Sure, here's an updated version of the Markdown file with more details:

Egy-Currency-Detectron

This repository contains code for detecting and recognizing Egyptian currency using YOLOv8.

Introduction

YOLOv8 is a popular and powerful object detection model that uses a single neural network to predict bounding boxes and class probabilities directly from full images in one evaluation. In this project, we use YOLOv8 to detect and recognize Egyptian currency notes.

Dataset

We used a dataset of Egyptian currency notes that we found at roboflow. The dataset consists of 5 classes of currency notes: 1 EGP, 5 EGP, 10 EGP, 20 EGP, 200 EGP , and 50 EGP. The dataset contains a total of 2700 images with bounding box annotations for each currency note.

Model

We fine-tuned the pre-trained YOLOv8 model on our dataset using Ultralytics implementation . Ultralytics is an open source repository that is used to train YOLO models. The YOLOv8 model is a modified version of YOLOv5 that uses a larger backbone network and more training data to achieve better performance.

Requirements

Python 3.x
PyTorch
OpenCV

Installation

Clone the repository:

git clone https://github.com/husseinmleng/Egy-Currency-Detectron.git

Install the dependencies:
```
pip install -r requirements.txt
```
Download the trained weights best.pt and place them in the weights directory.

Usage

To detect and recognize currency from an image, run the following command:

python detect.py --image <path_to_image>

For example:

python detect.py --image test_images/image1.jpg

Results

Below are some examples of the currency detection and recognition results:

As shown in the results, our model is able to accurately detect and recognize the different classes of Egyptian currency notes.