公开(公告)号：US20230343078A1

公开(公告)日：2023-10-26

申请号：US18304508

申请日：2023-04-21

Applicant: IMEC VZW

Inventor： Bappaditya Dey ,  Enrique Dehaerne ,  Sandip Halder

IPC: G06V10/776 ,  G06V10/77 ,  G06V10/70 ,  G06V10/82

CPC classification number: G06V10/776 ,  G06V10/7715 ,  G06V10/87 ,  G06V10/82

Abstract: The present disclosure related to a computer-implemented training and prediction method for defect detection, classification and segmentation in image data. The training method comprises providing an ensemble of learning structures, each learning structure comprising a feature extractor module, a region proposal module, a detection module, and a segmentation module. Each learning structure is trained individually and validated. Learning structures whose validation prediction score exceeds a predetermined threshold score are selected and their predictions combined, using a parametrized ensemble voting structure.

公开(公告)号：US20250076865A1

公开(公告)日：2025-03-06

申请号：US18462172

申请日：2023-09-06

Applicant: IMEC VZW ,  Katholieke Universiteit Leuven

Inventor： Bappaditya Dey ,  Enrique Dehaerne ,  Sandip Halder

IPC: G05B23/02 ,  G05B19/18

Abstract: A federated machine learning method is provided. The method includes providing, from a central model server, an initial trained machine learning (ML) model to a plurality of clients as a respective local ML model. The initial trained ML model is configured to identify defect features from scanning electron microscopy (SEM) images. The method additionally includes receiving, from at least one client by the central model server, information indicative of a respective updated local ML model. The method also includes determining, based on the information indicative of the respective updated local ML models, an updated global ML model.

公开(公告)号：US12243193B2

公开(公告)日：2025-03-04

申请号：US17366350

申请日：2021-07-02

Applicant: IMEC VZW

Inventor： Bappaditya Dey ,  Sandip Halder ,  Gouri Sankar Kar ,  Victor M. Blanco ,  Senthil Srinivasan Shanmugam Vadakupudhu Palayam

IPC: G06T5/70 ,  G06T5/50

Abstract: The disclosure relates generally to image processing. For example, the invention relates to a method and a device for de-noising an electron microscope (EM) image. The method includes the act of selecting a patch of the EM image, wherein the patch comprises a plurality of pixels, wherein the following acts are performed on the patch: i) replacing the value of one pixel, for example of a center pixel, of the patch with the value of a different, for example randomly selected, pixel from the same EM image; ii) determining a de-noised value for the one pixel based on the values of the other pixels in the patch; and iii) replacing the value of the one pixel with the determined de-noised value.

公开(公告)号：US20250076866A1

公开(公告)日：2025-03-06

申请号：US18462181

申请日：2023-09-06

Applicant: IMEC VZW ,  Katholieke Universiteit Leuven

Inventor： Bappaditya Dey ,  Enrique Dehaerne ,  Sandip Halder

IPC: G05B23/02 ,  G05B19/18

Abstract: A method for training a local machine learning model is provided. The method includes receiving a scanning electron microscope (SEM) image of semiconductor features. The method additionally includes determining a location and dimensions of a bounding box within the SEM image. The method yet further includes determining, whether a defect feature exists within the bounding box, based on an unsupervised object detection process. The method also includes, if the defect feature exists within the bounding box, receiving positive rewards. The method also includes, if the defect feature does not exist within the bounding box, receiving negative rewards.

公开(公告)号：US20220076383A1

公开(公告)日：2022-03-10

申请号：US17366350

申请日：2021-07-02

Applicant: IMEC VZW

Inventor： Bappaditya Dey ,  Sandip Halder ,  Gouri Sankar Kar ,  Victor M. Blanco ,  Senthil Srinivasan Shanmugam Vadakupudhu Palayam

IPC: G06T5/00

Abstract: The disclosure relates generally to image processing. For example, the invention relates to a method and a device for de-noising an electron microscope (EM) image. The method includes the act of selecting a patch of the EM image, wherein the patch comprises a plurality of pixels, wherein the following acts are performed on the patch: i) replacing the value of one pixel, for example of a center pixel, of the patch with the value of a different, for example randomly selected, pixel from the same EM image; ii) determining a de-noised value for the one pixel based on the values of the other pixels in the patch; and iii) replacing the value of the one pixel with the determined de-noised value.