US-12626351-B2 - Image processing method and system thereof

Abstract

An image processing system, including an input interface configured to receive a first direction image corresponding to a view of a semiconductor device in a first direction, and a second direction image corresponding to a view of the semiconductor device in a second direction which intersects the first direction at a first height at which the first direction image is generated; a processor configured to perform an edge detection operation for detecting an edge based on the first direction image, and to perform an image binarization operation on the first direction image; and a learning device configured to compare a first line width obtained based on the image binarization operation, and a second line width obtained based on the second direction image through machine learning, and to learn a condition of the image binarization operation which maximizes a correlation between the first line width and the second line width.

Inventors

• Min Su Kang
• Jang Hoon Kim
• Woo Jin Jung

Assignees

• SAMSUNG ELECTRONICS CO., LTD.

Dates

Publication Date

20260512

Application Date

20230501

Priority Date

20220704

Claims (20)

1. 1 . An image processing system comprising: at least one processor; and a memory configured to store one or more instructions which, when executed by the at least one processor, cause the image processing system to: receive a first direction image corresponding to a view of a semiconductor device in a first direction, and a second direction image corresponding to a view of the semiconductor device in a second direction which intersects the first direction at a first height at which the first direction image is generated, perform an edge detection operation for detecting an edge based on the first direction image, perform an image binarization operation on the first direction image, compare a first line width obtained based on the image binarization operation, and a second line width obtained based on the second direction image through machine learning, and learn a condition of the image binarization operation which maximizes a correlation between the first line width and the second line width.
2. 2 . The image processing system of claim 1 , wherein the memory is further configured to store the condition of the image binarization operation.
3. 3 . The image processing system of claim 1 , wherein the one or more instructions, when executed by the at least one processor, further cause the image processing system to: communicate with an external device which generates the first direction image and the second direction image.
4. 4 . The image processing system of claim 1 , wherein the first direction image and the second direction image are obtained using a scanning electron microscope (SEM) or a transmission electron microscope (TEM).
5. 5 . The image processing system of claim 1 , wherein the edge detection operation is performed based on a gradient value obtained by applying a gradient to brightness information about pixels of the first direction image.
6. 6 . The image processing system of claim 5 , wherein the edge detection operation determines pixels in which the gradient value is greater than a threshold gradient value, to be the edge.
7. 7 . The image processing system of claim 1 , wherein the image binarization operation determines pixels having a brightness greater than a threshold brightness to be an inner region, based on brightness information about pixels of the first direction image.
8. 8 . The image processing system of claim 7 , wherein the at least one processor is further configured to: calculate an edge detection ratio value by dividing a number of edges included in the inner region by a total number of edges detected in the first direction image; and perform the image binarization operation such that the edge detection ratio value is equal to a predetermined edge detection ratio value.
9. 9 . An image processing system comprising: at least one processor; and a memory configured to store a machine learning model for performing machine learning, and one or more instructions which, when executed by the at least one processor, cause the image processing system to: receive a first direction image corresponding to a view of a semiconductor device in a first direction, and a second direction image corresponding to a view of the semiconductor device in a second direction intersecting the first direction at a first height at which the first direction image is generated; perform the machine learning based on the first direction image and the second direction image; wherein to perform the machine learning, the one or more instructions which, when executed by the at least one processor, cause the image processing system to: perform an edge detection operation for detecting an edge based on the first direction image, perform an image binarization operation on the first direction image, compare a first line width obtained based on the image binarization operation with a second line width obtained based on the second direction image using the machine learning, and learn a condition of the image binarization operation at which a correlation between the first line width and the second line width is maximized by performing the machine learning.
10. 10 . The image processing system of claim 9 , wherein the memory is further configured to store the condition of the image binarization operation.
11. 11 . The image processing system of claim 9 , wherein the first direction image and the second direction image are obtained using a scanning electron microscope (SEM) or a transmission electron microscope (TEM).
12. 12 . The image processing system of claim 9 , wherein the edge detection operation is performed based on a gradient value obtained by applying a gradient to brightness information about pixels of the first direction image.
13. 13 . The image processing system of claim 12 , wherein the edge detection operation determines pixels in which the gradient value is greater than a threshold gradient value, to be the edge.
14. 14 . The image processing system of claim 9 , wherein the image binarization operation determines pixels having brightness greater than a threshold brightness to be an inner region, based on brightness information about pixels of the first direction image.
15. 15 . The image processing system of claim 9 , wherein the one or more instructions, when executed by the at least one processor, further cause the image processing system to: calculate an edge detection ratio value obtained by dividing a number of edges included in an inner region by a total number of edges detected in the first direction image; and perform the image binarization operation such that the edge detection ratio value is equal to a predetermined edge detection ratio value.
16. 16 . An image processing method comprising: receiving a first direction image corresponding to a view of a semiconductor device in a first direction, and a second direction image corresponding to a view of the semiconductor device in a second direction intersecting the first direction at a first height at which the first direction image is generated through an input interface; performing an edge detection operation for detecting an edge based on the first direction image, using a processor; performing an image binarization operation on the first direction image, using the processor; and comparing a first line width obtained based on the image binarization operation, and a second line width obtained based on the second direction image using machine learning, and learning a condition of the image binarization operation which maximizes a correlation between the first line width and the second line width, using a learning device.
17. 17 . The image processing method of claim 16 , wherein the edge detection operation is performed based on a gradient value obtained by applying a gradient to brightness information about pixels of the first direction image.
18. 18 . The image processing method of claim 17 , wherein the edge detection operation determines pixels in which the gradient value is greater than a threshold gradient value, to be the edge.
19. 19 . The image processing method of claim 16 , wherein the image binarization operation determines pixels having brightness greater than a threshold brightness to be an inner region, based on brightness information about pixels of the first direction image.
20. 20 . The image processing method of claim 19 , wherein an edge detection ratio value is obtained by dividing a number of edges included in the inner region by a total number of edges detected in the first direction image, and wherein the image binarization operation is performed such that the edge detection ratio value is equal to a predetermined edge detection ratio value.

Description

CROSS-REFERENCE TO RELATED APPLICATION This application is based on and claims priority under 35 U.S.C. § 119 to Korean Patent Application No. 10-2022-0081954 filed on Jul. 4, 2022 in the Korean Intellectual Property Office, the disclosure of which of which is incorporated by reference herein in its entirety. BACKGROUND 1. Field The disclosure relates to an image processing method and a system thereof. 2. Description of Related Art In recent years, as semiconductor devices have become highly integrated, the size of element inside a chip and the interval between the elements have been decreasing. Conversely, the decrease in the size of elements and decrease in the interval between the elements are a matter of being able to achieve high integration of the semiconductors. The high integration and high speed may be important not only in a memory field but also in a non-memory field. In a logic device such as a Central Processing Units (CPUs), the speed of the element may be increased by reducing dimensions such as a gate line width, which may be a width of a gate electrode, to achieve high speed of the signal. It may be important to adjust the gate line width for the high integration and speed improvement. SUMMARY Provided is an image processing system in which reliability of line width measured on the basis of an image during process of a semiconductor device is improved. Provided is an image processing method in which reliability of line width measured on the basis of an image during process of a semiconductor device is improved. Additional aspects will be set forth in part in the description which follows and, in part, will be apparent from the description, or may be learned by practice of the presented embodiments. In accordance with an aspect of the disclosure, an image processing system includes an input interface configured to receive a first direction image corresponding to a view of a semiconductor device in a first direction, and a second direction image corresponding to a view of the semiconductor device in a second direction which intersects the first direction at a first height at which the first direction image is generated; a processor configured to perform an edge detection operation for detecting an edge based on the first direction image, and to perform an image binarization operation on the first direction image; and a learning device configured to compare a first line width obtained based on the image binarization operation, and a second line width obtained based on the second direction image through machine learning, and to learn a condition of the image binarization operation which maximizes a correlation between the first line width and the second line width. In accordance with an aspect of the disclosure, an image processing system includes communication interface configured to communicate with an outside of the image processing system; an input interface configured to receive a first direction image corresponding to a view of a semiconductor device in a first direction, and a second direction image corresponding to a view of the semiconductor device in a second direction intersecting the first direction at a first height at which the first direction image is generated, through the communication interface; a learning device configured to perform machine learning based on the first direction image and the second direction image; a memory configured to store a machine learning model for performing the machine learning; and a processor configured to control the communication interface, the input interface, the learning device, and the memory, wherein the processor is further configured to: perform an edge detection operation for detecting an edge based on the first direction image, perform an image binarization operation on the first direction image, compare a first line width obtained based on the image binarization operation with a second line width obtained based on the second direction image using the machine learning, and learn a condition of the image binarization operation at which a correlation between the first line width and the second line width is maximized, using the learning device. In accordance with an aspect of the disclosure, an image processing method includes receiving a first direction image corresponding to a view of a semiconductor device in a first direction, and a second direction image corresponding to a view of the semiconductor device in a second direction intersecting the first direction at a first height at which the first direction image is generated through an input interface; performing an edge detection operation for detecting an edge based on the first direction image, using a processor; performing an image binarization operation on the first direction image, using the processor; and comparing a first line width obtained based on the image binarization operation, and a second line width obtained based on the second direction image using machine learning, a