Undergraduate Research
Research Topics
Deep Learning Image Processing
3D LiDAR Calibration
Industrial AI
Smart Material System
Recent Works
[Deep Learning Image Processing]
YoffleNet: Light-Weight Object Detection for Embedded System
Application of a Lightweight CNN Model for Detecting Defects in Substation Insulators with Drone EO/IR Camera
변전소 불량애자 검출을 위한 드론 EO/IR 영상 기반의 경량화 객체탐지 모델 응용 사례
See more research on this topic
Light-Weight Object Detection with SORT for Object Tracking
YOffleNet Deep Sort in PyTorch
This repository contains a modified version of Pytorch YOLOv3 + Deep Sort (https://github.com/mikel-brostrom/Yolov3_DeepSort_Pytorch). It is a program that counts the number of people and predicts the direction of movement using the bounding box coordinates obtained by tracking algorithm.
This project was carried out as a 2021-1 MIP by H.J. Kim
Reference
See more research on this topic Reference:
[Vehicle LiDAR]
Extrinsic Calibration of Vehicle 3D LiDAR
Accurate Alignment Inspection System for Low-Resolution Automotive LiDAR
LiDAR Sensor Alignment Inspection System for Automobile Productions Using 1-D Photodetector Arrays
Reference
- You, Ji Hwan, Seontaek Oh, Jae Eun Park, Hyeongseok Song, and Young Keun Kim, ‘A Novel LiDAR Sensor Alignment Inspection System for Automobile Productions Using 1-D Photodetector Arrays’, Measurement, 183 (2021), 109817 https://doi.org/10.1016/J.MEASUREMENT.2021.109817
- Oh, Seontaek, Ji-Hwan You, Azim Eskandarian, and Young-Keun Kim. “Accurate Alignment Inspection System for Low-Resolution Automotive LiDAR.” IEEE Sensors Journal 21, no. 10 (2021): 11961-11968.
See more research on this topic
[Industrial AI]
Prediction of Parking Free Space
포항시 스마트 시티 주차안내 시스템 - 주차 여유 공간 예측 알고리즘 개발: WeSEE X HGU SSS Lab
Unsupervised Defect Detection for Wire Rod, Steel Industry
Abstract in English and Korean
See more research on this topic
Vibration Control with Smart Material System
Reinforcement Learning based Stiffness Control of a Tunable Vibration Absorber
Abstract in English and Korean See more research on this topic