Traffic Sign Dataset for Rural Areas 

Connected and Automated Vehicles (CAVs) rely on accurate traffic sign recognition to ensure safe and efficient navigation. Despite advances in artificial intelligent tools, deep learning, and computer vision, the effectiveness of current traffic sign recognition models remains highly dependent on the comprehensiveness of training datasets. Existing traffic sign datasets, such as the LISA Traffic Sign Dataset, primarily include traffic signs commonly found in urban settings, leaving a gap in the representation of signs prevalent in rural areas. This limitation challenges the deployment of AVs in rural environments, where unique traffic signs and varying road conditions require specialized recognition capabilities. This dataset developed by CR2C2 researchers augments the LISA Traffic Sign Dataset with traffic signs commonly encountered in rural environments, such as deer crossing, cattle crossing, and farm machinery signs. This process involves both real-world data collection and synthetic data generation to enhance the robustness of the model against diverse environmental conditions, including occlusions, weather variations, and faded signage. This dataset was used for enhancing CR2C2 AVs performance in rural areas. Experimental results demonstrate that incorporating these rural-specific signs significantly improves detection performance in rural scenarios, providing a more comprehensive and robust dataset for future CAV applications.

 Link to Dataset: https://zenodo.org/records/15856539

Related publications:

o   Mohammed Zakaria, Tesfamichael Getahun, Ali Karimoddini, “Traffic Sign Dataset for Connected and Automated Vehicle Operations in Rural Areas,” Data in Brief, Under Review.

o   Mohammed Zakaria, Tesfamichael Getahun, Ali Karimoddini, “Traffic Sign Detection for Rural Operations of Connected and Automated Vehicles,” 2025 2nd International Conference on Intelligent Driving and Smart Transportation, Under review.

Cite Dataset: Zakaria, M., Getahun, T., & Karimoddini, A. (2025). Augmented Lisa Traffic Sign Dataset for Rural Areas (1.0) [Data set]. Center for Regional and Rural Connected Communities (CR2C2). DOI: 10.5281/zenodo.15856539