Invention Grant
- Patent Title: Neural network training device, system and method
-
Application No.: US16687349Application Date: 2019-11-18
-
Publication No.: US11200659B2Publication Date: 2021-12-14
- Inventor: Laurent Bidault
- Applicant: STMICROELECTRONICS (ROUSSET) SAS
- Applicant Address: FR Rousset
- Assignee: STMICROELECTRONICS (ROUSSET) SAS
- Current Assignee: STMICROELECTRONICS (ROUSSET) SAS
- Current Assignee Address: FR Rousset
- Agency: Seed IP Law Group LLP
- Main IPC: G06T7/00
- IPC: G06T7/00 ; G06K9/62 ; G06N3/04 ; G06N3/08

Abstract:
A device includes image generation circuitry and convolutional-neural-network circuitry. The image generation circuitry, in operation, generates a digital image representation of a wafer defect map (WDM). The convolutional-neural-network circuitry, in operation, generates a defect classification associated with the WDM based on: the digital image representation of the WDM and a data-driven model associating WDM images with classes of a defined set of classes of wafer defects and generated using a training data set augmented based on defect pattern orientation types associated with training images.
Public/Granted literature
- US20210150696A1 NEURAL NETWORK TRAINING DEVICE, SYSTEM AND METHOD Public/Granted day:2021-05-20
Information query