Training a neural network for defect detection in low resolution images

Patent metadata

Field	Value
Publication number	US-10599951-B2
Application number	US-201916364140-A
Country	US
Kind code	B2
Filing date	Mar 25, 2019
Priority date	Mar 28, 2018
Publication date	Mar 24, 2020
Grant date	Mar 24, 2020

Abstract

Methods and systems for training a neural network for defect detection in low resolution images are provided. One system includes an inspection tool that includes high and low resolution imaging subsystems and one or more components that include a high resolution neural network and a low resolution neural network. Computer subsystem(s) of the system are configured for generating a training set of defect images. At least one of the defect images is generated synthetically by the high resolution neural network using an image generated by the high resolution imaging subsystem. The computer subsystem(s) are also configured for training the low resolution neural network using the training set of defect images as input. In addition, the computer subsystem(s) are configured for detecting defects on another specimen by inputting the images generated for the other specimen by the low resolution imaging subsystem into the trained low resolution neural network.

First claim

What is claimed is:

1. A system configured to train a neural network for defect detection in low resolution images, comprising:
an inspection tool comprising a high resolution imaging subsystem and a low resolution imaging subsystem, wherein the high and low resolution imaging subsystems comprise at least an energy source and a detector, wherein the energy source is configured to generate energy that is directed to a specimen, and wherein the detector is configured to detect energy from the specimen and to generate images responsive to the detected energy;
one or more computer subsystems configured for acquiring the images of the specimen generated by the high and low resolution imaging subsystems; and
one or more components executed by the one or more computer subsystems, wherein the one or more components comprise a high resolution neural network and a low resolution neural network; and
wherein the one or more computer subsystems are further configured for:
generating a training set of defect images, wherein at least one of the defect images is generated synthetically by the high resolution neural network using at least one of the images generated by the high resolution imaging subsystem;
training the low resolution neural network using the training set of defect images as input; and
detecting defects on another specimen by inputting the images generated for the other specimen by the low resolution imaging subsystem into the trained low resolution neural network.

2. The system of claim 1 , wherein the training set of defect images comprises images of the specimen generated by more than one mode of the low resolution imaging subsystem.

3. The system of claim 2 , wherein the more than one mode of the low resolution imaging subsystem comprises all of the modes of the low resolution imaging subsystem.

4. The system of claim 2 , wherein the one or more computer subsystems are further configured for selecting one or more of the more than one mode of the low resolution imaging subsystem used for detecting the defects on the other specimen based on results of training the low resolution neural network with the images generated by the more than one mode of the low resolution imaging subsystem.

5. The system of claim 1 , wherein the inspection t configured as a macro inspection tool.

6. The system of claim 1 , wherein the defects detected on the other specimen are defects of a back end layer of the other specimen.

7. The system of claim 1 , wherein the defects detected on the other specimen are defects of a redistribution layer of the other specimen.

8. The system of claim 1 , wherein the defects detected on the other specimen are defects of a high noise layer of the other specimen.

9. The system of claim 1 , wherein the defects detected on the other specimen are defects of a layer comprising metal lines of the other specimen.

10. The system of claim 1 , wherein the other specimen on which the defects are detected is a post-dice specimen.

11. The system of claim 1 , wherein the high and low resolution neural networks are configured for single image defect detection.

12. The system of claim 1 , wherein the training set of defect images comprises one or more images of one or more programmed defects on the specimen, wherein the one or more computer subsystems are further configured for generating the one or more programmed defects by altering a design for the specimen to create the one or more programmed defects in the design, and wherein the altered design is printed on the specimen to create the one or more programmed defects on the specimen.

13. The system of claim 1 , wherein the training set of defects comprises one or more images of one or more synthetic defects, and wherein the one or more computer subsystems are further configured for generating the one or more synthetic defects by altering a design for the specimen to create the one or more synthetic defects in the design, generating simulated high resolution images for the one or more synthetic defects based on the one or more synthetic defects in the design, and adding the simulated high resolution images to the training set.

14. The system of claim 13 , wherein the one or more computer subsystems are further configured for generating the simulated high resolution images using the high resolution neural network, and wherein the high resolution neural network is configured as a deep generative model.

15. The system of claim 1 , wherein the training set of defects comprises one or more images of one or more synthetic defects, wherein the one or more computer subsystems are further configured for generating the one or more images of the one or more synthetic defects by altering a design for the specimen to create the one or more synthetic defects in the design, and wherein the one or more computer subsystems are further configured for generating simulated low resolution images for the one or more synthetic defects based on the one or more synthetic defects in the design.

16. The system of claim 15 , wherein the one or more computer subsystems are further configured for generating the simulated low resolution images using a deep generative model.

17. The system of claim 15 , wherein generating the simulated low resolution images is performed with a generative adversarial network or a variational Bayesian method.

18. The system of claim 1 , wherein the training set of defects comprises one or more synthetic defects, and wherein the one or more computer subsystems are further configured for generating the one or more synthetic defects by altering one or more of the images generated by the high resolution imaging subsystem and one or more of the images generated by the low resolution imaging subsystem to create a segmentation image, altering the one or more of the images generated by the high resolution imaging subsystem based on the segmentation image, and generating simulated low resolution images for the one or more synthetic defects based on the altered one or more images.

19. The system of claim 18 , wherein generating the simulated low resolution images is performed with a generative adversarial network or a variational Bayesian method.

20. The system of claim 1 , wherein the one or more computer subsystems are further configured for generating the at least one of the defect images synthetically by altering the at least one of the images generated by the high resolution imaging subsystem for the specimen to create high resolution images for known defects of interest.

21. The system of claim 1 , wherein the training set of defect images comprises one or more images of one or more artificial defects on the specimen generated by performing a process on the specimen known to cause the one or more artificial defects on the specimen.

22. The system of claim 1 , wherein the training set of defect images comprises one or more defects detected on the specimen in one or more of the images generated by the high resolution imaging subsystem.

23. The system of claim 22 , wherein the one or more computer subsystems are further configured for detecting the defects on the specimen in the images generated by the high resolution imaging subsystem by single image detection.

24. The system of claim 22 , wherein the one or more computer subsystems are further configured for detecting the defects on the specimen in the images generated by the high resolution imaging subsystem by die-to-database detection.

Assignees

• Kla Tencor Corp

Inventors

• Bhaskar Kris
• Karsenti Laurent
• Ries Brad
• Nicolaides Lena
• Yeoh Richard (Seng Wee)
• Hiebert Stephen

Classifications

• G06T2207/20084

  Artificial neural networks [ANN] · CPC title

• G06T2207/30148

  Semiconductor; IC; Wafer · CPC title

• G06T2207/20081

  Training; Learning · CPC title

• G06T7/0004 Primary

  Industrial image inspection · CPC title

• G06V30/2504

  Coarse or fine approaches, e.g. resolution of ambiguities or multiscale approaches · CPC title