Three projects from BASICS Farm System at Dongguk University
AS6 Level 5
MR6, AS6-05-10
Abstract:
TEAM UAHAN
To enhance students' access to information such as school notices and facilities usage, we developed a smart campus information dissemination service by recycling underutilized DIDs(Digital Information Display) on the campus. We attached cameras on the DID to count the number of people watching the information on the display. Then, we used a machine-learning algorithm to predict the degree of interest on the specific information to determine the length and frequency of displayed information. For future work, we will improve the performance of the system by including more factors such as users' pupil dilation, body reaction, and time spent on the display.
Name/ Major/ Grade
Choi, Sung June/ Computer Science and Engineering/ Rising Senior
Chang, Hyo Jung/ Computer Science and Engineering/ Rising Senior
Yang, Jin Na/ Computer Science and Engineering/ Rising Senior
Jeong, Young Jun/ Industrial and System Engineering/ Rising Junior
TEAM SMART BABY
We created a shared electric kickboard service called "Quick Dongguk" to realize the idea of collaborative economy on the campus. The hilly terrain of the Dongguk University campus makes the people on the campus hard to move fast. To alleviate this discomfort, we came up with the service near Chungmuro subway station in Seoul. What we thought crucial is that people overlook wearing helmets which is not only illegal but also dangerous. Thus, we designed a locking device which opens only when the user wears a helmet. An Arduino ultrasound sensor attached to the helmet determines whether it is worn and notifies the Raspberry Pi that controls the locking device using a servo motor, only if the given authentication number matches. For the payment system, we created Dongguk coin based on Ethereum.
Name/ Major/ Grade
Kim, Da Yeon/ Information Communication Engineering/ Rising Junior
Baek, Seung Yeon/ Information Communication Engineering/ Rising Junior
Lim, Hea Min/ Information Communication Engineering/ Rising Junior
Suk, Jae Wook/ Information Communication Engineering/ Rising Junior
Hwang, Chae Hwan/ Information Communication Engineering/ Rising Junior
TEAM YoLo POSE
We introduce a service called "Home PT" that combines deep learning with image processing. It provides a real-time feedback on the posture and speed of the user during exercise to improve the exercise experience. The correction service uses our pose estimation model based on YOLOv3, a real-time object detection model. We resolved the reversal problem of the existing pose estimation models by analyzing the characteristics of a set of sample exercise video data. In addition, we identify the most important factors for proper exercise postures based on kinematics and statistics to provide high-quality feedback.
Name/ Major/ Grade
Kim, Sung Jin/ Mathematics/ Rising Senior
Lee, Tae Min/ Mathematics/ Expected to graduate
Jeong, Ui Jeong/ Computer Science and Engineering/ Rising Senior
Chae, Jin Yeong/ Statistics/ Expected to graduate
Biodata:
The three projects are supervised by Professor Dongho Kim at Dongguk University who received his B.S. degree in Computer Engineering from Seoul National University, Korea in 1990 and his M.S. and Ph.D. degrees in Computer Science from University of Southern California, Los Angeles, California, U.S.A., in 1992 and 2002, respectively. He is now a professor of Dongguk Institute of Convergence Education at Dongguk University, Seoul, Korea. His research interests include distributed systems, networks and security.