Video-based road intelligent detection constitutes a critical component in modern intelligent transportation systems, serving as a crucial role for comprehensive transportation planning and emergency traffic management. Current traffic scene perception methodologies relying on conventional deep learning architectures present inherent limitations, including heavy dependence on extensive manual annotations of specific traffic scenarios and predefined rule configurations. These approaches demonstrate constrained semantic representation capacity and limited generalizability across heterogeneous traffic scenarios. To address these challenges, this study proposes a novel end-to-end multimodal foundation model architecture that jointly generates dynamic traffic event detection outcomes and semantic-rich contextual descriptions. Through integration of low-rank adaptation (LoRA) and prompt fine-tuning as parameter-efficient fine-tuning strategies, we develop the multimodal road traffic scene understanding foundation model (MoTIF), which establishes cross-modal alignment between visual patterns and textual semantics. This framework demonstrates enhanced capability in extracting salient traffic targets and generating hierarchical scene representations, significantly improving automated detection efficiency in road video analytics. Notably, MoTIF exhibits contextual reasoning capabilities for implicit traffic event interpretation. Extensive evaluations on two real-world datasets encompassing urban road intersection scenarios in Tianjin and highway monitoring systems in Shandong Province reveal that MoTIF achieves superior performance metrics: 65.81 average score on multimodal scene understanding assessment and 83.33% event detection accuracy, outperforming mainstream benchmarks in both precision and computational efficiency. This research advances multimodal learning paradigms for intelligent transportation systems while providing practical insights for adaptive traffic management applications.

Ride-sharing is one of the typical models in the context of the sharing economy. With the advantages of reducing travel costs and carbon emissions, it has occupied an important share in the urban travel market. Vehicle resource allocation is the core link in optimizing carpooling services. To realize the overall management of travel modes, this paper comprehensively considered the influence of multiple travel modes and passengers’selection behavior on the allocation scheme. Focusing on the supply and demand matching of the mixed travel mode based on ride-sharing travel, the internal correlation between the passenger travel mode selection behavior and the allocation of vehicle resources between regions was analyzed. A passenger-to-driver matching function was introduced to describe the relationship between vehicle supply and passenger demands. Then the passengers’travel selection behavior was simulated by the Logit model, and an inter-regional resource allocation optimization model was constructed. In solving the optimization model, the allocation problem was divided into two stages: passenger travel mode allocation and vehicle supply allocation. The Lagrangian relaxation and the gradient descent algorithms were integrated to solve the decision-making problems involved in each stage. The solutions of the two stages were updated based on the feedbacks from passenger selection preference and vehicle allocation, so as to establish the regional multi-period traffic resource allocation algorithm. Finally, the resource allocation algorithm was tested by random generation of examples on a road network with 100 sub-regions. Results show that the method proposed in this paper can effectively balance the mobility allocation among regions with mixed travel modes. As the number of ride-sharing vehicles decreased, the decreasing proportion of total revenue increased from 7.32% to 25.37%. Compared with the vehicle allocation method based on mileage, the vehicle allocation algorithm proposed in this paper helps to improve the total revenue.