公开(公告)号：US20240331132A1

公开(公告)日：2024-10-03

申请号：US18577684

申请日：2022-06-03

Applicant: ASML Netherlands B.V.

Inventor： Haoyi LIANG ,  Yani CHEN ,  Ming-Yang YANG ,  Yang YANG ,  Xiaoxia HUANG ,  Zhichao CHEN ,  Liangjiang YU ,  Zhe WANG ,  Lingling PU

IPC: G06T7/00 ,  G01N23/2251 ,  G06T7/73

CPC classification number: G06T7/001 ,  G01N23/2251 ,  G06T7/0006 ,  G06T7/74 ,  G01N2223/6116 ,  G06T2200/24 ,  G06T2207/10061 ,  G06T2207/20081 ,  G06T2207/30148

Abstract: Systems and methods for detecting a defect on a sample include receiving a first image and a second image associated with the first image; determining, using a clustering technique, N first feature descriptor(s) for L first pixel(s) in the first image and M second feature descriptor(s) for L second pixel(s) in the second image, wherein each of the L first pixel(s) is co-located with one of the L second pixel(s), and L, M, and N are positive integers; determining K mapping probability between a first feature descriptor of the N first feature descriptor(s) and each of K second feature descriptor(s) of the M second feature descriptor(s), wherein K is a positive integer; and providing an output for determining whether there is existence of an abnormal pixel representing a candidate defect on the sample based on a determination that one of the K mapping probability does not exceed a threshold value.

公开(公告)号：US20220404712A1

公开(公告)日：2022-12-22

申请号：US17772529

申请日：2020-10-01

Applicant: ASML NETHERLANDS B.V.

Inventor： Qiang ZHANG ,  Yunbo GUO ,  Yu CAO ,  Jen-Shiang WANG ,  Yen-Wen LU ,  Danwu CHEN ,  Pengcheng YANG ,  Haoyi LIANG ,  Zhichao CHEN ,  Lingling PU

IPC: G03F7/20 ,  G06V10/774 ,  G06V10/82 ,  G06T7/32 ,  G06T7/33

Abstract: A method for training a machine learning model to generate a predicted measured image, the method including obtaining (a) an input target image associated with a reference design pattern, and (b) a reference measured image associated with a specified design pattern printed on a substrate, wherein the input target image and the reference measured image are non-aligned images; and training, by a hardware computer system and using the input target image, the machine learning model to generate a predicted measured image.

公开(公告)号：US20220068590A1

公开(公告)日：2022-03-03

申请号：US17418741

申请日：2019-12-19

Applicant: ASML Netherlands B.V.

Inventor： Ying LUO ,  Zhonghua DONG ,  Xuehui YIN ,  Long DI ,  Nianpei DENG ,  Wei FANG ,  Lingling PU ,  Ruochong FEI ,  Bohang ZHU ,  Yu LIU

IPC: H01J37/147 ,  H01J37/20 ,  H01J37/21 ,  H01J37/28

Abstract: Systems and methods for irradiating a sample with a charged-particle beam are disclosed. The charged-particle beam system may comprise a stage configured to hold a sample and is movable in at least one of X-Y-Z axes. The charged-particle beam system may further comprise a position sensing system to determine a lateral and vertical displacement of the stage, and a beam deflection controller configured to apply a first signal to deflect a primary charged-particle beam incident on the sample to at least partly compensate for the lateral displacement, and to apply a second signal to adjust a focus of the deflected charged-particle beam incident on the sample to at least partly compensate for the vertical displacement of the stage. The first and second signals may comprise an electrical signal having a high bandwidth in a range of 10 kHz to 50 kHz, and 50 kHz to 200 kHz, respectively.

公开(公告)号：US20250078478A1

公开(公告)日：2025-03-06

申请号：US18882681

申请日：2024-09-11

Applicant: ASML Netherlands B.V.

Inventor： Wentian ZHOU ,  Liangjiang YU ,  Teng WANG ,  Lingling PU ,  Wei FANG

IPC: G06V10/774 ,  G06F18/214 ,  G06T7/00 ,  G06V10/776 ,  G06V10/98

Abstract: Disclosed herein is a method of automatically obtaining training images to train a machine learning model that improves image quality. The method may comprise analyzing a plurality of patterns of data relating to a layout of a product to identify a plurality of training locations on a sample of the product to use in relation to training the machine learning model. The method may comprise obtaining a first image having a first quality for each of the plurality of training locations, and obtaining a second image having a second quality for each of the plurality of training locations, the second quality being higher than the first quality. The method may comprise using the first image and the second image to train the machine learning model.

公开(公告)号：US20250006456A1

公开(公告)日：2025-01-02

申请号：US18787932

申请日：2024-07-29

Applicant: ASML Netherlands B.V.

Inventor： Wei FANG ,  Lingling PU ,  Bo WANG ,  Zhonghua DONG ,  Yongxin WANG

IPC: H01J37/22 ,  G01N23/2251 ,  G06T5/50 ,  G06T5/77 ,  G06T5/80 ,  H01J37/244 ,  H01J37/28

Abstract: An improved apparatus and method for enhancing an image, and more particularly an apparatus and method for enhancing an image through cross-talk cancellation in a multiple charged-particle beam inspection are disclosed. An improved method for enhancing an image includes acquiring a first image signal of a plurality of image signals from a detector of a multi-beam inspection system. The first image signal corresponds to a detected signal from a first region of the detector on which electrons of a first secondary electron beam and of a second secondary electron beam are incident. The method includes reducing, from the first image signal, cross-talk contamination originating from the second secondary electron beam using a relationship between the first image signal and beam intensities associated with the first secondary electron beam and the second secondary electron beam. The method further includes generating a first image corresponding to first secondary electron beam after reduction.

公开(公告)号：US20230076943A1

公开(公告)日：2023-03-09

申请号：US17985087

申请日：2022-11-10

Applicant: ASML Netherlands B.V.

Inventor： Lingling PU ,  Wei FANG ,  Zhong-wei CHEN

IPC: H01L21/02 ,  G03F7/20 ,  G06F30/39

Abstract: Systems and methods for in-die metrology using target design patterns are provided. These systems and methods include selecting a target design pattern based on design data representing the design of an integrated circuit, providing design data indicative of the target design pattern to enable design data derived from the target design pattern to he added to second design data, wherein the second design data is based on the first design data. Systems and methods can further include causing structures derived from the second design data to be printed on a wafer, inspecting the structures on the wafer using a charged-particle beam tool, and identifying metrology data or process defects based on the inspection. In some embodiments the systems and methods further include causing the charged-particle beam tool, the second design data, a scanner, or photolithography equipment to be adjusted based on the identified metrology data or process defects.

公开(公告)号：US20200211845A1

公开(公告)日：2020-07-02

申请号：US16730897

申请日：2019-12-30

Applicant: ASML Netherlands B.V.

Inventor： Lingling PU ,  Wei FANG ,  Zhong-wei CHEN

IPC: H01L21/02 ,  G03F7/20

Abstract: Systems and methods for in-die metrology using target design patterns are provided. These systems and methods include selecting a target design pattern based on design data representing the design of an integrated circuit, providing design data indicative of the target design pattern to enable design data derived from the target design pattern to be added to second design data, wherein the second design data is based on the first design data. Systems and methods can further include causing structures derived from the second design data to be printed on a wafer, inspecting the structures on the wafer using a charged-particle beam tool, and identifying metrology data or process defects based on the inspection. In some embodiments the systems and methods further include causing the charged-particle beam tool, the second design data, a scanner, or photolithography equipment to be adjusted based on the identified metrology data or process defects.

公开(公告)号：US20200211178A1

公开(公告)日：2020-07-02

申请号：US16718706

申请日：2019-12-18

Applicant: ASML Netherlands B.V.

Inventor： Wentian ZHOU ,  Liangjiang YU ,  Teng WANG ,  Lingling PU ,  Wei FANG

IPC: G06T7/00 ,  G06K9/62

Abstract: Disclosed herein is a method of automatically obtaining training images to train a machine learning model that improves image quality. The method may comprise analyzing a plurality of patterns of data relating to a layout of a product to identify a plurality of training locations on a sample of the product to use in relation to training the machine learning model. The method may comprise obtaining a first image having a first quality for each of the plurality of training locations, and obtaining a second image having a second quality for each of the plurality of training locations, the second quality being higher than the first quality. The method may comprise using the first image and the second image to train the machine learning model.

公开(公告)号：US20200018944A1

公开(公告)日：2020-01-16

申请号：US16508087

申请日：2019-07-10

Applicant: ASML Netherlands B.V

Inventor： Wei FANG ,  Lingling PU ,  Thomas Jarik HUISMAN ,  Erwin Paul SMAKMAN

IPC: G02B21/36 ,  G02B21/00 ,  G06T5/50 ,  G06T7/00 ,  G06K9/00 ,  G06N20/00

Abstract: Systems and methods for image enhancement are disclosed. A method for enhancing an image may include acquiring a first scanning electron microscopy (SEM) image at a first resolution. The method may also include acquiring a second SEM image at a second resolution. The method may further include providing an enhanced image by using the first SEM image as a reference to enhance the second SEM image. The enhanced image may be provided by using one or more features extracted from the first image to enhance the second SEM image, or using the first SEM image as a reference to numerically enhance the second SEM image.

公开(公告)号：US20250166161A1

公开(公告)日：2025-05-22

申请号：US18841032

申请日：2023-02-03

Applicant: ASML Netherlands B.V.

Inventor： Hairong LEI ,  Qian DONG ,  Cho Huak TEH ,  Lingling PU ,  Chih-Yu JEN ,  Chia Wen LIN

IPC: G06T7/00

Abstract: An automatic defect classification method may include obtaining a set of image data comprising a set of candidate defects from an inspection tool, developing a plurality of defect review types and a plurality of nuisance review types, and classifying the set of candidate defects according to the defect review types and nuisance review types using a machine learning classifier. Using the plurality of nuisance review types in the classification method reduces a nuisance rate.