Perdana, Reza Rizki (2019) Pengaruh Pemilihan Metode Ekstrasi Fitur pada Sistem Deteksi Plat Nomor Kendaraan. Other thesis, Institut Teknologi Sepuluh Nopember.
Preview |
Text
07111540000081-Undergraduate_Theses.pdf Download (4MB) | Preview |
Abstract
Pengenalan plat nomor kendaraan menjadi kebutuhan yang sangat krusial untuk menunjang ketertiban lalu lintas di jalan raya, yang dipantau menggunakan bantuan sistem monitoring dari Closed Circuit Television (CCTV) yang terpasang statis di atas lampu – lampu traffic light. Tujuan CCTV tersebut untuk melihat kondisi kendaraan yang lewat di bawahnya, khususnya memantau kendaraan yang melanggar aturan lalu lintas sehingga perlu di kenai sanksi berupa tilang. Target pantauan CCTV lebih ditujukan kepada lokasi plat nomor kendaraan, karena lebih mudah dilihat oleh kamera CCTV. Hanya saja jumlah kendaraan yang banyak ditambah kondisi pencahayaan yang kurang memadai sering kali terjadi kendala dalam hal tersebut. Untuk itu teknologi yang berbasis Artivisial Intelligent (AI) perlu diterapkan untuk meningkatkan kemampuan deteksi sehingga akurasi pengenalan plat nomor menjadi meningkat. Dalam Tugas Akhir ini akan dibuat aplikasi sistem deteksi lokasi plat nomor kendaraan, menggunakan algoritma ekstrasi fitur yaitu algoritma Dilation dan Haar Wavelet Transform (HWT) untuk mengambil fitur local citra yang diamati. Untuk menganalisa lebih lanjut menggunakan metode Horizontal Edge Processing dan Vertical Edge Processing, selain proses grey scale, binerisasi dan cropping citra. Adapun tahapan pembuatan sistem deteksi meliputi pra-prosrsing, ekstrasi fitur, seleksi fitur dan deteksi plat nomor. Selanjutnya klasifikasi deteksi akan dikelompokan dalam tiga kategori yaitu terdeteksi utuh, terdeteksi sebagian dan tidak terdeteksi sama sekali. Untuk menyelesaikan permasalahan Tugas Akhir ini digunakan alat bantu prengkat lunak Matlab dan dataset sebanyak 50 foto yang di buat sendiri. Dengan aplikasi sistem deteksi plat nomor kendaraan yang dibuat dalam Tugas Akhir, diharapkan dapat diimplementasikan untuk mendukung terciptanya sistem transportasi cerdas (Intelligent Transportation System).
================================================================================================================================
The introduction of vehicle number plates is a very crucial requirement to support traffic order on the highway, which is monitored using a Closed Circuit Television (CCTV) monitoring system installed on top of traffic lights. The purpose of the CCTV is to see the condition of the vehicle that passes below it, especially monitoring vehicles that violates the traffic rules so that sanctions need to be taken in the form of a ticket. The CCTV monitoring target is more aimed at the location of the vehicle number plate, because it is easier to be seen by CCTV cameras. It's just that a large number of vehicles plus inadequate lighting conditions often occur obstacles in this case. For this reason, technology based on Artivisial Intelligent (AI) needs to be applied to improve detection capabilities so that the accuracy of number plate recognition increases.
In this Final Project, a vehicle number plate location detection system application will be made, using feature extraction algorithms, the Dilation and Haar Wavelet Transform (HWT) algorithm to retrieve the observed local image features. To further analyze using the Horizontal Edge Processing and Vertical Edge Processing, in addition to the gray scale process, binaryization, and image cropping. The stages of making detection systems include pre-processing, feature extraction, feature selection and number plate detection. Furthermore, the detection classification will be grouped into three categories, which are detected intact, detected partially and not detected at all. To solve this Final Project problem Matlab software tools and datasets are used as many as 50 self-made photos. With the application of the vehicle number plate detection system made in the Final Project, it is expected to be implemented to support the creation of intelligent transportation systems (Intelligent Transportation System).
| Item Type: | Thesis (Other) |
|---|---|
| Additional Information: | RSE 006.42 Per p-1 2019 |
| Uncontrolled Keywords: | Dillation, Deteksi Plat Nomor, Haar Wavelet transform |
| Subjects: | T Technology > TA Engineering (General). Civil engineering (General) > TA1637 Image processing--Digital techniques T Technology > TK Electrical engineering. Electronics Nuclear engineering |
| Divisions: | Faculty of Electrical Technology > Electrical Engineering > 20201-(S1) Undergraduate Thesis |
| Depositing User: | Reza Rizki Perdana |
| Date Deposited: | 22 Feb 2023 08:07 |
| Last Modified: | 22 Feb 2023 08:07 |
| URI: | http://repository.its.ac.id/id/eprint/63821 |
Actions (login required)
 |
View Item |