Layout-based wafer defect identification and classification using separately generated predicted images and predicted layout designs

1 . A method comprising:

capturing, by a scanning electron microscope system, a scanning electron microscope (SEM) image of an electronic device manufactured according to a layout design describing the electronic device;

separately generating, by a computing system, both a predicted image of the electronic device and a predicted layout design of the electronic device, wherein the predicted image of the electronic device is generated directly from a portion of the layout design describing the electronic device, wherein the predicted layout design of the electronic device is generated directly from the SEM captured image of the electronic device, wherein the predicted image corresponds to an expected image of the electronic device to be captured by the scanning electron microscope system, wherein the predicted layout design of the electronic device is separate from the predicted image of the electronic device, and wherein the predicted layout design corresponds to an expected layout design utilized to manufacture the electronic design system in the captured image;

identifying, by the computing system, one or more manufacturing defects present in the electronic device based on at least one of differences between the predicted layout design of the electronic device and the layout design and differences between the predicted image of the electronic device and the captured SEM image of the electronic device; and

utilizing, by the computing system, the captured SEM image of the electronic device to classify the manufacturing defects identified based on the predicted image of the electronic device from the layout design.

2 . The method of claim 1 , wherein utilizing the layout design to generate the predicted image further comprising:

aligning the capture image of the electronic device to a portion of the layout design;

clipping the portion of the layout design aligned to the captured image of the electronic device; and

generating the predicted image of the electronic device from the clipped portion of the layout design.

3 . The method of claim 1 , further comprising:

comparing, by the computing system, the predicted image of the electronic device and the captured image of the electronic device; and

annotating, by the computing system, the captured image of the electronic device with the differences between the predicted image of the electronic device and the captured image of the electronic device.

4 . The method of claim 1 , further comprising:

generating, by the computing system, a manufacturing defect report identifying locations of the manufacturing defects in the electronic device and classifications of the manufacturing defects; and

performing, by the computing system, repair of the electronic device or modification of the layout design based, at least in part, on a root cause of the manufacturing defect identified using the manufacturing defect report.

5 . The method of claim 1 , further comprising utilizing, by the computing system, the captured SEM image of the electronic device to classify the manufacturing defects identified based on the differences between the predicted layout design of the electronic device and the layout design.

6 . The method of claim 1 , wherein generating the predicted image of the electronic device using the layout design is performed using a first machine-learning algorithm implemented by the computing system, and wherein utilizing the captured image of the electronic device to classify the manufacturing defects identified based on the predicted image of the electronic device from the layout design is performed using a second machine-learning algorithm implemented by the computing system.

7 . A system comprising:

a scanning electron microscope system configured to capture a scanning electron microscope (SEM) image of an electronic device manufactured according to a layout design describing the electronic device;

a memory system configured to store computer-executable instructions; and

a computing system, in response to execution of the computer-executable instructions, is configured to:

separately generate both a predicted image of the electronic device and a predicted layout design of the electronic device, wherein the predicted image of the electronic device is generated directly from a portion of the layout design describing the electronic device, wherein the predicted layout design of the electronic device is generated directly from the SEM captured image of the electronic device, wherein the predicted image corresponds to an expected image of the electronic device to be captured by the scanning electron microscope system, wherein the predicted layout design of the electronic device is separate from the predicted image of the electronic device, and wherein the predicted layout design corresponds to an expected layout design utilized to manufacture the electronic design system in the captured image;

identify one or more manufacturing defects present in the electronic device based on at least one of differences between the predicted layout design of the electronic device and the layout design and differences between the predicted image of the electronic device and the captured SEM image of the electronic device; and

utilize the captured SEM image of the electronic device to classify the manufacturing defects identified based on the predicted image of the electronic device from the layout design.

8 . The system of claim 7 , wherein the computing system, in response to execution of the computer-executable instructions, is further configured to:

align the capture image of the electronic device to a portion of the layout design;

clip the portion of the layout design aligned to the captured image of the electronic device; and

generate the predicted image of the electronic device from the clipped portion of the layout design.

9 . The system of claim 7 , wherein the computing system, in response to execution of the computer-executable instructions, is further configured to:

compare the predicted image of the electronic device and the captured image of the electronic device; and

annotate the captured image of the electronic device with the differences between the predicted image of the electronic device and the captured image of the electronic device.

10 . The system of claim 7 , wherein the computing system, in response to execution of the computer-executable instructions, is further configured to:

generate a manufacturing defect report identifying locations of the manufacturing defects in the electronic device and classifications of the manufacturing defects; and

perform repair of the electronic device or modification of the layout design based, at least in part, on a root cause of the manufacturing defect identified using the manufacturing defect report.

11 . The system of claim 7 , wherein the computing system, in response to execution of the computer-executable instructions, is further configured to utilize the captured image of the electronic device to classify the manufacturing defects identified based on the differences between the predicted layout design of the electronic device and the layout design.

12 . An apparatus comprising at least one non-transitory computer-readable memory device storing instructions configured to cause one or more processing devices to perform operations comprising:

separately generating both a predicted image of an electronic device and a predicted layout design of the electronic device, wherein the predicted image of the electronic device is generated directly from a portion of the layout design describing the electronic device, wherein the predicted layout design of the electronic device is generated directly from a captured scanning electron microscope (SEM) image of the electronic device, wherein the captured SEM image is an image of the electronic device captured by a scanning electron microscope (SEM) system, wherein the predicted image corresponds to an expected image of the electronic device to be captured by the scanning electron microscope system, wherein the predicted layout design of the electronic device is separate from the predicted image of the electronic device, and wherein the predicted layout design corresponds to an expected layout design utilized to manufacture the electronic design system in the captured image;

identifying one or more manufacturing defects present in the electronic device based on at least one of differences between the predicted layout design of the electronic device and the layout design and differences between the predicted image of the electronic device and the captured SEM image of the electronic device; and

utilizing the captured SEM image of the electronic device to classify the manufacturing defects identified based on the predicted image of the electronic device from the layout design.

13 . The apparatus of claim 12 , wherein utilizing the layout design to generate the predicted image further comprising:

aligning the capture image of the electronic device to a portion of the layout design;

clipping the portion of the layout design aligned to the captured image of the electronic device; and

generating the predicted image of the electronic device from the clipped portion of the layout design.

14 . The apparatus of claim 12 , wherein the instructions are configured to cause one or more processing devices to perform operations further comprising:

comparing the predicted image of the electronic device and the captured image of the electronic device; and

annotating the captured image of the electronic device with the differences between the predicted image of the electronic device and the captured image of the electronic device.

15 . The apparatus of claim 12 , wherein the instructions are configured to cause one or more processing devices to perform operations further comprising:

generating a manufacturing defect report identifying locations of the manufacturing defects in the electronic device and classifications of the manufacturing defects; and

performing repair of the electronic device or modification of the layout design based, at least in part, on a root cause of the manufacturing defect identified using the manufacturing defect report.

16 . The apparatus of claim 12 , wherein the instructions are configured to cause one or more processing devices to perform operations further comprising utilizing the captured SEM image of the electronic device to classify the manufacturing defects identified based on the differences between the predicted layout design of the electronic device and the layout design.

17 . The apparatus of claim 12 , wherein generating the predicted image of the electronic device using the layout design is performed using a first machine-learning algorithm implemented by the one or more processing devices, and wherein utilizing the captured image of the electronic device to classify the manufacturing defects identified based on the predicted image of the electronic device from the layout design is performed using a second machine-learning algorithm implemented by the one or more processing devices.