the top surface of the wafer and the die so as to clearly expose the plurality of edges of the photoresist structure;

capturing an image of a portion of the wafer having the die thereon;

detecting a die image within the captured image;

scanning the die image to detect the photoresist structure;

tracing the photoresist structure to produce primitives for the plurality of edges of the photoresist structure, thereby producing a production grammar; and

comparing the production grammar to a defect-free grammar representative of a defect-free die.

10. The method of claim 9 further comprising the step of converting the image into a pixel based image.

11. The method of claim 9 further comprising the step of producing a defect-free grammar by processing a computer aided drafting image of a defect-free die.

12. The method of claim 9 further comprising the step of producing a defect-free grammar by processing an image of a defect-free die.

13. The method of claim 9 further comprising the step of storing the defect-free grammar in a knowledge base.

14. The method of claim 9 further comprising the step of storing a production grammar in a knowledge base as a defect grammar when the production grammar differs from the defect-free grammar.

15. The method of claim 14 further comprising the step of organizing the defect grammar by a defect type.

16. The method of claim 9 further comprising the step of performing an Automatic Defect Classification on the production grammar.

17. A method of inspecting a structure on a surface of a production object, said structure having a plurality of edges, said method comprising the steps of:

emitting energy onto the surface of the object to illuminate the surface and the structure;

capturing, with an image capturing device, a plurality of sequential images of the illuminated surface of the object and the structure;

tracing the edges of the structure to generate primitives for each edge;

creating a histogram for each of the plurality of images, the histogram identifying a slope and a length of each edge of the structure as captured in each of the images;

aligning the images by aligning the histograms for the plurality of images;

generating a production structure grammar from the aligned images; and

comparing the production structure grammar with a reference structure grammar generated from a defect-free object to identify any defects on the production object.

18. The method of inspecting a structure on a surface of a production object of claim 17 wherein the object is supported on a moving platform while the images are captured, and the step of aligning the images by aligning the histograms for the plurality of images includes adjusting for wobble of the moving platform, said step of adjusting for wobble including the steps of:

comparing the slopes of the edges of the structure in two different histograms to identify differences in the slopes;

determining whether the differences in the slopes are a change in the slopes which is common to all edges; and

adjusting the slopes to be equal in each image upon determining that the differences in the slopes are a change in the slopes which is common to all edges.

19. The method of inspecting a structure on a surface of a production object of claim 18 wherein the step of aligning the images by aligning the histograms for the plurality of images includes adjusting for differences in magnification between the images, said step of adjusting for differences in magnification including the steps of:

comparing the lengths of the edges of the structure in two different histograms to identify differences in the lengths;

determining whether the differences in the lengths are a change in the lengths which is common to all edges; and

adjusting the lengths to match in each image upon determining that the differences in the lengths are a change in the lengths which is common to all edges.