🐶 Dog Breed Detection

Dog Breed Detection is a Streamlit web application that predicts the breed of a dog from an uploaded image using a deep learning model. It provides fast, accurate breed identification through a simple, user-friendly interface.

📖 Overview

This Streamlit-based application uses computer vision techniques to:

• Classify uploaded dog images into predefined breeds
• Preprocess and normalize images for consistent model input
• Display the predicted breed with confidence scores
• Enable quick experimentation with a trained CNN model in the browser

---

🚀 Features

• Upload dog images (JPG/PNG) directly in the browser

• Automatic resizing and preprocessing

• Prediction powered by a trained Keras .h5 model

• Clean, interactive Streamlit UI

• Ready for local use or cloud deployment (e.g., Streamlit Community Cloud)

• 🧠 Model

• Built with TensorFlow / Keras

• Trained on a multi-class dog breed dataset

• Saved as dog_breed.h5 and loaded at runtime

• Uses a CNN (optionally with transfer learning) for robust feature extraction

---

🔧 Installation

Prerequisites

• Python 3.8 or higher
• pip (Python package manager)

Step 1: Clone the Repository

-git clone https://github.com/rivu-intel45/dogbreed-deepvision.git

Step 2: Install Dependencies

-pip install -r requirements.txt

---

▶️ Running the App

-streamlit run main_app.py

Then open the URL shown in the terminal (usually http://localhost:8501).

---

🖼️ How to Use

1. Start the app with streamlit run main_app.py.
2. Upload a clear image of a dog.
3. Wait for the model to process the image.
4. View the predicted breed and confidence.

---

📈 Future Improvements

• Support more breeds and larger datasets
• Show top-k predictions with probabilities

💡 Contributing

Contributions are welcome! Feel free to fork, submit issues, or make pull requests.