Perancangan Sistem Deteksi Pantulan Reflektor Life Jacket pada Malam Hari Menggunakan Night Vision dan Image Processing

Utomo, Zaydan (2026) Perancangan Sistem Deteksi Pantulan Reflektor Life Jacket pada Malam Hari Menggunakan Night Vision dan Image Processing. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5019221073-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (3MB) | Request a copy

Abstract

Operasi Search and Rescue (SAR) pada malam hari menghadapi kendala akibat minimnya pencahayaan sehingga proses pencarian korban di laut menjadi lebih sulit. Salah satu komponen keselamatan yang dapat dimanfaatkan sebagai target deteksi adalah material retro-reflective pada life jacket yang mampu memantulkan cahaya inframerah dari kamera Night Vision. Penelitian ini bertujuan merancang sistem deteksi pantulan reflektor life jacket yang mengintegrasikan kamera Night Vision analog dan sensor LiDAR secara berkabel dengan komputer sebagai pusat pemrosesan, serta mengevaluasi kinerja sistem dalam mendeteksi reflektor dan memberikan informasi jarak objek. Metode pengolahan citra yang digunakan meliputi Gaussian Smoothing, Morphological Top-Hat Transform, Otsu Thresholding, dan Blob Analysis untuk mengisolasi pantulan reflektor dari latar belakang pada kondisi minim cahaya. Pengujian sistem dilakukan melalui pengujian jarak sensor LiDAR, Detection Rate, variasi sudut, dan tracking objek. Hasil pengujian menunjukkan bahwa sensor LiDAR mampu memberikan informasi jarak dengan tingkat error yang relatif kecil sesuai rentang pengujian. Pengujian keberhasilan deteksi menghasilkan Detection Rate sebesar 100% pada pengujian objek tunggal dan objek pertama pada pengujian dua objek, sedangkan objek kedua memperoleh Detection Rate sebesar 80% akibat satu kejadian False Negative. Selain itu, pengujian sudut menunjukkan bahwa sistem masih dapat mendeteksi pantulan reflektor pada variasi sudut pengujian, sedangkan pengujian tracking menunjukkan seluruh target tetap terdeteksi pada setiap skenario tanpa ditemukan False Negative, meskipun pada beberapa kondisi masih terdapat False Positive. Berdasarkan hasil tersebut, sistem yang dirancang mampu mendeteksi pantulan reflektor life jacket sekaligus memberikan informasi jarak objek secara real-time, sehingga berpotensi menjadi salah satu alternatif teknologi pendukung operasi SAR pada kondisi minim cahaya
=======================================================================================================================================
Nighttime Search and Rescue (SAR) operations face significant challenges due to limited illumination, making it more difficult to locate victims at sea. One of the safety components that can be utilized as a detection target is the retro-reflective material attached to a life jacket, which reflects infrared light emitted by a Night Vision camera. This study aims to design a life jacket reflector detection system by integrating an analog Night Vision camera and a LiDAR sensor through a wired connection to a computer as the central processing unit, as well as to evaluate the system's performance in detecting reflectors and providing object distance information. The proposed image processing method consists of Gaussian Smoothing, Morphological Top-Hat Transform, Otsu Thresholding, and Blob Analysis to isolate reflector reflections from the background under low-light conditions. The system performance was evaluated through LiDAR distance measurement, Detection Rate, detection angle variation, and object tracking tests. The experimental results show that the LiDAR sensor is capable of providing object distance measurements with a relatively low error within the tested range. The reflector detection test achieved a Detection Rate of 100% for the single-object scenario and the first object in the two-object scenario, while the second object achieved a Detection Rate of 80% due to one False Negative occurrence. Furthermore, the angle variation test demonstrated that the system was able to detect reflector reflections across the tested angles, while the object tracking test showed that all targets remained successfully detected in every testing scenario without any False Negative occurrences, although several False Positive detections were observed. Based on these results, the proposed system is capable of detecting life jacket reflector reflections while simultaneously providing real-time object distance information, making it a potential supporting technology for nighttime Search and Rescue (SAR) operations

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Image Processing, Life Jacket, LiDAR, Night Vision, Search and Rescue (SAR), Image Processing, Life Jacket, LiDAR, Night Vision, Search and Rescue (SAR).
Subjects:	T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK5102.9 Signal processing. T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK7871.674 Detectors. Sensors T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK7882.P3 Pattern recognition systems
Divisions:	Faculty of Marine Technology (MARTECH) > Marine Engineering > 36202-(S1) Undergraduate Thesis
Depositing User:	Zaydan Utomo
Date Deposited:	07 Aug 2026 13:07
Last Modified:	07 Aug 2026 13:07
URI:	http://repository.its.ac.id/id/eprint/142790

Actions (login required)

View Item