Sistem Estimasi Kecepatan Kendaraan Menggunakan YOLOv8 Berbasis Video Drone Pada Jetson Nano

Ahmad, Syahrul Fathoni (2025) Sistem Estimasi Kecepatan Kendaraan Menggunakan YOLOv8 Berbasis Video Drone Pada Jetson Nano. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5024211007-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (16MB) | Request a copy

Abstract

Sistem ini dirancang untuk melakukan estimasi kecepatan kendaraan pada NVIDIA Jetson Nano dengan menggabungkan deteksi objek menggunakan YOLOv8 yang telah dioptimasi. Algoritma pelacakan multi-objek OC-SORT yang ringan namun akurat dalam mempertahankan identitas tiap kendaraan antar frame. Kecepatan setiap kendaraan dihitung berdasarkan perpindahan koordinat dalam satuan piksel per frame. Berkat memanfaatkan TensorRT untuk inferensi dan menyusun pipeline image preprocessing yang efisien sistem mampu mencapai throughput antara 8 hingga 12 FPS saat menjalankan deteksi dan pelacakan secara simultan Dengan konfigurasi yang ringkas dan penggunaan perangkat keras yang terjangkau, solusi ini menawarkan alternatif praktis untuk penerapan pengawasan di perkotaan, tanpa memerlukan infrastruktur
pencahayaan atau pemrosesan terpusat yang kompleks.
=====================================================================================================================================
This system is designed to estimate vehicle speed on the NVIDIA Jetson Nano by integrating optimized YOLOv8 object detection with the lightweight yet accurate OC-SORT multi-object tracking algorithm to preserve each vehicle’s identity across frames. Speed is calculated from the per-frame pixel displacement of bounding box coordinates. By leveraging TensorRT for inference and an efficient image-preprocessing pipeline, the system achieves a throughput of 8–12 FPS for simultaneous detection and tracking. With its compact configuration and cost-effective hardware, this solution offers a practical alternative for urban surveillance without requiring additional lighting infrastructure or centralized processing

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Traffic Control, Drone, Speed Estimation, Jetson Nano, YOLOv8, Pengawasan, Drone, Estimasi kecepatan, Jetson Nano, YOLOv8
Subjects:	T Technology > TA Engineering (General). Civil engineering (General) > TA1637 Image processing--Digital techniques. Image analysis--Data processing. T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK5105.546 Computer algorithms T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK6592.A9 Automatic tracking.
Divisions:	Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Computer Engineering > 90243-(S1) Undergraduate Thesis
Depositing User:	Syahrul Fathoni Ahmad
Date Deposited:	20 Jun 2025 03:58
Last Modified:	20 Jun 2025 03:58
URI:	http://repository.its.ac.id/id/eprint/119181

Actions (login required)

View Item