公开(公告)号：US20250045904A1

公开(公告)日：2025-02-06

申请号：US18230127

申请日：2023-08-03

Applicant: Applied Materials Israel Ltd.

Inventor： Gilad VERED ,  Dror ALUMOT ,  Rafael BISTRITZER ,  Hadar SHLOMAI-NAPARSTEK ,  Yarden ZOHAR

IPC: G06T7/00 ,  G06T7/10 ,  G06T7/50 ,  G06V10/46 ,  G06V10/764

Abstract: A system for examining a semiconductor specimen that includes a plurality of layers at respective different depths, and a plurality of holes. Each hole has a top portion at the surface of the specimen, and a bottom portion accommodated in one of the layers. The system includes a processing and memory circuitry (PMC) configured to provide an inspection image indicative of the holes, and process a hole image in the inspection image, without using a shape characterizing model. The processing includes segmenting the inspection image and determining data indicative of a contour of the top portion of the hole, and further segmenting the inspection image and determining data indicative of a contour of a shape enclosed within the contour of the top of the hole.

公开(公告)号：US20240153043A1

公开(公告)日：2024-05-09

申请号：US17983181

申请日：2022-11-08

Applicant: Applied Materials Israel Ltd.

Inventor： Tamir EINY ,  Dror ALUMOT ,  Yarden ZOHAR ,  Anna LEVANT

IPC: G06T5/00 ,  G06T5/50 ,  G06T7/30

CPC classification number: G06T5/002 ,  G06T5/50 ,  G06T7/30 ,  G06T2207/10016 ,  G06T2207/20081 ,  G06T2207/20212 ,  G06T2207/30148

Abstract: There is provided an image generation system and method. The method comprises obtaining a runtime image of a semiconductor specimen with a low Signal-to-noise ratio (SNR), and processing the runtime image using a machine learning (ML) model to obtain an output image with a high SNR. The ML model is previously trained using a training set comprising a plurality of low SNR images associated with a high SNR image. The plurality of low SNR images correspond to a plurality of sequences of frames acquired in a plurality of runs of scanning a first site of the specimen. The high SNR image is generated based on the plurality of low SNR images. The training comprises, for each low SNR image: processing the low SNR image by the ML model to obtain predicted image data, and optimizing the ML model based on the predicted image data and the high SNR image.