Claim Missing Document
Check
Articles

Found 4 Documents
Search

Vehicle Theft Detection Using YOLO Based on License Plates and Vehicle Ownership Bradika Almandin Wisesa; M. Hizbul Wathan; Evvin Faristasari; Sirlus Andreanto Jasman Duli; Silvia Agustin; Better Swengky
International Journal of Informatics and Computation Vol. 7 No. 1 (2025): International Journal of Informatics and Computation
Publisher : University of Respati Yogyakarta, Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35842/ijicom.v7i1.105

Abstract

Detection of vehicle theft requires innovative approaches to address an increasing number of cases in Indonesia. This study presents a YOLOv11-based system for detecting vehicle theft by combining real-time object detection with a vehicle ownership database. The proposed system identifies license plates, detects vehicle owners using facial recognition, and analyzes suspicious activity to determine theft occurrences. The proposed method can produce model effectiveness with an accuracy = 70%. Key improvements in architecture, including enhanced feature fusion and dynamic anchor assignment, contribute to the object’s detection in complex environments. This research can be a potential technique to provide efficient, scalable, and real-time security solutions in dynamic surveillance applications.
Simulation of Solar Panel Design as an Energy Source for Catfish Ponds Maulana, Ade Putra; Duli, Sirlus Andreanto Jasman; Istoto, Enggar Hero; Peprizal, Peprizal; Faristasari, Evvin
Journal of Technology and Engineering Vol 3 No 1 (2025): Journal of Technology and Engineering
Publisher : Yayasan Banu Haji Samsudin

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.59613/journaloftechnologyandengineering.v3i1.214

Abstract

The increasing demand for sustainable energy solutions has driven the adoption of solar photovoltaic (PV) systems in various industries, including aquaculture. This study designs and simulates a solar power system for small-scale catfish (Lele) pond operations using the System Advisor Model (SAM). The methodology includes assessing energy requirements, selecting system components, conducting simulations, and performing an economic feasibility analysis. The results indicate that the designed 12-panel, 3-battery solar system effectively meets the pond’s daily energy demand while ensuring continuous operation during low sunlight conditions. The SAM simulation confirms stable electricity generation throughout the year, with seasonal variations minimally affecting efficiency. The economic analysis reveals that PLTS costs Rp. 150,365 per month, compared to Rp. 151,620 for PLN electricity, showing small but valuable long-term savings. Despite the high initial investment, solar power offers price stability, energy independence, and reduced reliance on fossil fuels. This study demonstrates that solar energy is a viable, cost-effective, and sustainable alternative for aquaculture operations. Future research should focus on optimizing system efficiency and integrating hybrid energy solutions to further enhance performance and financial benefits.
Penerapan YOLOv11 untuk Penghitungan Otomatis Jumping Jack pada Video Latihan Fisik Wisesa, Bradika Almandin; Putri, Vivin Mahat; Faristasari, Evvin; Duli, Sirlus Andreanto Jasman; Irawan, Indra; Agustin, Silvia
Jurnal Pengembangan Sistem Informasi dan Informatika Vol. 6 No. 3 (2025): Jurnal Pengembangan Sistem Informasi dan Informatika
Publisher : Training & Research Institute - Jeramba Ilmu Sukses

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47747/jpsii.v6i3.2795

Abstract

The Jumping Jack Counter is an image processing-based application developed to automatically count the number of jumping jack movements in exercise videos. This study aims to implement the YOLOv11 model to detect and count jumping jack movements by analyzing body posture. YOLOv11 is utilized to identify body positions categorized into two main classes: "open" (arms and legs spread apart) and "closed" (arms and legs together). The dataset consists of 15,000 video frames collected from various exercise videos, with research stages including data collection, data labeling, preprocessing, model training, and testing. The results demonstrate that YOLOv11 achieves a 92% accuracy rate in counting jumping jack movements. These findings are expected to assist coaches and users in monitoring physical exercise in real-time, thereby enhancing training effectiveness. The majority of movement detections (78%) were for the open position, followed by the closed position (20%), with 2% detection errors attributed to lighting variations or camera angles. [1].
Laplacian Kernel and Deep Learning for Palmprint Classification Duli, Sirlus Andreanto Jasman; Wisesa, Bradika Almandin; Faristasari, Evvin; Peprizal, Peprizal; Putri, Vivin Mahat; Fadila, Resma
Journal of Artificial Intelligence and Software Engineering Vol 5, No 2 (2025): Juni On-Progress
Publisher : Politeknik Negeri Lhokseumawe

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30811/jaise.v5i2.6978

Abstract

Palmprint classification is a robust biometric method for personal identification due to its uniqueness and stability. This study explores the use of deep learning combined with the Laplacian Kernel and Deep Morphological Processing Network (DMPN) for palmprint classification. We trained the proposed system on a dataset of palmprint images collected from 10 participants, each contributing 10 palm images. The results demonstrated that the model achieved an accuracy of 90%, with weighted precision, recall, and F1-score all at 0.9007, indicating a well-balanced classification performance. Additionally, the model achieved a weighted precision of 0.9045, emphasizing its ability to minimize false positives. The average Equal Error Rate (EER) of 0.0917 indicates an effective balance between the false acceptance rate (FAR) and false rejection rate (FRR). The system was tested under various conditions, including different orientations, lighting, and backgrounds, demonstrating its robustness in real-world scenarios. This study also compares the results with recent palmprint classification techniques, such as deep learning, GANs, and few-shot learning, and discusses potential improvements, including incorporating multi-spectral data fusion and few-shot learning to enhance performance in real-world applications.