Selected Projects
- Predicting Barge Presence and Quantity on Inland Waterways using Vessel Tracking Data using Machine Learning
- The study introduces a machine learning approach to predict the number of barges transported by vessels on inland waterways using Automatic Identification System (AIS) tracking data, which does not inherently track barge presence or quantity.
- Using features derived from AIS data and manual observations, the study developed models with high F1 scores for predicting both the presence and quantity of barges, offering valuable insights for waterway management and infrastructure planning.
- Real-Time Barge Detection Using Traffic Cameras and Deep Learning on Inland Waterways
- Developed a deep learning model designed to analyze traffic camera feeds and accurately predict the presence of barge traffic on inland waterways, achieving an accuracy rate of 96% on 116 test images.
- Authored a research paper detailing the project's methodology, findings, and implications, currently under review for publication in the Transportation Research Record (TRR).
- Improving Rider Satisfaction and Efficiency of Razorback Transit in Fayetteville, Arkansas
- Designed comprehensive questionnaires aimed at capturing rider feedback and preferences, focusing on enhancing overall satisfaction with Razorback Transit services in Fayetteville.
- Implemented a beta survey to test the effectiveness and usability of the questionnaires among a sample of riders.
- Investigating the Relationship Between Particulate Matter Pollution, Land Use, and Population Growth in Long Beach, California Using GIS
- Leveraged ArcGIS spatial analysis and modeling techniques to create a comprehensive model predicting the relationship between particulate matter (PM) pollution, land use patterns, and population growth in Long Beach, California.
- Visualized the analysis results using GIS tools, generating 12 maps and reports to effectively communicate complex spatial relationships.
- Transportation Engineering Game for K-12 Students
- Conducted extensive research to gain insights into key concepts and principles of Transportation Engineering most relevant and engaging for K-12 students.
- Created and tested an interactive game challenging about 20 students to solve Transportation Engineering problems, resulting in a positive feedback score of 4.75 and effectively promoting problem-solving skills.
- 2023 AI City Challenge - Track 5 (Detecting Violation of Helmet Rule for Motorcyclists)
- Employed state-of-the-art computer vision techniques, including Convolutional Neural Networks (CNNs) and object detection algorithms, to analyze video feeds from traffic cameras and identify helmet rule violations among motorcyclists, achieving an mAP score of 0.526.
- Authored a research paper detailing the project's methodology, results, and implications, which has been published, contributing to the field of computer vision and traffic safety.