Defect inspection apparatus

• US 20050002560A1
• Filed: 05/28/2004
• Published: 01/06/2005
• Est. Priority Date: 05/29/2003
• Status: Active Grant

• Current Assignee: Nidek Company Limited
• Original Assignee: Nidek Company Limited
• Inventors: Yonezawa, Eiji, Umezaki, Taizo, Yamamoto, Takayasu
• Granted Patent: US 7,333,650 B2
• Time in Patent Office: Days
• US Class Current: 382/156
• CPC Class Codes: G06T 2207/30148 Semiconductor; IC; Wafer
• CPC Class Codes: G06T 7/0004 Industrial image inspection
• CPC Class Codes: G06V 10/25 Determination of region of interest [ROI] or a volume of interest [VOI]

Abstract

A defect inspection apparatus for inspecting an object to be inspected for a defect by processing an image taken from the object, includes: neural networks provided respectively for individual defect types to be classified; a learning unit which makes the neural networks learn based on the corresponding defect types to be classified; and a defect detection unit which classifies and detects defect types using the neural networks that have learned.

Claims

1. A defect inspection apparatus for inspecting an object to be inspected for a defect by processing an image taken from the object, the defect inspection apparatus comprising:

neural networks provided respectively for individual defect types to be classified;

a learning unit which makes the neural networks learn based on the corresponding defect types to be classified; and

a defect detection unit which classifies and detects defect types using the neural networks that have learned.

2. A defect inspection apparatus according to claim 1, wherein the neural networks include at least two of a neural network for a flaw produced during fabrication of the object, a neural network for a partial defocus produced during exposure of the object, a neural network for a shot defocus produced during exposure of the object, and a neural network for a post-development defect produced during development of the object.

3. A defect inspection apparatus according to claim 1, wherein

the neural networks include a neural network for a shot defocus produced during exposure of the object, and

the apparatus further includes an image size conversion unit which converts size of the taken image by an affine transformation so as to correspond to size of an input layer before the taken image is entered into the neural network for a shot defocus.

4. A defect inspection apparatus according to claim 1, further comprising an interpolation unit which, where the taken image contains outside of an objective region of the object, interpolates data about the outside of the objective region based on data about brightness of the objective region before the taken image is entered into the neural networks.

5. A defect inspection apparatus according to claim 1 further comprising an artificial defect image creation unit which creates an artificial defect image for each defect type to be classified by image processing,

wherein the learning unit makes the neural networks learn by the artificial defect image for the corresponding defect type.

6. A defect inspection apparatus according to claim 5, wherein the artificial defect image creation unit creates the artificial defect image by determining at least one of a position, brightness, and a shape of the defect using random numbers.

7. A defect inspection apparatus according to claim 5, wherein the learning unit makes the neural networks to learn by determining a learning pattern of the artificial defect images and good product images for the corresponding defect type using random numbers.

8. A defect inspection apparatus for inspecting an object to be inspected for a defect by processing an image taken from the object, the defect inspection apparatus comprising:

a neural network;

an artificial defect image creation unit which creates an artificial defect, image by image processing;

a learning unit which makes the neural network learn by the created artificial defect image; and

a defect detection unit which detects the defect using the neural network that have learned.

9. A defect inspection apparatus according to claim 8, wherein the artificial defect image creation unit creates the artificial defect image by determining at least one of a position, brightness, and a shape of the detect using random numbers.