Search Results (3)

Accurate vehicle counts from traffic videos are fundamental to traffic measurement and to estimating roadway demand for infrastructure planning and maintenance. However, many vision-based traffic datasets and pretrained models under-represent vehicle types that are prevalent in developing countries, such as the keke (globally known as auto-rickshaw/three-wheeler), which can bias traffic composition estimates and downstream workload indicators. This paper presents a keke-aware vehicle detection and counting pipeline that combines fine-tuned YOLO-based detectors with BoT-SORT/ByteTrack tracking and ROI-based counting, together with a newly curated and publicly released traffic-video dataset that includes a dedicated keke class. The detectors are fine-tuned from pretrained weights on a six-class dataset (bicycle, bus, car, motorcycle, truck, keke) and evaluated on held-out roadside test videos with a manual counting baseline. On the validation split (2088 images; 8400 instances), the fine-tuned YOLO11l model achieves $P=0.752$, $R=0.696$, $\mathrm{mAP}@0.5=0.766$, and $\mathrm{mAP}@0.5\text{:}0.95=0.578$, with the keke class attaining $\mathrm{mAP}@0.5=0.772$, while YOLO26l achieves slightly higher overall precision ($P=0.766$) and stronger keke recall and $\mathrm{mAP}@0.5\text{:}0.95$. In system-level counting, the selected tuned ROI-based variants produce the most reliable results on the Yola Road downward flow, where keke counts remain close to the manual baseline, but performance is strongly direction- and scene-dependent, with substantially larger errors in the Yola upward flow and the more challenging Mubi Road scene. Flow-rate and ESAL-rate analyses further show that class misclassification can severely distort pavement-loading estimates even when total traffic flow appears close to baseline, underscoring the need for localized class ontologies and robust heavy-vehicle discrimination in mixed-traffic ITS deployments. The released dataset and baseline pipeline provide a practical reference for keke-aware traffic monitoring and for infrastructure-relevant traffic measurement in developing-country contexts. Full article

The engineering sector is undergoing digital transformation (DT) alongside shifts in labor patterns. This study concentrates on piping design within plant engineering, aiming to develop a system for optimal piping route design using artificial intelligence (AI) technology. The objective is to overcome limitations related to time and costs in traditional manual piping design processes. The ultimate aim is to contribute to the digitalization of engineering processes and improve project performance. Initially, digital image processing was utilized to digitize piping and instrument diagram (P&ID) data and establish a line topology set (LTS). Subsequently, three-dimensional (3D) modeling digital tools were employed to create a user-friendly system environment that visually represents piping information. Dijkstra’s algorithm was implemented to determine the optimal piping route, considering various priorities during the design process. Finally, an interference avoidance algorithm was used to prevent clashes among piping, equipment, and structures. Hence, an auto-routing system (ARS), equipped with a logical algorithm and 3D environment for optimal piping design, was developed. To evaluate the effectiveness of the proposed model, a comparison was made between the bill of materials (BoM) from Company D’s chemical plant project and the BoM extracted from the ARS. The performance evaluation revealed that the accuracy in matching pipe weight and length was 105.7% and 84.9%, respectively. Additionally, the accuracy in matching the weight and quantity of fittings was found to be 99.7% and 83.9%, respectively. These findings indicate that current digitalized design technology does not ensure 100% accurate designs. Nevertheless, the results can still serve as a valuable reference for attaining optimal piping design. This study’s outcomes are anticipated to enhance work efficiency through DT in the engineering piping design sector and contribute to the sustainable growth of companies. Full article

In this paper, we present the need for specialized artificial intelligence (AI) for counterfeit and defect detection of PCB components. Popular computer vision object detection techniques are not sufficient for such dense, low inter-class/high intra-class variation, and limited-data hardware assurance scenarios in which accuracy is paramount. Hence, we explored the limitations of existing object detection methodologies, such as region based convolutional neural networks (RCNNs) and single shot detectors (SSDs), and compared them with our proposed method, the electronic component localization and detection network (ECLAD-Net). The results indicate that, of the compared methods, ECLAD-Net demonstrated the highest performance, with a precision of 87.2% and a recall of 98.9%. Though ECLAD-Net demonstrated decent performance, there is still much progress and collaboration needed from the hardware assurance, computer vision, and deep learning communities for automated, accurate, and scalable PCB assurance. Full article