http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
2차원 영상에서의 깊이 추정을 위한 객체 인식 제어 방법
장서윤(Seo-Yun Jang),이상열(Sangyul Lee) 제어로봇시스템학회 2022 제어·로봇·시스템학회 논문지 Vol.28 No.4
In this paper, we propose a method for controlling a manipulator using a single 2D camera and an experimental system. Typically, to control the manipulator, the camera is constantly moved using a motor to navigate or detect the reference object in a fixed frame. The proposed control method does not require conventional high-performance depth cameras and can derive depth values from images acquired using a low-cost 2D camera. A real-time image frame detects objects that are the criteria and configures the learning model and communication node to draw an object-bounding box accordingly. In this work, we convert the change in size of this bounding box into a depth value. When the depth value is thus derived from the 2D image, the target position value of the manipulator can be determined. Finally, the manipulator system is configured with detection control.
기계 학습과 유전 알고리즘을 결합한 파라핀 도포 균일도 최적화
박정준(Jeong Jun Park),김진우(Jin Woo Kim),장서윤(Seo Yun Jang),이상열(Sang Yul Lee) 대한기계학회 2024 大韓機械學會論文集A Vol.48 No.6
본 연구에서는 자동화된 파라핀 도포 공정의 균일도를 최적화하기 위한 기계학습과 최적화 알고리즘을 결합한 기법을 제안한다. 먼저 파라핀의 도포율 예측 모델을 개발하기 위해 실제 공정에 온 · 습도 센서를 설치하였다. 공정의 말단에 3대의 카메라를 설치하고 영상처리 알고리즘을 결합하여 도포율을 수치화하였다. 실제 수집된 온 · 습도와 수치화된 도포율을 기반으로 랜덤 포레스트 모델을 구축하였다. 이후 학습된 기계학습 모델에 최적화 기법인 유전 알고리즘을 적용하여 높은 파라핀 도포 균일도를 달성하기 위해 실제 공정에서 변경 가능한 파라미터를 최적화하였다. 최적화된 온도와 실제 수집된 온도와의 오차는 각 1.1°C, 4.0°C로 매우 낮았다. 이는 실제 시나리오에서의 유용성을 검증하며, 파라핀을 이용한 다양한 공정 및 시스템에 유용성과 가치가 있을 것으로 판단된다. In this study, we propose a method that combines machine learning and optimization algorithms to optimize the uniformity of the automated paraffin coating process. Initially, to develop a model predicting the rate of paraffin application, temperature and humidity sensors were installed in the actual process. At the end of the process, three cameras were installed, and an image processing algorithm was integrated to quantify the application rate. Based on the actual collected temperature and humidity data and the quantified application rate, a random forest model was constructed. Subsequently, an optimization technique, the genetic algorithm, was applied to the trained machine learning model to optimize parameters that could be changed in the actual process to achieve high paraffin coating uniformity. The errors between the optimized temperature and the actual collected temperature were extremely low, at 1.1°C and 4.0°C, respectively. This validates its usefulness in real-world scenarios, suggesting its potential value and utility in various processes and systems utilizing paraffin.
장서윤,최선남,안지연,박인전,안옥희 영남대학교 자원문제연구소 1996 資源問題硏究 Vol.15 No.-
The purposes of this study were to investigate the degree of using study room and level of the satisfaction about it, and to explore the variables concerned with the satisfaction of using study room. For these purposes, the data were collected by using questionaire distributed to 186 adolescents living in Taegu and Andong city. The data were analyzed using descriptive statistics,X2-test, one way ANOVA and multiple regression analysis. The results of the research were as follows ; 1. 38% of adolescent had their own study room, and the rest of them didn't have their own study room, but used their study room with other people such as sibiling or relatives. 2. They usually use their room during 9 hours or 10 hours per a day. 3. The level of satisfaction using their study room was ranked by middle satisfaction, a little satisfaction, and great satisfaction respectively. 4. Among the independent variables, good landscope, quiet surrounding, duration of staying in home and study room had significant effect on the satisfaction of using study room.