Perancangan Program Pendeteksi Posisi Life Jacket pada Adrift Survivor Berbasis Computer Vision dengan Metode Image Processing Berbasis Warna

Khalif, Affan (2025) Perancangan Program Pendeteksi Posisi Life Jacket pada Adrift Survivor Berbasis Computer Vision dengan Metode Image Processing Berbasis Warna. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5019211055-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (3MB) | Request a copy

Abstract

Penelitian ini bertujuan merancang sistem pendeteksi posisi life jacket pada adrift survivor di laut menggunakan metode pengolahan citra berbasis warna dan sensor LIDAR. Sistem ini menggunakan kamera untuk mendeteksi warna oranye khas life jacket melalui segmentasi HSV, dan mengintegrasikannya dengan data jarak dan sudut dari RPLIDAR A1 untuk memperoleh posisi spasial objek secara real-time. Program dikembangkan menggunakan bahasa C++ dan pustaka OpenCV dengan pendekatan multithreading agar kamera dan LIDAR bekerja sinkron. Pengujian dilakukan di lingkungan perairan pada siang dan malam hari dengan variasi jarak dan sudut 0. Hasil menunjukkan bahwa sistem mampu mendeteksi objek secara akurat, dengan rata-rata error jarak sebesar 2,60% di siang hari dan 3,09% di malam hari. Sedangkan rata-rata error sudut sebesar 0,002% di siang dan malam hari Luas piksel yang terdeteksi menurun seiring bertambahnya jarak, menunjukkan korelasi antara ukuran visual objek dan jaraknya terhadap kamera. Sistem ini menunjukkan potensi besar untuk diterapkan dalam operasi SAR (search and rescue) maritim berbasis otonom.
=================================================================================================================================
This research aims to design a system for detecting the position of life jackets on adrift survivors at sea using color-based image processing methods and LIDAR sensors. The system uses a camera to detect the distinctive orange color of life jackets through HSV segmentation, and integrates this with distance and angle data from the RPLIDAR A1 to obtain the real-time spatial position of the object. The program was developed using the C++ language and the OpenCV library with a multithreading approach to ensure that the camera and LIDAR work synchronously. Testing was conducted in a water environment during the day and at night with varying distances and angles of 0. The results show that the system is able to detect objects accurately, with an average distance error of 2.60% during the day and 3.09% at night. Meanwhile, the average angle error was 0.002% during both daytime and nighttime. The detected pixel area decreases as the distance increases, indicating a correlation between the visual size of the object and its distance from the camera. This system shows great potential for application in autonomous maritime search and rescue (SAR) operations.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Adrift Survivor, Computer Vision, Deteksi Warna, HSV, LIDAR, Life Jacket,Adrift Survivor, Color Detetction, Computer Vision, HSV, LIDAR, Life Jacket
Subjects:	T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK7871.674 Detectors. Sensors V Naval Science > VK > VK200 Merchant marine--Safety measures
Divisions:	Faculty of Marine Technology (MARTECH) > Marine Engineering > 36202-(S1) Undergraduate Thesis
Depositing User:	Affan Khalif
Date Deposited:	04 Aug 2025 06:43
Last Modified:	04 Aug 2025 06:43
URI:	http://repository.its.ac.id/id/eprint/127020

Actions (login required)

View Item