AutoLPR: Automated License Plate Recognition Based on Itera

Thesis: Identification Of Police Number Characters On Vehicle Plates Using Convolutional Neural Networks

Author

• Miftakhul Aziz
• [email protected]

---

This repository contains the implementation of my final assignment on character identification on vehicle plates using Convolutional Neural Networks (CNNs). The project is divided into three main stages:

1. Object Detection Plate using SSD ResNet-50
2. Object Detection Plate using YOLO-V3
3. Character Recognition using MobileNet-V2

Table of Contents

• Datasets
• Models
• Source Code
• Configuration
• Thesis Document

---

Datasets

The datasets used in this thesis are provided below:

1. Dataset for Object Detection (SSD ResNet-50):

   • Download Link

2. Dataset for Object Detection (YOLO-V3):

   • Download Link

3. Dataset for Character Recognition (MobileNet-V2):

   • Download Link

---

Models

Pre-trained models used in this project can be accessed through the following links:

1. SSD ResNet-50 Models:

   • Download Link

2. YOLO-V3 Models:

   • Download Link

---

Source Code

The source code for testing the YOLO-V3 model can be found at the following link:

• YOLO-V3 Testing Source Code

---

Configuration

Configuration files for the SSD ResNet-50 model can be downloaded from the link below:

• SSD ResNet-50 Configuration Files

---

Thesis Document

The final thesis document, which provides a detailed explanation of the methodologies and results, can be accessed here:

• Final Thesis Document

---

How to Use

1. Clone the Repository:

   git clone https://github.com/mftakhullaziz/automate-lpr.git
   cd automate-lpr