Deep learning based adaptive regions of interest for critical dimension measurements of semiconductor substrates

1. A system comprising:

a characterization sub-system configured to acquire one or more images of a specimen, the characterization sub-system including an illumination source and a detector assembly; and

a controller including one or more processors configured to execute a set of program instructions stored in memory, the set of program instructions configured to cause the one or more processors to:

receive one or more training images of a specimen from the characterization sub-system;

receive one or more training region-of-interest (ROI) selections within the one or more training images;

generate a machine learning classifier based on the one or more training images and the one or more training ROI selections, wherein the machine learning classifier is configured to identify one or more measurements of interest of a specimen based on the one or more training images and the one or more training ROI selections;

receive one or more product images of a specimen from the characterization sub-system;

generate one or more classified regions of interest with the machine learning classifier; and

determine one or more measurements of the specimen within the one or more classified regions of interest.

2. The system of claim 1 , wherein the generating one or more classified regions of interest with the machine learning classifier comprises:

receiving one or more product ROI selections within the one or more product images;

adaptively modifying one or more characteristics of the one or more product ROI selections with the machine learning classifier to generate the one or more classified regions of interest.

3. The system of claim 2 , wherein at least one product ROI selection is received from a user via a user interface.

4. The system of claim 2 , wherein the adaptively modifying one or more characteristics of the one or more product ROI selections with the machine learning classifier comprises:

adaptively modifying at least one of a size or a shape of at least one product ROI selection with the machine learning classifier.

5. The system of claim 2 , wherein the adaptively modifying one or more characteristics of the one or more product ROI selections with the machine learning classifier comprises:

adaptively modifying an orientation of at least one product ROI selection with the machine learning classifier to generate a classified region of interest which has been rotated with respect to the at least one product ROI selection.

6. The system of claim 1 , wherein the generating one or more classified regions of interest with the machine learning classifier comprises:

receiving one or more product pattern of interest (POI) selections within the one or more product images; and

generating the one or more classified regions of interest based on the one or more product POI selections.

7. The system of claim 1 , wherein the one or more measurements comprise a critical dimension measurement within the one or more classified regions of interest.

8. The system of claim 1 , wherein the characterization sub-system comprises at least one of a scanning electron microscopy (SEM) sub-system or an optical characterization sub-system.

9. The system of claim 1 , wherein at least one training ROI selection is received from a user via a user interface.

10. The system of claim 1 , wherein the machine learning classifier comprises at least one of a deep learning classifier, a convolutional neural network (CNN), an ensemble learning classifier, a random forest classifier, or an artificial neural network.

11. A system comprising:

a controller including one or more processors configured to execute a set of program instructions stored in memory, the set of program instructions configured to cause the one or more processors to:

receive one or more training images of a specimen;

receive one or more training region-of-interest (ROI) selections within the one or more training images;

generating a machine learning classifier based on the one or more training images and the one or more training ROI selections, wherein the machine learning classifier is configured to identify one or more measurements of interest of a specimen based on the one or more training images and the one or more training ROI selections;

receive one or more product images of a specimen;

generate one or more classified regions of interest with the machine learning classifier; and

determine one or more measurements of the specimen within the one or more classified regions of interest.

12. The system of claim 11 , wherein the generating one or more classified regions of interest with the machine learning classifier comprises:

receiving one or more product ROI selections within the one or more product images; and

adaptively modifying one or more characteristics of the one or more product ROI selections with the machine learning classifier to generate one or more modified regions of interest.

13. The system of claim 12 , wherein the adaptively modifying one or more characteristics of the one or more product ROI selections with the machine learning classifier comprises:

adaptively modifying at least one of a size or a shape of at least one product ROI selection with the machine learning classifier.

14. The system of claim 12 , wherein the adaptively modifying one or more characteristics of the one or more product ROI selections with the machine learning classifier to generate one or more modified regions of interest comprises:

adaptively modifying an orientation of at least one product ROI selection with the machine learning classifier to generate a classified region of interest which has been rotated with respect to the at least one product ROI selection.

15. The system of claim 11 , wherein the one or more measurements comprise a critical dimension measurement within the one or more classified regions of interest.

16. The system of claim 11 , wherein at least one training ROI selection is received from a user input device of a user interface.

17. The system of claim 11 , wherein the machine learning classifier comprises at least one of a deep learning classifier, a convolutional neural network (CNN), an ensemble learning classifier, a random forest classifier, or an artificial neural network.

18. A method comprising:

acquiring one or more training images of a specimen with a characterization sub-system;

receiving one or more training region-of-interest (ROI) selections within the one or more training images;

generating a machine learning classifier based on the one or more training images and the one or more training ROI selections, wherein the machine learning classifier identifies one or more measurements of interest of a specimen based on the one or more training images and the one or more training ROI selections;

acquiring one or more product images of a specimen with the characterization sub-system;

generating one or more classified regions of interest with the machine learning classifier; and

determining one or more measurements of the specimen within the one or more classified regions of interest.