Release candidate · August 2026
TH-LDP: designing low-distortion coordinates
Phisan Santitamnont’s RC-1 proposal brings together projection definitions, Python code and regional evaluation reports.
Explore this work ↗Civil engineering, geospatial science and AI. Explore the people and research behind infrastructure that serves society.
Center of Excellence in Infrastructure Management
Scientific Research & Methods
Explore the questions, methods and published work across physical infrastructure and AI.
A road-asset detector combines a pretrained Vision Transformer with YOLOX and a feature-pyramid decoder. The study evaluates detection on Thai highway imagery, including small roadside objects such as kilometer stones.
This study transforms discrete crash features into grayscale images using Weight of Evidence, then trains a VGG-based neural network to classify accident severity. It uses Thai highway crash records from 2011–2023.
Researchers applied GeoSAM to extract building footprints from UAV true orthophotos with a 5 cm ground sampling distance. The study covers science and engineering areas at Chulalongkorn University.
A convolutional network predicts text, offset and border masks to locate curved and closely spaced text. Polygon offsetting and border augmentation help separate neighboring text instances in natural images.
Understanding road risk, travel behaviour and network performance through safety assessment, traffic information and transport research.
Connecting pavement condition, material performance and life-cycle costs to maintenance planning and infrastructure decisions.
Developing knowledge and tools around geographic information systems, surveying and infrastructure information management.
Studying how information supports decisions during disasters, alongside knowledge exchange on preparedness and response.
Initiative · AI for Physical Sciences
Integrating computer vision, foundation models, and statistical learning into large-scale civil engineering systems. Our initiatives focus on automated satellite segmentation, real-time roadside asset intelligence, probabilistic highway safety modeling, and low-distortion geodetic coordinate transformations — backed by open code and reproducible benchmarks.
Faculty & Fellows
Explore expertise and publications across our researcher directory.
Pavement materials and performance
View portfolio & papers →
Geotechnical engineering
View portfolio & papers →
Image analysis and intelligent transport
View portfolio & papers →
Environmental engineering
View portfolio & papers →
Structural engineering
View portfolio & papers →
AI for road safety and sustainable mobility
View portfolio & papers →
Geospatial engineering and infrastructure mapping
View portfolio & papers →
Infrastructure evaluation and construction management
View portfolio & papers →
Computer vision and scene understanding
View portfolio & papers →
Traffic operations and transit planning
View portfolio & papers →
GNSS positioning and cadastral surveying
View portfolio & papers →Projects & impact
Release candidate · August 2026
Phisan Santitamnont’s RC-1 proposal brings together projection definitions, Python code and regional evaluation reports.
Explore this work ↗Programme outcomes · January report
iRAP reports 500 km of Bangkok roads assessed and 360 high-risk sites upgraded through ThaiRAP’s work with BMA. Kasem Choocharukul is identified as ThaiRAP lead.
Explore this work ↗Industry prototype · July presentation
Boonchai Sangpetngam and Tipco Asphalt presented a prototype combining reclaimed asphalt concrete with a waste-palm-oil rejuvenator.
Explore this work ↗Peer-Reviewed Research
A source-linked selection spanning infrastructure and computing.
Engineering Journal
Compares four ways to interpret down-hole seismic measurements at three Thai sites, examining how overburden pressure affects inferred soil velocities.
IEEE Access, 13, pp. 51645–51659
Studies conformal prediction for quantifying uncertainty in HER2 status classification from breast-cancer immunohistochemistry images.
Results in Engineering, 27, 106155
A VGG-based convolutional model predicts crash severity from Thai highway records for 2011–2023. Categorical features become grayscale images, with class balancing and Bayesian optimisation; evaluation uses F1-score and AUC.
Software & tools
Python + external browser tool
Projection-design examples from Phisan Santitamnont, with an author-linked converter for multiple coordinate systems.
Explore this work ↗Public Python code
Tools for traverse adjustment, setting-out, geodetic networks and route geometry, with documented examples.
Explore this work ↗Paper implementation · MIT
Feature extraction, classifier training and analysis notebooks for HER2 classification in a medical-imaging collaboration coauthored by Thanarat Chalidabhongse.
Explore this work ↗Teaching & resources
Recorded workshop · UCL / Chula
UCL’s workshop page provides the programme and recording of the June 2023 event, including a presentation by Kasem Choocharukul.
Explore this work ↗Reference material + code
Phisan’s collection includes digital-camera material and a geotag plotting example for UAV map production.
Explore this work ↗News & Knowledge Exchange
Activities and knowledge exchange, with their original dates.
Partners & collaboration
Joint activities and knowledge exchange with industry and government, documented in the center’s archive.
Industry ·
Co-organized an asset and infrastructure management seminar with the research unit for local-government personnel.
View activity record ↗Government ·
Kasem Choocharukul delivered iRAP Star Rating training for department personnel on 9–10 July 2020.
View activity record ↗Government ·
Boonchai Sangpetngam contributed to a knowledge-transfer seminar on integrating rural-road network data using iAmR.
View activity record ↗Years refer to the activities described. These records do not confirm current partnership or sponsorship status.
Study & collaborate
Explore graduate study, find a researcher working in your field, and connect through the university’s official channels.
Connect with the center about research collaboration, academic services and knowledge exchange.
Study & research enquiries