Department Publications

Vehicle license plate recognition is an integral component of intelligent transportation systems. In this work, we present an embedded real-time license plate recognition system customized for developing countries. We address the challenge of handling complex, unstructured traffic scenes with diverse vehicle types while implementing the system on an embedded platform for low-cost deployment. Our method consists of license plate detection on a multi-vehicle image, followed by character recognition on the detected license plates. Both steps use lightweight convolutional neural networks to balance accuracy and efficiency. We also introduce the SL-LPR dataset of Sri Lankan road images, which contains a variety of vehicle types and traffic conditions typically seen in developing countries. On this dataset, the license plate detection and character recognition models achieved 93.6% mAP and 87.88% accuracy, respectively, and were competitive against larger models on several public datasets. To achieve real-time performance in a resource-constrained embedded environment, we applied low-bitwidth quantization using the Brevitas library and implemented FPGA acceleration for the models using the FINN framework. The end-to-end system can operate at 11.5~FPS when implemented on the Xilinx Kria KV260 platform. These results demonstrate that our system is effective for real-time license plate recognition on an embedded device, even in complex traffic scenarios. The SL-LPR dataset is available for research use at: https://github.com/sl-lpr-uom/SL-LPR.git.

Traffic signal design in developing countries frequently underperforms due to the direct application of classical models—Webster's delay formula, HCM, ARR123, and ORN13—developed for homogeneous, lane-disciplined traffic conditions. Sri Lankan intersections exhibit fundamentally different characteristics: heterogeneous vehicle streams comprising motorcycles, three-wheelers, non-motorized vehicles, and pedestrians operating without strict lane discipline under geometric constraints. This systematic review synthesizes empirical evidence explaining why conventional signal design methods fail in such contexts. Following PRISMA principles, 138 records were screened, yielding 41 high-quality references spanning foundational theory, classical assumptions, regional adaptations, and Sri Lankan empirical data. The review demonstrates that local intersections systematically violate foundational assumptions regarding lane usage, headway uniformity, start-up behavior, and pedestrian compliance. Observed saturation flows reach 2574 pcu/h/lane—substantially exceeding HCM benchmarks—while behaviors such as motorcycle seepage, lateral filtering, initial surge, and dynamic clustering create discharge patterns that static PCU values cannot capture. A mathematical framework is presented mapping these deviations to quantifiable variables: lateral efficiency, surge discharge dynamics, dynamic vehicle equivalence, headway expansion, and geometric constraints affecting intersection capacity. The paper concludes by proposing context-sensitive design pathways incorporating dynamic PCU factors, behavioral modeling, and simulation …