@conference {123, title = {Applications of Earth Observation Data from Small-scale Satellites for Disaster Management by Combinations with Open Geospatial Data}, booktitle = {3rd Human Resource Development and Space Data Utilization for Disaster}, year = {2020}, month = {01/2020}, address = {Bali, Indonesia}, author = {Hiroyuki Miyazaki} } @booklet {128, title = {Vec2Instance: Parameterization for Deep Instance Segmentation}, year = {2020}, author = {N. Lakmal Deshapriya and Matthew N. Dailey and Manzul Kumar Hazarika and Hiroyuki Miyazaki} } @conference {122, title = {Development of an Automated Settlement Mapping System using High-Resolution Satellite Images and Deep Learning}, booktitle = {The 60th Annual Meeting for the Japanese Society of Tropical Medicine}, year = {2019}, month = {11/2019}, address = {Okinawa, Japan}, abstract = {We developed an automated system of building with the use of high-resolution satellite images and deep learning, comprising a geospatial data management system, an image data processing system, and a quality control system. The system development has achieved the component of the geospatial data management and image data processing, and performed building mapping of some large extents, while the development of quality control systems is ongoing. Because we developed the system with open-source and web-based software, anyone can participate in the preparation of training data just only with a computer and the Internet. The system is expected to be a platform for the large-scale mapping of buildings and other ground objects with international collaborations of local partners. The building maps developed by this system are expected to be a basis of analyzing demography with possible risks and impacts of communicable diseases.}, author = {Hiroyuki Miyazaki and Wataru Ohira and Satoshi Kaneko and Ryosuke Shibasaki} } @proceedings {115, title = {Development of micro population data for each building: Case study in Tokyo and Bangkok}, year = {2019}, address = {Chiang Mai, Thailand}, abstract = {In order to carry out sustainable development and management of cities, it is necessary to design and implement appropriate city planning and traffic planning. Indispensable information for designing them is the population distribution. However, population data with high spatial resolution, such as building units, are rarely maintained in cities in developing countries. Therefore, this study examined the development of methods for estimating the number of residents per building in Tokyo and Bangkok using detailed building maps and population census in subdistrict units. In addition, using these methods, we tried to develop micro population data (MPD) across Tokyo and Bangkok. Moreover, the reliability of MPD was verified by comparing it with population census with higher resolution than subdistrict unit in Tokyo. As a result, it has become possible to develop MPDs that are strongly correlated with the population census of various aggregation units and have small errors.}, keywords = {building, census, disaggregation, micro geodata, population}, url = {https://ieeexplore.ieee.org/document/9018851}, author = {Yuki Akiyama and Hiroyuki Miyazaki and Sirinya Sirikanjanaanan} } @proceedings {117, title = {Effect of Public Transport Network on Urban Core and the Future Perspective in Bangkok, Thailand}, year = {2019}, month = {12/2019}, address = {Chiang Mai, Thailand}, abstract = { City center is an essential components of urban structure that rules urban activities including economy, transport, and social interactions. In Bangkok, Thailand, the railway network is expanding and the expansion is expected to affect the city center locations. In this study we attempt to capture the effect of public transport network on the accumulation of three types of urban core facilities based on the spatial statistical approach, and estimate the future perspective of locations of those facilities. As a result we found that expected number of facilities in current urban core in Bangkok decreases and the number of facilities at stations on planned railways increases under certain conditions. The results can be utilized to estimate the future travel pattern and residential locations. }, keywords = {location probability, network centrality, point of interest, railway network, urban core}, url = {https://ieeexplore.ieee.org/document/9018769}, author = {Masanobu Kii and Apantri Peungnumsai and Varameth Vichiensan and Hiroyuki Miyazaki} } @proceedings {118, title = {Impact of Rainfall on Urban Traffic Flow based on Probe Vehicle Data in Bangkok}, year = {2019}, month = {12/2019}, address = {Chiang Mai, Thailand}, abstract = {Adverse weather frequently affects the capacities and travel speeds on roadways, which result in worsened traffic congestion and incurred productivity loss. Further, with climate change predicted to increase rainfall in various cities in Southeast Asia, the risk of flood damage in this region is not only anticipated to increase and affect urban function but may also significantly aggravate daily traffic flow. This study highlighted an analysis of the effect of rainfall on urban traffic flow through the use of probe vehicle data and rainfall data in the center of Bangkok, which is known in Southeast Asia for problems with respect to maintenance of pumps and drainage channels and for many flooded roads after heavy rainfalls. The experimental results demonstrated that the average travel speed decreased by 0.02 km/hour per 1 mm of daily rainfall. In particular, at the time of peak traffic demand, the travel speed was notably reduced when passengers preferred automobile traffic. In 2018, the economic loss estimate in central Bangkok due to annual rainfall was approximately 0.01\% of the city{\textquoteright}s GDP. Future rainfall forecast data makes it possible to assess the risk of climate change on urban traffic flow.}, keywords = {Climate change, Probe vehicle data, Rainfall impact, Regression model, Travel speed}, url = {https://saki.siit.tu.ac.th/stud2019/uploads_final/111__18076cee1637baa6dafa754962eb2939/FinalFile_stud19_takano_v7_en.pdf}, author = {Tsuyoshi Takano and Hiroyoshi Morita and Shinichiro Nakamura and Hiroyuki Miyazaki and Wasan Pattara-atikom and Napaporn Piamsa-nga} } @inbook {106, title = {Measurement of Inter- and Intra-city Connectivity Using Vehicle Probe Data}, booktitle = {Measuring Connectivity Within and Among Cities in ASEAN}, number = {26}, year = {2019}, publisher = {JETRO Bangkok/IDE-JETRO}, organization = {JETRO Bangkok/IDE-JETRO}, chapter = {2}, address = {Bangkok}, abstract = {This chapter analyzes intra- and inter-city connectivity using the vehicle probe data for selected 48-hour slots in March and September in 2017 and 2018. I demonstrate the potential analyses by aggregating the probe data of commercial vehicles with overlay to geographical extents of the majo r cit ies ident ified by night-time light satellite image data.The cit ies could be classified into more vehicles in the daytime or night time, which were likely associated with drivers{\textquoteright} preference on traffic conditions by the time. Some cities indicated notable changes of driving speeds by the time, possibly owing to traffic condition with people{\textquoteright}s commuting as well as transport infrastructure, such as highways. More than half of the vehicles were traveling only two cities within the 48-hour periods, which were possibly shuttle trips between two cities. Some cities in the large industrial areas and inland cities indicated high proportion of vehicles were travelling more than two cities, indicating contribution to connectivity among the city.}, keywords = {Connectivity, Probe data, Urban}, url = {https://www.ide.go.jp/English/Publish/Download/Brc/26.html}, author = {Hiroyuki Miyazaki} } @proceedings {114, title = {A Review of MATSim: A Pilot Study of Chatuchak, Bangkok}, year = {2019}, month = {12/2019}, address = {Chiang Mai, Thailand}, abstract = {Transportation is one of the basic infrastructures that has become an important factor for urban planning and development. In order to develop a better transportation system can lead to better infrastructure, studying the traffic system, current situation and its behavior, is necessary. However, to reveal every object and its dynamic that happens in the traffic system is impossible without a tool and techniques. MATSim is a simulation model software used to assign the traffic between origins and destinations. Most of MATSim applications have been used for developed countries. Nevertheless, Bangkok is one of several cities challenging on the over-saturated situation on road traffic. To check the situation, the simulation can be used to explore highly concentrated traffic flow. Thus, the objective of this study is to examine the applicability of the Multi-Agents Transportation Simulation (MATSim) framework to Bangkok situation. For the travel demand forecasting, it commonly referred to as the four step model. And MATSim framework is one model for the fourth step of the model which is traffic assignment or route assignment. Therefore, this study explored MATSim by experimenting with two plans of agents represented by people travelling from home to work and work to home over Chatuchak district, Bangkok. The sample size of agents using in the simulation are 10, 100, and 500 agents. The results show the traffic flow differently because of the volume of agent effect on the traffic flow.}, keywords = {Agent-based modeling, Road transportation, Transportation}, url = {https://jiist.aiat.or.th/assets/uploads/1588686091433SRGAmjiist.aiat.or.th-23.pdf}, author = {Apantri Peungnumsai and Hiroyuki Miyazaki and Apichon Witayangkurn and Masanobu Kii} } @proceedings {116, title = {Urban Growth Modeling using Historical Landsat Satellite Data Archive on Google Earth Engine}, year = {2019}, month = {12/2019}, address = {Chiang Mai, Thailand}, abstract = {This paper presents a pilot of data analysis for urban growth modeling using historical Landsat satellite data archive on Google Earth Engine and SLEUTH cellular automata model. The systems were organized for non-expert so that it could be useful for other applications. The developed system was applied to urban growth modeling for the cities of Hue, Ha Giang, and Vinh Yen in Viet Nam. Although the results indicated that further tuning will be needed in applying SLEUTH for urban growth modeling, the system was well established enabling users to efficiently polish the quality of the modeling results.}, url = {https://ieeexplore.ieee.org/document/9018846}, author = {Hiroyuki Miyazaki and Himanshu Bhushan and Kotone Wakiya} } @proceedings {113, title = {Utilizing User Generated Contents to describe Tourism Areas of Interest}, year = {2019}, month = {12/2019}, address = {Chiang Mai, Thailand}, abstract = {The use of available place databases (like GeoNamesand traditional maps) to obtain descriptive keywords of a user defined place is not possible because such data sources mainlymaintain location definitions of the well-known places only.Traditional sources may not be updated dynamically and maynot ensure diverse information. Additionally, they do not give anyinformation on the popularity, e.g., which is more popular amongthe places indexed by the same keyword. A bottom-up approach,based on real user attention, can address these problems. Wepropose a method to describe tourism area of interest (TAOI) byaggregating user generated social media text. We match the cooccurrence of important keywords in a particular location andselect such words to describe TAOIs. We applied the proposedmethod to data on micro blogging service Twitter and photosharing service Flickr and confirmed that our method made itpossible to extract TAOI description. The recommended bottomup approach enables the extraction of valuable information thatis not possible by using traditional top-down approaches.}, keywords = {Flickr, TFIDF, Tourism Area of Interest, Twitter, User Generated Contents}, url = {https://ieeexplore.ieee.org/document/9018810}, author = {Bidur Devkota and Hiroyuki Miyazaki and Niraj Pahari} } @proceedings {100, title = {Deep Domain Adaptation for Single-Shot Vehicle Detector in Satellite Images}, year = {2018}, doi = {10.1109/IGARSS.2018.8519129}, url = {https://ieeexplore.ieee.org/abstract/document/8519129}, author = {Yohei Koga and Hiroyuki Miyazaki and Ryosuke Shibasaki} } @proceedings {99, title = {Development of Population Distribution Map and Automated Human Settlement Map Using High Resolution Remote Sensing Images}, year = {2018}, doi = {10.1109/IGARSS.2018.8517827}, url = {https://ieeexplore.ieee.org/abstract/document/8517827}, author = {Uttam Dwivedi and Zhiling Guo and Hiroyuki Miyazaki and Mohamed Batran and Ryosuke Shibasaki} }