Journal of Science and Transport Technology

Violation Detection on Traffic Light Area Based on Image Classification Using Dimensionality Reduction and Deep Learning

2024-07-12T09:22:59+00:00

The smart city concept is closely related to efficient traffic management, especially using technology to improve safety and smooth traffic flow, including at traffic lights. In this context, traffic light integration into a smart city system uses sensors, surveillance cameras, and intelligent algorithms to adaptively manage traffic based on vehicle volume and real-time traffic conditions. The CCTV already installed in several junction road at Bandung City, Indonesia. Image classification using deep learning is an essential and rapidly growing application in artificial intelligence. When the number of images in the dataset collected from CCTV gets larger, the total dimension of the data will also increase significantly. The large dimension of image datasets makes the data analysis and processing process more complex and requires extensive computational resources. To reduce computational resource, dimensionality reduction using Principal Component Analysis (PCA) can handle high-dimensional data. PCA is used to process and analyze image datasets efficiently. We proposed combination of deep learning and PCA to solve classification problem to high dimensional traffic image dataset. Experimental result showed that the non-PCA deep learning model achieved an accuracy of 73.11%, while the deep learning model with PCA achieved an accuracy of 72.73%. In other hand, the combination of deep learning and PCA showed a much shorter training time of only 2.95 seconds compared to the non-PCA deep learning model, which took 80.43 seconds.

Engineering Evaluation of Freshwater Lake Coastal Sand Dunes: A Study of Sandbanks Provincial Park, Kingston, Canada

2024-10-21T03:01:35+00:00

This study provides a comprehensive engineering evaluation of the freshwater coastal sand dunes at Sandbanks Provincial Park, located on the southern edge of Prince Edward County, Kingston, Ontario, Canada. Sandbanks is renowned for its extensive and dynamic sand dune system, which plays a crucial role in coastal management by absorbing wave energy, mitigating storm surges, and preventing sand deposition on infrastructure. Shaped by historical glacial activities, wind, and wave action, these dunes are critical for stabilizing the beach environment, controlling wind erosion, and supporting diverse plant and animal communities. This research uniquely combines historical data with recent field studies to offer new insights into the engineering properties, geomorphological processes, and ecological dynamics that govern the formation, stability, and resilience of the dunes. The study addresses contemporary challenges such as climate change, human impact, and erosion, and proposes actionable conservation strategies that balance ecological preservation with practical land management. While focused on a specific section of Ontario’s coastal dunes, the findings contribute to a deeper understanding of coastal dune systems more broadly, offering valuable guidance for their long-term sustainability and management across similar environments.

Comparative Study on the Probabilistic Safety of Truss Structures Designed Using U.S. and Vietnamese Codes

2024-12-03T02:43:17+00:00

This study evaluates and compares the reliability of a truss designed according to the Vietnamese TCVN 5575-2012 and the American AISC 360-16 steel design codes. The same truss configuration and applied loads are considered, with sections designed according to each standard. Their probabilistic safety levels are then evaluated to provide deeper insights into the differences between the design codes. Results indicate that, under identical loading, trusses designed with AISC are lighter than those designed with TCVN, as TCVN requires larger sections for compression members. Reliability indexes (RIs) for tension behavior are similar between codes; however, TCVN yields higher RIs for buckling, indicating a conservative approach for compression members compared to AISC. Although TCVN does not specify a target RI, its deterministic and probabilistic safety levels exceed those of AISC, suggesting a target RI above AISC’s 3.0. Consequently, TCVN-based designs generally involve higher costs, emphasizing the importance of understanding safety implications in code selection. Finally, the conservative results from TCVN are examined through equivalent safety factors, providing insights into its design assumptions.