Perancangan Sistem Monitoring Rip Current Pada UAV Cerdas Dengan Integrasi Computer Vision Dan Deep Learning

Prayoga, Muhammad Nur (2026) Perancangan Sistem Monitoring Rip Current Pada UAV Cerdas Dengan Integrasi Computer Vision Dan Deep Learning. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5022221152-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (4MB) | Request a copy

Abstract

Rip Current (arus balik) merupakan fenomena arus laut berbahaya yang sulit dikenali secara visual namun menjadi penyebab utama kecelakaan laut di pantai. Kurangnya sistem pemantauan otomatis dan keterbatasan deteksi visual oleh manusia menjadi tantangan besar dalam mitigasi risiko terhadap wisatawan pantai di Indonesia, khususnya pantai selatan pulau Jawa. Oleh karena itu, penelitian ini bertujuan untuk merancang sistem monitoring rip current berbasis Unmanned Aerial Vehicle (UAV) cerdas yang mengintegrasikan teknologi computer vision dan deep learning untuk mendeteksi dan memetakan area rip current secara real-time. Metodologi penelitian diawali dengan studi literatur untuk mengidentifikasi ciri visual rip current dan teknik deteksi berbasis citra. UAV dilengkapi dengan kamera resolusi tinggi digunakan untuk akuisisi citra pantai dari udara. Data yang diperoleh diproses melalui model Convolutional Neural Network (CNN), khususnya varian segmentasi YOLOv8-seg, yang telah dilatih dengan dataset lokal dan diperluas melalui augmentasi. Sistem diprogram untuk mengenali karakteristik visual seperti perbedaan warna air, buih yang mengarah ke laut, dan permukaan air yang tenang. Sistem ini mampu mendeteksi rip current secara cepat dan akurat, serta memberikan visualisasi area rawan kepada operator pantai. Hasil yang ditargetkan mencakup kemampuan segmentasi spasial area rip current, serta integrasi operasional UAV untuk pemantauan adaptif. Penelitian ini diharapkan menjadi solusi inovatif dalam upaya keselamatan pesisir berbasis kecerdasan buatan yang dapat diterapkan di berbagai wilayah pantai tropis.
===================================================================================================================================
Rip currents are hazardous ocean phenomena that are difficult to recognize visually yet are a leading cause of water-related accidents along beaches. The absence of automated monitoring systems and the limited ability of humans to detect rip currents in real-time present serious challenges for ensuring tourist safety in coastal areas. This study aims to design an intelligent rip current monitoring system using Unmanned Aerial Vehicles (UAVs) integrated with computer vision and deep learning technologies to detect and map of rip current zones in real-time. The research methodology begins with a literature review to identify the visual characteristics of rip currents and image-based detection techniques. High-resolution cameras mounted on UAVs are employed to acquire aerial imagery of coastal areas. The collected data is processed using a Convolutional Neural Network (CNN), specifically the YOLOv8-seg segmentation variant, trained on localized datasets and enhanced through data augmentation. The system is programmed to detect visual indicators such as darker water patterns, foam drifting seaward, and calm surface zones. This system provided rapid and accurate detection of rip currents while visually highlighting hazard zones for beach operators. The anticipated outcomes include spatial segmentation detection of rip current areas, and seamless UAV integration for adaptive coastal monitoring. Ultimately, this research aims to offer an innovative, AI-driven solution for coastal safety that can be applied across a range of tropical beach environments.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Computer Vision, Deep Learning, Sistem Pemantauan, Rip Current, UAV (Unmanned Aerial Vehicle).
Subjects:	T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK1007 Electric power systems control T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK7888.3 Digital computers
Divisions:	Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Electrical Engineering > 20201-(S1) Undergraduate Thesis
Depositing User:	Muhammad Nur Prayoga
Date Deposited:	23 Jul 2026 04:12
Last Modified:	23 Jul 2026 04:12
URI:	http://repository.its.ac.id/id/eprint/134185

Actions (login required)

View Item