🧠 Image Classifier Using VGG19

A web-based image classifier built using Flask and TensorFlow Keras that allows users to upload an image and get real-time predictions using a pre-trained VGG19 model.

---

🚀 Features

• Upload images using drag & drop or file input
• Real-time preview with loading animation
• Predict the object in the image using VGG19
• Displays top prediction using decode_predictions()

---

🖼️ Demo

🔹 Upload Area

🔹 Prediction Result

---

---

🧪 How it Works

1. User uploads or drags an image into the UI.
2. JavaScript displays preview and enables the Predict button.
3. On clicking Predict, image is sent to Flask backend.
4. Flask loads the image, preprocesses it to 224x224, and runs prediction using VGG19.
5. Top prediction is returned and shown to the user.

---

⚙️ Requirements

• Python 3.7+
• TensorFlow
• Flask
• NumPy
• Pillow

---

💻 Installation

git clone https://github.com/yourusername/image-classifier-vgg19.git
cd image-classifier-vgg19

# Install dependencies
pip install -r requirements.txt

🧠 Run the App

python app.py

Then open your browser and go to:

http://127.0.0.1:5000/

---

📸 Add Your Own Model (Optional)

If you want to fine-tune or use a custom model, replace vgg19.h5 with your own .h5 model file.

Make sure to update the preprocessing accordingly in model_predict().

---

🤝 Acknowledgements

• VGG19 model via TensorFlow Keras Applications
• Frontend styled with Bootstrap and FontAwesome