Faculty Profiles - KAWAGUCHI, Nobuo

HASC-PAC2016: Large scale human pedestrian activity corpus and its baseline recognition Reviewed

FloodEye: Real-time Flash Flood Prediction System for Urban Complex Water Flow

Hiroi Kei, Kawaguchi Nobuo

2016 IEEE SENSORS page： . 2016

Haruyuki Ichino, Katsuhiko Kaji, Ken Sakurada, Kei Hiroi, Nobuo Kawaguchi

UbiComp 2016 Adjunct - Proceedings of the 2016 ACM International Joint Conference on Pervasive and Ubiquitous Computing page： 705 - 714 2016

Language：Japanese Publishing type：Research paper (international conference proceedings) Publisher：Association for Computing Machinery, Inc

Human activity recognition by wearable sensors will enable a next-generation human-oriented ubiquitous computing. However, most of the existing research on human activity recognition is based on a small number of subjects, and lab-created-data. To overcome this problem, we hold HASC Challenge as a technical challenge to collect the data for activity recognition. In addition to HASC Challenge, we collected indoor pedestrian sensing data of 107 people with a balance of gender and age (HASC-IPSC). Through these data collection, we gathered 111,968 sensor files of 510 subjects. For the convenience of the future researchers in this field, we combined them as a single corpus named HASC-PAC2016 and make it public. Baseline recognition result of HASC-PAC2016 segmented data is 73.4% accuracy for overall, 81.4% for limited by terminal position, and 85.1% with file-based recognition. For sequence data, we only get 73.4% even for limited subjects. This shows we need further research of activity recognition using HASC-PAC2016.

DOI： 10.1145/2968219.2968277

A Location Estimation Method using BLE Tags Distributed Among Participants of a Large-Scale Exhibition Reviewed

Measurement Methods of Spatial Ability using a Virtual Reality System

Shimada Kosuke, Hiroi Kei, Kawaguchi Nobuo, Kaji Katsuhiko

2016 NINTH INTERNATIONAL CONFERENCE ON MOBILE COMPUTING AND UBIQUITOUS NETWORKING (ICMU) page： 80-85 2016

Multi-Algorithm On-Site Evaluation System for PDR Challenge

Kaji Katsuhiko, Kanagu Kohei, Murao Kazuya, Nishio Nobuhiko, Urano Kenta, Iida Hirokazu, Kawaguchi Nobuo

2016 NINTH INTERNATIONAL CONFERENCE ON MOBILE COMPUTING AND UBIQUITOUS NETWORKING (ICMU) page： 74-79 2016

Kenta Urano, Kei Hiroi, Katsuhiko Kaji, Nobuo Kawaguchi

ADJUNCT PROCEEDINGS OF THE 13TH INTERNATIONAL CONFERENCE ON MOBILE AND UBIQUITOUS SYSTEMS: COMPUTING NETWORKING AND SERVICES (MOBIQUITOUS 2016) Vol. 28-November-2016 page： 124 - 129 2016

Language：Japanese Publishing type：Research paper (international conference proceedings) Publisher：ASSOC COMPUTING MACHINERY

Indoor location estimation is essential technology when we analyse the participants' activities in large-scale exhibition. There are some problems with existing methods such as PDR, ultrasound and laser range finder: installation location of measurement equipment at large site, cost for measurement equipment, and necessity of smartphone application. We focus on Bluetooth Low Energy(BLE). Currently, BLE technology is used as proximity notification for smart phones and cannot detect the exact distance between two BLE devices. This is because BLE radiowave is unstable and signal strength changes every time at the same distance. In this paper, we propose a location estimation method which utilizes BLE beacon tags and single board computers. In contrast to conventional ways using location-fixed BLE beacons, BLE beacon tags are distributed to participants at the event. Signal strengths of BLE advertising packets captured by multiple location-fixed scanners are used for location estimation. The method requires less cost for equipment and labor of installing an application. It can be used without complex initial setting. We had a data collection experiment at real large-scale exhibition. We estimate locations of participants by applying the proposed method to the collected data, then evaluate accuracy of estimation and analyse the activities such as the time of longer stay booths based on participants' occupation. As a result, we could track the movement of the participants to some meters error and find the characteristic booths based on longer-stay booths per occupations.

DOI： 10.1145/3004010.3004051

4th workshop on human activity sensing corpus and applications: Towards open-ended context awareness Reviewed

Nobuo Kawaguchi, Nobuhiko Nishio, Daniel Roggen, Sozo Inoue, Susanna Pirttikangas, Kristof Van Laerhoven

UbiComp 2016 Adjunct - Proceedings of the 2016 ACM International Joint Conference on Pervasive and Ubiquitous Computing page： 690 - 695 2016

Language：Japanese Publishing type：Research paper (international conference proceedings) Publisher：Association for Computing Machinery, Inc

Current motion sensors in wearable devices are primarily used for simple orientation and motion sensing. They provide however signals related to more complex and subtle human behaviours which will enable next-generation human-oriented computing in scenarios of high societal value. This requires large scale human activity corpuses and improved methods to recognise activities and their context. This workshop deals with the challenges of designing reproducible experimental setups, running large-scale dataset collection campaigns, designing robust activity and context recognition methods and evaluating systems in the real world. As a special topic, we wish to reflect on the challenges and approaches to recognise activities outside of a pre-defined set to achieve an open-ended activity and context awareness. Following the success of previous years, this workshop is the place to share experiences on human activity corpus and their applications and to discuss the future of activity sensing, in particular towards open-ended contextual intelligence.

DOI： 10.1145/2968219.2968294

A better positioning with BLE tag by rssi compensation through crowd density estimation Reviewed

Kei Hiroi, Yoichi Shinoda, Nobuo Kawaguchi

UbiComp 2016 Adjunct - Proceedings of the 2016 ACM International Joint Conference on Pervasive and Ubiquitous Computing page： 831 - 840 2016

Language：Japanese Publishing type：Research paper (international conference proceedings) Publisher：Association for Computing Machinery, Inc

We present a scheme that improves accuracy of 2.4GHz RF tag based indoor positioning. The accuracy of indoor positioning using 2.4GHz RF tags is affected by propagation loss caused by human body shielding, especially in crowded situations. This paper proposes a RSSI compensation scheme that estimate a crowd density level based on detected 2.4 GHz RSSI. We developed a scanner system to sense BLE (Bluetooth Low Energy) tag for 2.4 GHz RSSI compensation. We deployed 40 BLE scanners and over 100 BLE tags, and collected data for 100 participants at an actual event over a period of four days. Using the collected data, we estimated crowd density level that can be applied to the proposed compensation scheme based on theoretical methods on propagation loss. Our scheme resulted in 59.8% higher accuracy than simple positioning method without the compensation.

DOI： 10.1145/2968219.2968293

安定センシング区間検出に基づく3次元歩行軌跡推定手法 Reviewed

梶克彦, 河口信夫

情報処理学会論文誌 Vol. 57 ( 1 ) page： 12 - 24 2016

Multi-Algorithm On-Site Evaluation System for PDR Challenge Reviewed

Katsuhiko Kaji, Kohei Kanagu, Kazuya Murao, Nobuhiko Nishio, Kenta Urano, Hirokazu Iida, Nobuo Kawaguchi

2016 NINTH INTERNATIONAL CONFERENCE ON MOBILE COMPUTING AND UBIQUITOUS NETWORKING (ICMU) page： 74 - 79 2016

Language：English Publishing type：Research paper (international conference proceedings) Publisher：IEEE

PDR (Pedestrian Dead Reckoning) is a very promising technology for indoor positioning. We held a technical challenge, entitled the UbiComp/ISWC 2015 PDR Challenge, consisting of the following three categories: a PDR algorithm category; a PDR Evaluation method category; and an exhibition. In this paper, we especially focus on several systems for the PDR algorithm category. A PDR skeleton was prepared for the participants. Using an Android skeleton, participants focus on implementing the PDR algorithm because of the skeleton's various functions, such as sensor data acquisition, trajectory visualization, and sensor data upload. The evaluation server evaluates the accuracy of each PDR algorithm automatically as often as sensor data is uploaded to the server and provides a trajectory image file so that participants can compare their PDR algorithms in real time.

DOI： 10.1109/ICMU.2016.7742094

FloodEye: Real-time Flash Flood Prediction System for Urban Complex Water Flow Reviewed

Kei Hiroi, Nobuo Kawaguchi

2016 IEEE SENSORS 2016

Language：English Publishing type：Research paper (international conference proceedings) Publisher：IEEE

This study presents a water-level sensor system that has multiple functions: real-time river-level monitoring and high accurate flood prediction for urban complex water flow (CWF) flooding caused by Localized Heavy Rain. Traditional water-level sensors for large-scale river can detect water-level rising, though they have limitations in detecting urban river dyke height to alert about flooding. This is a result of various riverside environmental limitations. Additionally, previous prediction methods are difficult to detect CWF flooding with high accuracy because water rising differ depends on a location's environment. Therefore, we propose a detection scheme for CWF by developing a water-level sensor system that works in various installation environments using infrared image processing with both low installation and operation cost. We implemented a CWF prediction system that produces accurate and early predictions using a Linear Regression (LR) of deep learning approach with data assimilation. Evaluating based on actual installations (over 1,000 days, 13 locations), our system proved to accurately monitor and predict water-level rising after 5 minutes.

Proceedings of the 7th ACM SIGSPATIAL International Workshop on Indoor Spatial Awareness, ISA 2015: Foreword

Li K.

Proceedings of the 7th ACM SIGSPATIAL International Workshop on Indoor Spatial Awareness, ISA 2015 page： . 2015.11

A proposal of IndoorGML extended data model for pedestrian-oriented voice navigation system Reviewed

Hirokazu Iida, Kei Hiroi, Katsuhiko Kaji, Nobuo Kawaguchi

Proceedings of the 7th ACM SIGSPATIAL International Workshop on Indoor Spatial Awareness, ISA 2015 page： 9 - 14 2015.11

Language：English Publishing type：Research paper (international conference proceedings) Publisher：Association for Computing Machinery, Inc

We propose Landmark-Conscious Voice Navigation as one type of a pedestrian navigation system, which navigate users by only voice guidance. It is necessary to standardize data model in order to use this system widely. In a previous paper[1], we constructed a basic voice navigation system, which uses Open Street Map based data model. In this paper, at first, we conduct an experiment of voice navigation at an underground shopping area of Nagoya Station with two types of landmark descriptions. After that, we discuss what data structure is necessary to describe landmark information for voice navigation. Therefore, we propose to extend IndoorGML1.0 by adding landmark space as a new defined data model for voice navigation. The main contribution of this paper is that we conduct an experiment of voice navigation and research how different landmark descriptions affect users
furthermore, we discuss a IndoorGML extended data model for voice navigation.

DOI： 10.1145/2834812.2834814

Velobug: A mobile system for 3D indoor mapping Reviewed

Weimin Wang, Katsuhiko Kaji, Kenji Yamakawa, Nobuo Kawaguchi, Kei Hiroi

UbiComp and ISWC 2015 - Proceedings of the 2015 ACM International Joint Conference on Pervasive and Ubiquitous Computing and the Proceedings of the 2015 ACM International Symposium on Wearable Computers page： 337 - 340 2015.9

Language：English Publishing type：Research paper (international conference proceedings) Publisher：Association for Computing Machinery, Inc

In this work, we implement a mobile system called Velobug to measures 3D data of the environment. Velobug could generate dense and colored point cloud to reconstruct the environment for 3D mapping with long effective range. Velobug is mainly consisted of a Velodyne HDL-32e LiDAR senor and a Point Grey Research Ladybug3 panoramic camera. The LiDAR sensor is utilized to acquire 3D data of the environment and generate point cloud. Pictures captured by the panoramic camera are used for complementing RGB information of the point cloud after the calibration between the panoramic camera and the LiDAR sensor. We could maker point cloud denser by registering multiple frames of point cloud with Generalized Iterative Closet Point(Generalized-ICP) method. We evaluate Velobug in an underground shopping street near Nagoya Station. The experiment result shows the mobility, practicability and efficiency in reconstruction for 3D indoor mapping.

DOI： 10.1145/2800835.2800878

Design and implementation of algorithm for estimation of elevator travel distance using smartphone accelerometer Reviewed

Antonio Martinez, Kei Hiroi, Nobuo Kawaguchi

Language：English Publishing type：Research paper (international conference proceedings) Publisher：Association for Computing Machinery, Inc

Since the advent of smartphones equipped with sophisticated sensing hardware, human activity recognition research has moved from utilizing dedicated sensing devices to using commercial smartphones. This paper presents the design of an algorithm to recognize and estimate travel distance when riding an elevator and its corresponding implementation within an app for a smartphone running Apple's mobile operating system (iOS). The algorithm receives solely the signal of the smartphone's accelerometer, recognizes that it belongs to an elevator ride, and proceeds to calculate the distance traveled by the rider. The algorithm has been designed in a way that simplifies the necessary operations to calculate the travel distance with the objective of minimizing processing power, while keeping the corresponding estimations highly accurate. This is an initial attempt towards the building of a robust, but simple and fast, real-time human activity recognition service on a wearable platform.

DOI： 10.1145/2800835.2801620