IoT Sensor technologies to Address Issues of ASEAN Region
Prof. Dr. Koichiro Ishibashi
The University of Electro-Communications, Tokyo, Japan
Trillion sensor universe in which trillions of sensors are distributed to gather the data of the internet is expected in early 2020s, thereby addressing various issues such as agriculture, aquaculture, environment, energy, healthcare, and so on. These are many possibilities that these IoT sensor technologies play important roles on addressing various issues on ASEAN region, where various issues have been occurred. This paper introduces technologies to realize IoT sensors, and some application examples using the IoT sensors which could address the issues on those countries.
Water quality monitoring by sensors in shrimp firms: Sensor networks are adopted to monitoring water quality of shrimp firms in Vietnam. Long term data from shrimp firms in various districts of south in Vietnam can be acquired by the system. Mechanisms of the water quality change can be analyzed, and we show that cost of firming shrimps can be reduced.
Vital sign acquisition by contactless sensors: We can acquire vital signs of human by observing of skin-surface movement with 24GHz Doppler radar. The output signal of the radar is analyzed by peak detection algorithm to get Heat Beat(HB) and Respiration Rate(RR) so that the data is utilized to screening system of infection diseases. This system could be applicable to prevent predicted pandemic of infection diseases. Data acquisition from more than 400 patients was done in the National Hospital of Tropical Diseases in Hanoi in Aug. 2017, when dengue fever has been one of big issue in ASEAN countries.
Beat sensor technology for analog edge computing: We have proposed “Beat Sensors” as ultra-low power IoT sensors. Data of the sensor is wirelessly transmitted by the interval time of transmitted ID code. Analog edge computing can be executed, resulting in ultra - low communication power. Power, Temperature, and DC current Beat Sensors have been presented up to now. We are planning to develope various types of beat sensors to address issues on energy, environment, healthcare, and disaster and so on.
Prof. Dr. Koichiro Ishibashi has been a professor of The University of Electro-
Communications, Tokyo, Japan since 2011. He has been serving a guest professor at Ho Chi
Minh City University of Technology and Ho Chi Minh City University of Science since 2012.
He received PH. D degree from Tokyo Institute of Technology in 1985. He joined
Central Research Laboratory, Hitachi Ltd. in 1985, where he had investigated low power
technologies for Super H microprocessors and high density SRAMs. From 2004 to 2011, he was
in Renesas Electronics where he developed low power IPs mainly for SOCs used in mobile
He has presented more than 150 academic papers at international conferences and
journals. He edited and authored a book entitled Low Power and Reliable SRAM Memory Cell
and Array Design
from Springer, and more than 7,000 sections of the book have been
downloaded in total. He was awarded R&D 100 for the development of SH4 Series
Microprocessor in 1999. He is a member of IEICE and a Fellow of IEEE.
His current interests are IoT technologies including Ultra low power LSI design technology,
Technologies for energy harvesting sensor networks and applications, and Bio sensor
A Study of Noise Robust Speech Recognition on ROBOT
Prof. Dr.Eng. Yoshikazu Miyanaga
Hokkaido University, Sapporo, Japan
This topic introduces the design of a noise robust speech recognition on a communication robot. For the valuable systems as communication robots, the high performance of a strong noise robust speech recognition should be implemented. Although the dialog mechanism as human communications is one of interesting features into the robot, the command-based communications to the robot must be more important. When we consider the development of a command-based automatic speech recognition (ASR) system, a noise robust ASR should be also considered against various noise circumstances.
In this presentation, some advanced speech analysis techniques have been introduced as noise reduction techniques. In order to develop the robustness under low SNR, the different kinds of running spectrum analysis (RSA) have been proposed and they focus on the speech feature adjustment with an important speech components by reducing any noise components.
Even if these proposed algorithms are applied to noisy speech, it may be difficult to recognize several similar pronunciation phrases. Discrimination of similar pronunciation phrases is more difficult than that of normal phrases under low SNR circumstances. In this topic, we also propose the optimum noise reduction RSA for such similar pronunciation phrases. With the combination of various kinds of RSA, it can be shown that the proposed system can improve speech recognition. Evaluation results demonstrate that the proposed system achieves better speech recognition performance at low SNR conditions.
Prof. Dr.Eng. Yoshikazu Miyanaga received the B.S., M.S., and Dr. Eng. degrees from Hokkaido University, Sapporo, Japan, in 1979,
1981, and 1986, respectively. Since 1983 he has been with Hokkaido University. He is now
Professor at Division of Information Communication Systems in Graduate School of Information
Science and Technology, Hokkaido University. He is also the dean of Graduate School of
Information Science and Technology, Hokkaido University (2014-present). From 1984 to 1985, he
was a visiting researcher at Department of Computer Science, University of Illinois, USA. He is also
the adjunct professors of King Mongkut's University of Technology Thonburi (KMUTT, 2014-
present) and University of Technology Sydney (UTS, 2016-present).
Study on the comparison of the difference of human behavior between the real and the virtual environment using HMD
Prof. Dr. Kazuhiko Hamamoto
Tokai University, Japan
A purpose of this research is to investigate a difference of human cognitive behavior (in other word, ecological cognition) in VR using HMD and the real environment and to examine a method to compensate the difference. Therefore, this research conducts a comparative experiment using a cognitive behavior called “Stepping-over” and “Passing-under”. First, a subject estimate whether the subject can step over a rope which is set in a height and set at 2 meters away from the subject. After that, the subject comes close to the rope and estimates the same thing again, and if the estimation is “can step over”, the subject tries to step over the rope. The result of success or failure is investigated in the virtual space and the real space. As the result, there was no difference between the prior estimation and the actual behavior in both the virtual and the real spaces. However, in simplified virtual environment where almost of all environmental information is removed, the number of the subject who makes the difference between the prior estimation and the actual behavior was increased. The result shows the human cognitive behavior comes from the environmental information. As the result, it was suggested that an amount of environmental information in the virtual space is important to realize the same human cognitive behavior as one in the real space.
Kazuhiko HAMAMOTO, was born in Nagasaki prefecture, Japan in 1966. He received B.D, M.D and Ph.D from Tokyo University of Agriculture and Technology in 1989, 1991 and 1994 respectively. He was assistant professor in Dept. of Communications Eng., Tokai University in 1994, Associate Professor in 1999, and Professor in Dept. of Information Media Technology, School of Information and Tele-communication Eng., Tokai University in 2009 and Currently, he is the Dean of School of Information and Tele-communication Eng., Tokai University. His research interest is information architecture, especially, medical image processing, human interface and virtual reality. He joins IEICE, IEEJ, IEEE, VRSJ, JSST, etc. He is a member of Technical Committee of Medical and Biological Engineering, Society of Electronics, Information and Systems, IEEJ.
Intelligent Robot Technologies for Care and Lifestyle Support
Prof. Dr. Jun Miura
Toyohashi University of Technology, Japan
Care and lifestyle support robot is one of the promising areas to which robotic technologies can be applied. As we are facing the aged society, the need for robots that can help the elderly and/or the disabled in their daily lives is increasing. Supporting caregivers by taking over their routine works and making them focus on more important activities is also an important application. I have been working on the mobile robot domain and the related technologies can be applied to various tasks in care and lifestyle support. Example tasks are: fetching a user-specified object in a distant room and moving around in a hospital for checking the residents' status.
In this talk, I first survey representative mobile robot technologies such as SLAM (simultaneous localization and mapping), people detection and tracking, and object and scene recognition. I also show several robot systems built on a combination of these technologies.
I then present three of our recent results which could be used for care and lifestyle support. The first one is a multiple feature-based person re-identification method, which is effective in tracking a specific person in crowded environments. The second one is an estimation of persons' unusual poses (e.g., incumbent poses) using depth images. An effective way of generating a large amount of training data for deep neural networks is presented. The last one is face recognition under various illumination conditions. We develop an illumination normalization method which can generate consistent face images irrespective of illumination conditions.
Jun Miura is a Professor at Department of Computer Science and Engineering, Toyohashi University of Technology (TUT). He received PhD in Information Engineering from the University of Tokyo in 1989. After having spent 18 years at Department of Mechanical Engineering, Osaka University, as an Assistant Professor and an Associate Professor, he joined TUT in 2007. Prof. Miura has published over 190 papers in the areas of mobile robots, robot vision, human robot interaction, and robotic applications to lifestyle support. He has also received many awards including the RSJ Best Paper Award in 1997, ICRA-1995 Best Paper Award Finalist, and ICRA-2013 Best Service Robotics Paper Award Finalist.