Article Per Year (5 Year)
p-Index From 2021 - 2026
0.408
P-Index
This Author published in this journals
All Journal International Journal of Advances in Intelligent Informatics Jurnal Teknimedia: Teknologi Informasi dan Multimedia
Thomas Brian
Politeknik Perkapalan Negeri Surabaya
Computer Science & IT Control & Systems Engineering Engineering

Published : 2 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 2 Documents
Search

 1 
Iron welding spot segmentation using Nested UNet (UNet++) enhanced with diverse convolutional modules Thomas Brian; Oskar Natan; Yohanes Yohanie Fridelin Panduman; Anggarjuna Puncak Pujiputra
International Journal of Advances in Intelligent Informatics Vol 12, No 2 (2026): May 2026
Publisher : Universitas Ahmad Dahlan

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

Welding inspection plays an essential role in manufacturing industries to ensure the integrity and quality of weld joints. However, the prevalent manual inspection procedures are inherently subjective, prone to bias, and result in inconsistent quality assessments. Therefore, there is a strong need for an automated, intelligent system capable of objectively detecting welding spots. To address this, we propose an advanced segmentation model based on deep learning and computer vision techniques, specifically utilizing a Nested UNet (UNet++) architecture enhanced by extensive architectural modifications and comprehensive hyperparameter tuning. To further optimize segmentation performance, we systematically compare various convolutional blocks integrated into the bottleneck of the network architecture. Our experimental evaluation demonstrates that employing a VGG convolutional block at the bottleneck of Nested UNet achieves the highest performance, reaching an Intersection over Union (IoU) score of 76.18% and a validation loss of 0.1713 on our collected dataset.
SISTEM DETEKSI KERUSAKAN PANEL PLTS APUNG DI EMBUNG SIDOBANDUNG BERBASIS CONVOLUTIONAL NEURAL NETWORK DENGAN VISUALISASI AUGMENTED REALITY Thomas Brian; Immanuel Freddy Augustino; Parman Parman; Muhamad Sukarno
TEKNIMEDIA: Teknologi Informasi dan Multimedia Vol. 7 No. 1 (2026): June 2026
Publisher : Badan Penelitian dan Pengabdian Masyarakat (BP2M) STMIK Syaikh Zainuddin NW Anjani

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.46764/teknimedia.v7i1.364

Abstract

This study aims to develop an Augmented Reality (AR) application integrated with a Convolutional Neural Network (CNN) as an interactive system for detecting damage in floating solar power plant (PLTS) panels at Embung Sidobandung in order to maintain the efficiency of the photovoltaic energy system. Conventional manual inspection methods are considered inefficient and prone to errors due to human factors. Therefore, a deep learning approach is employed to automatically and interactively detect and classify solar panel damage. AR technology is utilized to display panel condition information directly through a mobile device camera, enabling real-time damage monitoring. The dataset consists of 615 solar panel images, including 472 images of physical damage and 143 images of electrical damage. Experimental results show that the system is capable of classifying solar panel damage types in real time, achieving a precision of 93.48%, recall of 89.58%, and an F1-score of 91.49% for physical damage, and a precision of 70.59%, recall of 80.00%, and an F1-score of 75.00% for electrical damage, with an overall accuracy of 87.30%. Although the developed application provides interactive and informative visualization, varying lighting conditions in aquatic environments and differences in image acquisition angles remain challenges that affect system accuracy. Overall, the integration of CNN and AR has the potential to serve as an effective and efficient solution for developing damage detection systems for floating solar power plant (PLTS) panels.
Co-Authors Anggarjuna Puncak Pujiputra Immanuel Freddy Augustino Muhamad Sukarno Oskar Natan Parman Parman Yohanes Yohanie Fridelin Panduman