Computer Vision Image Processing

Toan-Khoa Nguyen +84-853******

# adx983@r.postjobfree.com

# adx983@r.postjobfree.com

§ GitHub Profile

ï LinkedIn Profile

Education

•National Taiwan University of Science and Technology, Taiwan 2020-2022 Master of Science in Electrical Engineering - Specialization in Anomaly Object Segmentation GPA: 4.26/4.3

•Ho Chi Minh University of Technology, Vietnam 2013-2018 Bachelor of Automotive Engineering

Experience

•R&D Samsung SDS 12/2022-Now

AI Research Engineer Hanoi, Vietnam

– Working as a research engineer in the Vision Inspection AI team related to artificial intelligence applications including computer vision, image processing, and machine vision.

– Proposing and developing solutions for checking anomaly defection and detecting missing screws, tight connector inspection, scatter screws detection and laser printed defect inspection on mainboard for laptop manufacturing at Samsung Electronics Vietnam.

– Led the execution of computer vision projects for manufacturing clients, resulting in 25% efficiency gains and 95% defect reduction.

– Collaborated with hardware, software developers, and manufacturing specialists to deliver high-quality projects on schedule.

– Communicated with clients, providing regular updates, resolving issues, and building strong relationships.

•Autonomous and Soft Robotics Lab 8/2020-7/2022

Research Assistant National Taiwan University, Taiwan

– Research topics: segmentation technologies for autonomous mobile robots.

– Developed self-supervised learning methods for drivable area and road anomalies segmentation for mobile robots.

– Proposed automatic system to generate segmentation labels for drivable area and road obstacles based on disprep- ancy network with RGB-D data as inputs to reduce labor and time consuming.

– Built up real-time drivable area and road anomalies segmentation using OAK-D camera, which resulted in the winner list phase 1 of OpenCV competion in 2021, and got a publication on MDPI Sensors.

•National Taiwan University of Science and Technology Teaching Assistant

– IoT Programming and Practice Course - Instructor: Prof. Minh-Quang Tran

– Fundamental of Self-Driving Cars Course - Instructor: Prof. Shu-Hao Liang Projects & Involvements

•Anomaly Scatter Screw and Missing Screw Detection 12/2022-4/2023 Inspection system for laptop manufacturing at Samsung Electronics Vietnam, R&D Samsung SDS

– Collected data and developed missing screw detection, tight connector inspection, and anomaly scatter screw framework based on small object detection and image classification approach by doing the template matching and scanning area to area using tilling approach.

– Working on 5 lines and 2 sections at manufacturing site, which resulted in 25% efficiency gains and 95% defect reduction.

– Tools & technologies used: Python, Pytorch, ONNX

•Laser Printed Defect Detection 4/2023-Now

Inspection system for laptop manufacturing at Samsung Electronics Vietnam, R&D Samsung SDS

– Constructed lights and cameras for inspection system which is proper for a variety of colors of texts on laptops and laptop types.

– Collected data and eveloped an approach using feature matching based on template laptop set, similarity and structural similarity index measure, and perceptual loss for defect detection.

– Doing research on semi-supervised learning for defect segmentation based on memory-based approach.

– Tools & technologies used: Python, Pytorch, ONNX Technical Skills

OS System: Windows, Linux

Languages: Python, MATLAB

Frameworks: OpenCV, Tensorflow, Pytorch, Git, Docker Analytical Skills

Machine Learning, Regression, Image Processing, Object Detection, Image Segmentation, Neural Networks Honors and Awards

•Phase 1 Finalist, OpenCV Competion 2021

•Full scholarship, National Taiwan University of Science and Technology 2020 Publications

•Toan-Khoa Nguyen, Phuc Thanh-Thien Nguyen, Dai-Dong Nguyen, Chung-Hsien Kuo. Effective Free-driving Region Detection for Mobile Robots by Uncertainty Estimation Using RGB-D Data. MDPI Sensors, IF: 3.847.

•Minh-Quang Tran, Meng-Kun Liu, Quoc-Viet Tran, Toan-Khoa Nguyen. Effective Fault Diagnosis Based on Wavelet and Convolutional Attention Neural Network for Induction Motors. IEEE Transactions on Instrumentations and Measurement, IF: 5.6.

Contact this candidate