Development of driving support system for electric vehicle by using image processing technology

Yuxiang Sun,Nan Wu,Shigeyuki Tateno,Harutoshi Ogai 제어로봇시스템학회 2012 제어로봇시스템학회 국제학술대회 논문집 Vol.2012 No.10

Aiming to realize the safe driving for the elderly people and novice drivers, intelligent vehicle technologies are becoming an important research topic in recent years. While driving, sometimes they cannot handle the car in an emergency due to the poor concentration and reaction. In fact, there are a lot of car accidents caused by such people who occasionally step on the accelerator instead of the brake unexpectedly when intending to have an emergency stopping. This paper describes a recognition method for recognizing driving situation. For supporting driving by traffic situation recognition, real-time detection and robustness are increasingly required. In indoor experiments, the RoboCar was used to verify the recognition methods. The algorithm which determines the routes are able to run safety by the CCD camera was used to recognize the white line on both sides of the road. And the image matching algorithm was used to recognize the traffic signs on the routes. The experimental result with high accuracy was successfully obtained.

Comparison and Performance Analysis of App and Web-based One-stop Attendance Management

Sun YuXiang,YoungWook Cha,ChoonHee Kim 보안공학연구지원센터 2015 International Journal of u- and e- Service, Scienc Vol.8 No.12

This paper presents the structure of an electronic attendance system composed of OAA (One-stop Attendance and Absence)-server and client. We compare simultaneous check, fraudulent attendance, prevention of leaving the classroom, and extra equipment according to check methods, and also present a method of app and web-based attendance management. NFC and BLE-based app and location-based web for attendance management were applied to classrooms. Issues with electronic attendance management and resolution methods were also presented. RESTful web service was defined for attendance management and performance analysis was conducted according to thread numbers of OAA-server and parser types of XML and JSON.

Intelligent Auxiliary Diagnostic Architecture based on FHIR

Yuxiang Sun(순위샹),Jehyun Jung(정제현),Yongju Lee(이용주) 한국정보과학회 2020 한국정보과학회 학술발표논문집 Vol.2020 No.7

Entity Classification Based on Graph Convolutional Networks for Knowledge Graphs

Yuxiang Sun(순위샹),Hongzhu Duan(단홍주),Yongju Lee(이용주) 한국정보기술학회 2022 Proceedings of KIIT Conference Vol.2022 No.6

지식 그래프의 엔티티 분류는 엔터티를 연속적인 저차원 공간에 투영하고 엔티티의 벡터 값을 기반으로 클러스터링을 수행하는 것을 목표로 한다. 본 논문에서는 엔터티를 2차원 공간에 임베드하기 위한 그래프 컨볼루션 네트워크 모델을 제안하고, 밀도 기반 공간 클러스터링 알고리즘을 사용하여 다양한 임베딩 모델의 클러스터링 성능을 평가한다. 기존의 임베딩 기반 접근 방식과 비교하여 제안된 그래프 컨볼루션 네트워크 모델은 더 나은 클러스터링 성능을 보여주고 이의 우수한 엔티티 표현 능력도 증명된다. The entity classification of knowledge graphs aims to project entities into a continuous low-dimensional space and perform clustering based on the vector values of entities. In this paper, we proposed a graph convolutional network model to embed entities into a two-dimensional space. Afterward, a density-based spatial clustering algorithm is utilized to evaluate clustering performance of different embedding models. Compare to traditional embedding-based approaches, the proposed graph convolutional network model shows a better clustering performant, its excellent entity representation ability is also proved.

Research on Action Strategies and Simulations of DRL and MCTS-based Intelligent Round Game

Yuxiang Sun,Bo Yuan,Yongliang Zhang,Wanwen Zheng,Qingfeng Xia,Bojian Tang,Xianzhong Zhou 제어·로봇·시스템학회 2021 International Journal of Control, Automation, and Vol.19 No.9

The reinforcement learning problem of complex action control in multiplayer online battlefield games has brought considerable interest in the deep learning field. This problem involves more complex states and action spaces than traditional confrontation games, making it difficult to search for any strategy with human-level performance. This paper presents a deep reinforcement learning model to solve this problem from the perspective of game simulations and algorithm implementation. A reverse reinforcement-learning model based on high-level player training data is established to support downstream algorithms. With less training data, the proposed model is converged quicker, and more consistent with the action strategies of high-level players’ decision-making. Then an intelligent deduction algorithm based on DDQN is developed to achieve a better generalization ability under the guidance of a given reward function. At the game simulation level, this paper constructs Monte Carlo Tree SearchIntelligent Decision Model for turn-based antagonistic deduction games to generate next-step actions. Furthermore, a prototype game simulator that combines offline with online functions is implemented to verify the performance of proposed model and algorithm. The experiments show that our proposed approach not only has a better reference value to the antagonistic environment using incomplete information, but also accurate and effective in predicting the return value. Moreover, our work provides a theoretical validation platform and testbed for related research on game AI for deductive games.

First Principles Investigation of the Mechanical, Thermodynamic and Electronic Properties of FeSn5 and CoSn5 Intermetallic Phases under Pressure

Wenming Sun,Jing Liu,Hong Wang,Zhenwei Zhang,Liang Zhang,Yuxiang Bu 한국물리학회 2017 THE JOURNAL OF THE KOREAN PHYSICAL SOCIETY Vol.70 No.4

For guidance for developing Fe/Co-Sn-based anode materials for lithium-ion batteries, the mechanical, thermodynamic and electronic properties of FeSn5 and CoSn5 intermetallic phases under pressures ranging from 0 to 30 GPa have been investigated systematically using first-principles total-energy calculations within the framework of the generalized gradient approximation. The pressure was found to have significant effects on the mechanical, thermodynamic and electronic properties of these compounds. In the selected pressure range, CoSn5 has a more negative formation enthalpy than FeSn5. Based on the calculated elastic constants, the bulk modulus, shear modulus, and Young’s modulus were determined via the Viogt-Reuss-Hill averaging scheme. The variations of specific heats at constant volume for FeSn5 and CoSn5 in a wide pressure (0 - 30 GPa) and temperature (0 - 1000 K) range are also predicted from phonon density of states calculation. The calculated results suggested that both FeSn5 and CoSn5 are mechanically stable at pressure from 0 to 30 GPa. FeSn5 is dynamically stable at pressure up to, 30 GPa, at least, however, CoSn5 is dynamically stable no higher than 15 GPa.

Numerical investigation of the dynamic response of a simplified floating structure subjected to two underwater explosion bubbles

Xudong Jiang,Yuxiang Wang,Cheng Sun 대한기계학회 2023 JOURNAL OF MECHANICAL SCIENCE AND TECHNOLOGY Vol.37 No.11

The interaction of the two coupling bubble with a floating structure is modelled to investigate the dynamic behavior of the warship under an underwater explosion (UNDEX) by coupled Eulerian-Lagrangian (CEL) method. In this study, the CEL method is first verified by the experimental results for free field underwater explosion. Then, a series of cases for various initial distance and delay intervals between the bubble pair are implemented to analyze the deformation and damage modes of the floating structure. It is observed that for large initial distance between the bubble pair synchronously generated, the floating structure can be damaged with longitudinal bending buckling in the middle domain. When the initial distance between the bubble pair decreases to a critical value, the floating structrure is damaged with a remarkable crevasse besides the longitudinal bending buckling in the middle domain. However, under two bubbles with different phase differences, the floating structure a the deformation of the whipping motion without remarkable local structural damage. This is because there is the conspicuous despression effect between two out-of-phase bubbles, which weakens the bubble pulsation load. These phenomena present a theoretical reference for further studies on the destruction of underwater explosion on the ship structure caused by multiple bubbles generated with various phase differences.

Sample Preparation and Stability of Human Serum and Urine Based on HPLC-DAD for Metabonomics Studies

Liu, Yun,Sun, Xiaoming,Di, Duolong,Feng, Yuxiang,Jin, Fengling Korean Chemical Society 2012 Bulletin of the Korean Chemical Society Vol.33 No.7

Many literatures focus on the biological relevance and the identification of biomarkers for disease activity assessment while less attention has been paid to the development of standard procedures for sample preparation and storage based on liquid chromatography technique. The influencing factors including protein precipitation, storage temperature, storage time, and reconstitution by ultra pure water were analyzed employing HPLC-DAD. The effects were investigated from five participants over three months by principal components analysis (PCA) and the values of percent changes (PC). The samples with protein precipitation might slow the rate of bacterial enzymatic conversion. After protein precipitation, the average PC of urine samples ($0.136{\pm}0.013$, n = 5) is relatively less than that of the serum samples ($0.173{\pm}0.026$, n = 5) for three months. Minimal effects on metabolic profiles of serum and urine (PC < 0.15) are reasonable for metabolomic studies after protein precipitation and storage at $-20^{\circ}C$ for two months.

Implications of Sarcopenia and Glucometabolism Parameters of Muscle Derived From Baseline and End-of-Treatment 18F-FDG PET/CT in Diffuse Large B-Cell Lymphoma

Tan Xiaoyue,Sun Xiaolin,Chen Yang,Wang Fanghu,Shang Yuxiang,Zhang Qing,Yuan Hui,Jiang Lei 대한영상의학회 2024 Korean Journal of Radiology Vol.25 No.3

Objective: We previously found that the incidence of sarcopenia increased with declining glucose metabolism of muscle in patients with treatment-naïve diffuse large B-cell lymphoma (DLBCL). This study aimed to investigate the relationship between sarcopenia and muscle glucometabolism using 18F-FDG PET/CT at baseline and end-of-treatment, analyze the changes in these parameters through treatment, and assess their prognostic values. Materials and Methods: The records of 103 patients with DLBCL (median 54 years [range, 21–76]; male:female, 50:53) were retrospectively reviewed. Skeletal muscle area at the third lumbar vertebral (L3) level was measured, and skeletal muscle index (SMI) was calculated to determine sarcopenia, defined as SMI < 44.77 cm2/m2 and < 32.50 cm2/m2 for male and female, respectively. Glucometabolic parameters of the psoas major muscle, including maximum standardized uptake value (SUVmax) and mean standardized uptake value (SUVmean), were measured at L3 as well. Their changes across treatment were also calculated as ΔSMI, ΔSUVmax, and ΔSUVmean; Δbody mass index was also calculated. Associations between SMI and the metabolic parameters were analyzed, and their associations with progression-free survival (PFS) and overall survival (OS) were identified. Results: The incidence of sarcopenia was 29.1% and 36.9% before and after treatment, respectively. SMI (P = 0.004) was lower, and sarcopenia was more frequent (P = 0.011) at end-of-treatment than at baseline. The SUVmax and SUVmean of muscle were lower (P < 0.001) in sarcopenia than in non-sarcopenia at both baseline and end-of-treatment. ΔSMI was positively correlated with ΔSUVmax of muscle (P = 0.022). Multivariable Cox regression analysis showed that sarcopenia at end-of-treatment was independently negatively associated with PFS (adjusted hazard ratio [95% confidence interval], 2.469 [1.022–5.965]), while sarcopenia at baseline was independently negatively associated with OS (5.051 [1.453–17.562]). Conclusion: Sarcopenic patients had lower muscle glucometabolism, and the muscular and metabolic changes across treatment were positively correlated. Sarcopenia at baseline and end-of-treatment was negatively associated with the prognosis of DLBCL.

A Monocular Ranging Algorithm Combining Visual Markers and Convolutional Neural Networks

Nan Yin(인난),Yuxiang Sun(순위샹),Hyunsoo Kim(김현수),Youngin Kim(김영인),Jae Soo Kim(김재수) 한국통신학회 2023 한국통신학회 학술대회논문집 Vol.2023 No.11