🧠 AI Image Classification System

A powerful deep learning-based image classifier trained on the CIFAR-10 dataset using Transfer Learning (MobileNetV2).

🌟 Overview

This project demonstrates the power of Transfer Learning by fine-tuning the MobileNetV2 architecture to classify images into 10 distinct categories with high accuracy. It features a full-stack implementation with a Flask web interface for real-time predictions.

✨ Features

• 🚀 Deep Learning: Powered by MobileNetV2 (pre-trained on ImageNet).
• 📂 Standard Dataset: Trained on the robust CIFAR-10 dataset (60,000 images).
• 💻 User Interface: specific Flask web app for easy image uploading and testing.
• 📊 Visualization: Includes real-time confidence charts and confusion matrices.
• ⚡ Fast Predictions: Optimized for speed using TensorFlow/Keras.

📊 Dataset Stats

Metric	Count
Total Images	60,000
Training Set	50,000
Test Set	10,000
Classes	10 (Airplane, Auto, Bird, Cat, Deer, Dog, Frog, Horse, Ship, Truck)

🛠️ Tech Stack

• Languages: Python
• Libraries: TensorFlow, Keras, NumPy, Pandas, Matplotlib, Seaborn
• Web Framework: Flask
• Frontend: HTML5, CSS3

🚀 Getting Started

1. Installation

Clone the repository and install the required dependencies:

git clone https://github.com/VibeCoder-Saad/Ai-Image-classification-system-
cd Ai-Image-classification-system-
pip install -r requirements.txt

2. Training the Model (Optional)

If you want to retrain the model from scratch:

python train_dl_model.py

This will generate model files in models/dl_models and plots in static/plots.

3. Running the App

Launch the web interface:

python app.py

Visit http://localhost:5000 in your browser!

📂 Project Structure

Ai-Image-classification-system-/
├── app.py                 # 🚀 Main Flask Application
├── train_dl_model.py      # 🧠 Training Script
├── requirements.txt       # 📦 Dependencies
├── utils/                 # 🛠️ Utility Scripts
│   ├── preprocessing.py
│   └── evaluation.py
├── models/                # 💾 Saved Models
├── static/                # 🎨 Static Assets (CSS, JS, Images)
└── templates/             # 📄 HTML Templates

---

Made with ❤️ by Saad