DESYKA, SHELLA NOVA (2023) PERANCANGAN SISTEM PENDETEKSI PLAT NOMOR GANJIL GENAP MENGGUNAKAN METODE YOLOV5. S1 thesis, Universitas Mercu Buana Jakarta.
|
Text (COVER)
Cover.pdf Download (570kB) | Preview |
|
|
Text (ABSTRAK)
Abstra.pdf Download (133kB) | Preview |
|
![[img]](https://repository.mercubuana.ac.id/style/images/fileicons/text.png) |
Text (BAB I)
Bab I.pdf Restricted to Registered users only Download (184kB) |
|
|
Text (BAB II)
Bab II.pdf Restricted to Registered users only Download (575kB) |
|
|
Text (BAB III)
Bab III.pdf Restricted to Registered users only Download (274kB) |
|
|
Text (BAB IV)
Bab IV.pdf Restricted to Registered users only Download (1MB) |
|
|
Text (BAB V)
Bab V.pdf Restricted to Registered users only Download (130kB) |
|
|
Text (DAFTAR PUSTAKA)
Daftar Pustaka.pdf Restricted to Registered users only Download (129kB) |
Abstract
The odd-even rule on vehicle license plates aims to reduce congestion that occurs. The application of these regulations is constrained by the limitations of the manual oversight function by officers. This problem can be overcome by implementing intelligence in the form of detecting license plate objects with the YOLO v5 algorithm. Object detection technology will detect objects in the form of vehicle plates. YOLO is a new approach to object detection systems, targeted for real-time processing. YOLO frames object detection as a single regression problem, where from image pixels go directly to spatially separated bounding boxes and their associated class probabilities. The google colab library was used to complete this research. Based on this experiment, the yolov5s model obtained a [email protected] value of 55.8% which was in a batch size of 15 with an epochs of 200 to the recall value. The peak value of the average recall gets a value of 0.96 at a confidence value of 0.00. And the total value of accuracy gets a value of 64%. The implementation of the YOLO Algorithm has succeeded in detecting license plates with odd and even categories. Keyword: You Only Look Once, Objek Detection, YOLOv5s, GitHub, Python Aturan ganjil genap pada pelat nomor kendaraan bertujuan untuk mengurangi kemacetan yang terjadi. Penerapan peraturan tersebut terkendala oleh keterbatasan fungsi pengawasan manual oleh petugas. Permasalahan tersebut dapat diatasi dengan mengimplementasikan kecerdasan berupa pendeteksian objek plat nomor dengan algoritma YOLO v5. Teknologi pendeteksi objek akan mendeteksi objek berupa plat kendaraan. YOLO adalah sebuah pendekatan baru untuk sistem pendeteksan objek, yang ditargetkan untuk pemrosesan secara real-time. YOLO membingkai pendeteksian objek sebagai masalah regresi tunggal, dimana dari piksel gambar langsug ke kotak pembatas spasial yang terpisah dan probabilitas kelas yang terkait. Digunakan library google colab untuk menyelesaikan penelitian ini. Berdasarkan percobaan ini model yolov5s memperoleh nilai mAP@0,5 sebesar 55,8% yang berada pada batch size 15 dengan epochs 200 terhadap nilai recall. Nilai puncak rata-rata recall mendapatkan nilai 0,96 pada nilai confidence 0,00. Dan total nilai akurasi mendapatkan nilai 64%. Implementasi Algoritma YOLO berhasil mendeteksi plat nomor dengan kategori ganjil dan genap. Kata Kunci: You Only Look Once, Deteksi Objek, YOLOv5s, GitHub, Python
| Item Type: | Thesis (S1) |
|---|---|
| Call Number CD: | FT/ELK. 23 124 |
| Call Number: | ST/14/23/105 |
| NIM/NIDN Creators: | 41419110116 |
| Uncontrolled Keywords: | You Only Look Once, Deteksi Objek, YOLOv5s, GitHub, Python |
| Subjects: | 600 Technology/Teknologi > 620 Engineering and Applied Operations/Ilmu Teknik dan operasi Terapan 600 Technology/Teknologi > 620 Engineering and Applied Operations/Ilmu Teknik dan operasi Terapan > 621 Applied Physics/Fisika terapan > 621.3 Electrical Engineering, Lighting, Superconductivity, Magnetic Engineering, Applied Optics, Paraphotic Technology, Electronics Communications Engineering, Computers/Teknik Elektro, Pencahayaan, Superkonduktivitas, Teknik Magnetik, Optik Terapan, Tekn |
| Divisions: | Fakultas Teknik > Teknik Elektro |
| Depositing User: | Annas Tsabatulloh |
| Date Deposited: | 20 Sep 2023 03:05 |
| Last Modified: | 20 Sep 2023 03:05 |
| URI: | http://repository.mercubuana.ac.id/id/eprint/80778 |
Actions (login required)
 |
View Item |
