Tam Bang

Tam Bang

Auto-generated profile, based on publicly available information

Tam Bang

Center for Urban Informatics and Progress (CUIP), The University of Tennessee at Chattanooga, Chattanooga, TN

Research Focus & Metrics

Academic Career Length

4Years (2023-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
24 December 2025 in IEEE Transactions on Instrumentation and Measurement
Verified
Journal Article
Closed access
LCLei Cheng
Lei Cheng
LGLihao Guo
Lihao Guo
TZTianya Zhang
Tianya Zhang
TBTam Bang
Tam Bang
AHAustin Harris
Austin Harris
SCSiyang Cao
Siyang Cao
Accurate multi-sensor calibration is essential for deploying robust perception systems in applications such as autonomous driving and intelligent transportation. Existing LiDARcamera calibration methods often rely on manually placed targets, preliminary parameter estimates, or intensive data pre-processing, limiting their scalability and adaptability in real-world settings. In this work, we propose a fully automatic, targetless, and online calibration framework, CalibRefine, which directly processes raw LiDAR point clouds and camera images. Our approach is divided into four stages: (1) a Common Feature Discriminator that leverages relative spatial positions, visual appearance embeddings, and semantic class cues to identify and generate reliable LiDAR-camera correspondences, (2) a coarse homography-based calibration that uses the matched feature correspondences to estimate an initial transformation between the LiDAR and camera frames, serving as the foundation for further refinement, (3) an iterative refinement to incrementally improve alignment as additional data frames become available, and (4) an attention-based refinement that addresses non-planar distortions by leveraging a Vision Transformer and cross-attention mechanisms. Extensive experiments on two urban traffic datasets demonstrate that CalibRefine achieves high-precision calibration with minimal human input, outperforming state-of-the-art target-less methods and matching or surpassing manually tuned base-lines. Our results show that robust object-level feature matching, combined with iterative refinement and self-supervised attention-based refinement, enables reliable sensor alignment in complex real-world conditions without ground-truth matrices or elaborate preprocessing. Code is available at https://github.com/radar-lab/Lidar_Camera_Automatic_Calibration.

19 July 2023 in Electronics
Verified
Journal Article
Open access
PDF file
HTHuy Tran
Huy Tran
TVTien Vu-Van
Tien Vu-Van
TBTam Bang
Tam Bang
TLThanh-Van Le
Thanh-Van Le
HPHoang-Anh Pham
Hoang-Anh Pham
NHNguyen Huynh-Tuong
Nguyen Huynh-Tuong
It is crucial to review and update course materials regularly in higher education. However, in the course evaluation process, it is debatable what a difficult learning topic is. This paper proposes a data mining approach to detect learning topics requiring attention in the improvement process of teaching materials by analyzing the discrepancy between formative and summative assessments. In addition, we propose specific methods involving clustering and noise reduction using the OPTICS algorithm and discrepancy calculation steps. Intensive experiments have been conducted on a dataset collected from accurate assessment results of the data structures and algorithms (DSA) course for IT major students at our university. The experimental results have shown that noise reduction can assist in identifying underperforming and overperforming students. In addition, our proposed method can detect learning topics with a high discrepancy for continuously improving teaching materials, which is essential for question recommendation in adaptive learning systems.


1 January 2026 in IEEE Sensors Letters
Verified
Journal Article
Closed access
LGLihao Guo
Lihao Guo
JTJiahao Tang
Jiahao Tang
TBTam Bang
Tam Bang
TZTianya Zhang
Tianya Zhang
AHAustin Harris
Austin Harris
SCSiyang Cao
Siyang Cao


29 June 2026 in Transportation Research Record: Journal of the Transportation Research Board
Verified
Journal Article
Closed access
MKMehdi Khaleghian
Mehdi Khaleghian
TBTam Bang
Tam Bang
MMMahshid Malazizi
Mahshid Malazizi
MSMina Sartipi
Mina Sartipi
Road diets represent a cost-effective strategy for enhancing roadway safety; however, comprehensive frameworks integrating operational assessment with surrogate safety evaluation remain limited. This study develops an end-to-end microscopic simulation and machine learning framework to quantify operational and safety impacts of an urban road diet. A 0.22 mi segment of Frazier Avenue in Chattanooga, Tennessee, U.S.—recently converted from a four-lane undivided cross-section to a two-lane facility with a center left-turn lane—is modeled in SUMO using multi-month GridSmart trajectory counts, speed measurements, annual average daily traffic, and OpenStreetMap-derived geometry. A one-way analysis-of-variance-based sensitivity analysis identifies nine influential driver and vehicle parameters, which are calibrated via a genetic algorithm that minimizes speed root mean square error; validation against turning-movement counts using the Geoffrey E. Havers statistic satisfies FHWA acceptance criteria. Three 24 h scenarios are simulated (pre-diet, post-diet, and post-diet geometry with pre-diet demand), complemented by a demand sensitivity analysis up to +30% average daily trips. Corridor-wide, the road diet yields modest mobility penalties—average travel time increases by 4.3%, average speed decreases by 5.2%, and average waiting time increases by 40.7%—with more pronounced degradation westbound and intersection-specific queue growth. Surrogate safety measures at the major signalized intersection indicate substantial safety gains: total vehicle conflicts with time-to-collision (TTC) <3 s decrease by 45.7%, and critical conflicts with TTC <1 s decrease by 62.3%, with conflict hotspots markedly attenuated. Finally, an XGBoost regressor using pre-diet queue length, density, travel time, and speed predicts post-diet waiting times with R 2 = 0.905, demonstrating the potential of simulation-driven predictive models to support proactive corridor management and design of future road diet projects.