Vehicle-Level Temporal Image Sequences is a dataset of vehicle image sequences designed to provide standardized data for research on dangerous driving behavior recognition (sudden braking provocation, forced entry, emergency braking, serpentine driving) and vehicle-level video temporal classification. It is mainly used to support research in areas such as intelligent transportation systems, advanced driver assistance systems, and class imbalance learning. This dataset contains 434 vehicle-level time-series image sequences, covering normal driving and dangerous driving behaviors. It is organized according to 94 internally labeled source video groups, with each sequence corresponding to a tracked vehicle and accompanied by a behavior label metadata file. In the original five-class classification, there were 351 sequences of normal driving, 53 of sudden braking and provocation, 21 of forced entry, 5 of sudden braking, and 4 of zigzag driving. In the experiment, forced entry, sudden braking, and zigzag driving were merged into other categories.

Citation

Galvín-Terrero, M., Guerrero-Contreras, G. & Balderas-Díaz, S. (2026). Vehicle-Level Temporal Image Sequences for Driving Behavior Recognition (Version 1.0) [Dataset]. Zenodo. https://doi.org/10.5281/zenodo.21427132