Charged particle image inspection

1 . A system comprising:

a memory storing a set of instructions; and

at least one processor configured to execute the set of instructions to cause the system to perform:

obtaining an inspection image of a wafer resulting from a lithography manufacturing process;

obtaining layout data associated with the inspection image;

obtaining a probability map derived from the layout data wherein the probability map identifies a probability of a first type of defect resulting from the lithography manufacturing process occurring in a region of the layout data;

identifying a potential defect resulting from the lithography manufacturing process in the inspection image occurring at a first location;

evaluating a confidence score for the potential defect based on a threshold corresponding to a known defect type;

indicating the potential defect as an unknown defect type if the confidence score does not satisfy the threshold; and

classifying the unknown defect type based on the probability map and the first location.

2 . The system of claim 1 , wherein, in classifying the unknown defect type, the at least one processor is configured to execute the set of instructions to further cause the system to perform:

classifying the unknown defect type as the first type of defect when the first location corresponds to the region of the layout data.

3 . The system of claim 2 , wherein, in classifying the unknown defect type, the at least one processor is configured to execute the set of instructions to further cause the system to perform:

determining the probability that the unknown defect type is above a threshold probability of being the first type of defect; and

in response to a determination that the probability is below the threshold probability, classifying the unknown defect type as a new defect type.

4 . The system of claim 1 , wherein the first type of defect is one of a predetermined set of known defect types.

5 . The system of claim 1 , wherein the first type of defect is one of a hard bridge defect, a soft bridge defect, a hard break defect, a soft break defect, a line-end pullback defect, or a particle defect.

6 . The system of claim 1 , wherein, in evaluating the confidence score, the at least one processor is configured to execute the set of instructions to further cause the system to perform:

evaluating the confidence score based on a threshold corresponding to a known defect type among a predetermined set of known defect types.

7 . The system of claim 1 , wherein the layout data is in Graphic Database System (GDS) format, Graphic Database System II (GDS II) format, Open Artwork System Interchange Standard (OASIS) format, or Caltech Intermediate Format (CIF).

8 . A non-transitory computer readable medium that stores a set of instructions that is executable by at least one processor of a computing device to cause the computing device to perform a method of defect classification, the method comprising:

obtaining an inspection image of a wafer resulting from a lithography manufacturing process;

obtaining layout data associated with the inspection image;

obtaining a probability map derived from the layout data wherein the probability map identifies a probability of a first type of defect resulting from the lithography manufacturing process occurring in a region of the layout data;

identifying a potential defect resulting from the lithography manufacturing process in the inspection image occurring at a first location;

evaluating a confidence score for the potential defect based on a threshold corresponding to a known defect type;

indicating the potential defect as an unknown defect type if the confidence score does not satisfy the threshold; and

classifying the unknown defect type based on the probability map and the first location.

9 . The non-transitory computer readable medium of claim 8 , wherein, in classifying the unknown defect type, the set of instructions that is executable by at least one processor of the computing device to cause the computing device to further perform:

classifying the unknown defect type as the first type of defect when the first location corresponds to the region of the layout data.

10 . The non-transitory computer readable medium of claim 9 , wherein the set of instructions that is executable by at least one processor of the computing device to cause the computing device to further perform:

determining the probability that the unknown defect type is above a threshold probability of being the first type of defect; and

in response to a determination that the probability is below the threshold probability, classifying the unknown defect type as a new defect type.

11 . The non-transitory computer readable medium of claim 8 , wherein the first type of defect is one of a predetermined set of known defect types.

12 . The non-transitory computer readable medium of claim 8 , wherein the first type of defect is one of a hard bridge defect, a soft bridge defect, a hard break defect, a soft break defect, a line-end pullback defect, or a particle defect.

13 . The non-transitory computer readable medium of claim 8 , wherein, in evaluating the confidence score, the set of instructions that is executable by at least one processor of the computing device to cause the computing device to further perform:

evaluating the confidence score based on a threshold corresponding to a known defect type among a predetermined set of known defect types.

14 . A method of defect classification comprising:

obtaining an inspection image of a wafer resulting from a lithography manufacturing process;

obtaining layout data associated with the inspection image;

obtaining a probability map derived from the layout data, wherein the probability map identifies a probability of a first type of defect resulting from the lithography manufacturing process occurring in a region of the layout data;

identifying a potential defect resulting from the lithography manufacturing process in the inspection image occurring at a first location;

evaluating a confidence score for the potential defect based on a threshold corresponding to a known defect type;

indicating the potential defect as an unknown defect type if the confidence score does not satisfy the threshold; and

classifying the unknown defect type based on the probability map and the first location.

15 . The method of claim 14 , wherein classifying the defect further comprises:

classifying the unknown defect type as the first type of defect when the first location corresponds to the region of the layout data.

16 . The method of claim 15 , wherein classifying the unknown defect type further comprises:

determining the probability that the unknown defect type is above a threshold probability of being the first type of defect; and

in response to a determination that the probability is below the threshold probability, classifying the unknown defect type as a new defect type.

17 . The method of claim 14 , wherein the first type of defect is one of a predetermined set of known defect types.

18 . The method of claim 14 , wherein the first type of defect is one of a hard bridge defect, a soft bridge defect, a hard break defect, a soft break defect, a line-end pullback defect, or a particle defect.