Claim Missing Document
Check
Articles

Found 3 Documents
Search

YOLOv11-Based Detection of Indonesian Traffic Signs: Transfer Learning vs. From-Scratch Training Ramadhan, Ibnu Cipta; Hendriawan, Akhmad; Oktavianto, Hary
Journal of Applied Informatics and Computing Vol. 9 No. 4 (2025): August 2025
Publisher : Politeknik Negeri Batam

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30871/jaic.v9i4.9718

Abstract

Traffic sign detection is a fundamental component in intelligent transportation systems (ITS), autonomous driving, and advanced driver assistance systems (ADAS), enabling vehicles to interpret road conditions and enhance safety. Developing robust traffic sign detection models for specific regions requires high-quality, well-annotated local datasets, which are often challenging and costly to create. Even when such datasets are available, training deep learning models from scratch demands substantial computational resources and time. This study compares models trained from scratch and those using transfer learning based on the lightweight YOLOv11s architecture on an Indonesian traffic sign dataset. Evaluations using precision, recall, mean Average Precision at IoU 0.5 (mAP@0.5), and mean Average Precision across IoU thresholds 0.5 to 0.95 (mAP@0.5:0.95) demonstrate that the transfer learning model consistently outperforms the from-scratch model across all metrics. These findings highlight the effectiveness and efficiency of transfer learning for developing accurate and practical traffic sign detection systems adapted to local contexts.
A Low-Cost Salinity Meter Based On Ultrasonic Wave Gunawan, Agus Indra; Hendriawan, Akhmad; Taufiqurrahman, Taufiqurrahman; Nurmaida, Firnanda Pristiana
Jurnal Rekayasa Elektrika Vol 21, No 3 (2025)
Publisher : Universitas Syiah Kuala

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.17529/jre.v21i3.43940

Abstract

Monitoring the quality of shrimp pond water is crucial for shrimp growth, with salinity being one of the most significant parameters. Currently, salinity sensors for pond water are designed for momentary measurements, which are unsuitable for continuous monitoring. This study introduces a method for continuous salinity measurement using ultrasonic signals. The proposed approach utilizes a measuring chamber equipped with ultrasonic sensors to determine the Time-of-Flight (ToF). To ensure accuracy, four ToF methods were compared, with the cross-correlation method identified as the most accurate. This method was subsequently used to calculate the ToF, which was then applied to determine the acoustic speed. Since the acoustic speed in water is influenced by salinity, temperature, and pressure, changes in salinity cause detectable changes in the acoustic speed. The acoustic speed was further used as input for the modified Del Grosso equation to derive the salinity. Experimental results showed an average error of 4.83% for saline solutions and 1.81% for shrimp pond water. These findings demonstrate that the proposed method provides sufficient accuracy for water salinity measurement.
Early Warning Safety System Development for Electric Vehicle Batteries to Prevent Fires and Accidents: Implementation in Urban Public Transportation Happyanto, Dedid Cahya; Anita, Jelia; Hendriawan, Akhmad
JURNAL INFOTEL Vol 17 No 4 (2025): November
Publisher : LPPM INSTITUT TEKNOLOGI TELKOM PURWOKERTO

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.20895/infotel.v17i4.1383

Abstract

The increasing adoption of electric vehicles (EVs) in urban public transportation has raised significant safety concerns, particularly regarding thermal runaway incidents that may lead to catastrophic fires. Existing battery monitoring systems often provide inadequate warning times and lack predictive capabilities to mitigate failures before they reach critical conditions. This study proposes an intelligent early warning system for EV battery safety in public transportation fleets by employing predictive analytics. The system integrates a distributed Internet of Things (IoT) sensor network that monitors temperature, voltage, current, and gas emissions, combined with machine learning algorithms—specifically, Random Forest and Support Vector Machine—to analyze battery performance patterns. The proposed architecture incorporates edge computing for real-time data processing and cloud infrastructure for centralised fleet monitoring. Field validation involving 50 electric buses operating under Jakarta's TransJakarta network over a twelve-month period achieved a prediction accuracy of 94.7% for thermal runaway events, with an average warning time of 8.3 minutes. The system successfully prevented 23 potential battery failures while maintaining a false alarm rate below 2.1%. An economic analysis further indicated a favourable cost-benefit ratio of 1:7.4. The proposed solution demonstrates significant potential in enhancing EV battery safety through predictive analytics and automated emergency response, offering a scalable model for broader industry adoption.