🐱 Cat vs Dog Image Classifier using CNN

🎓 Internship Details

Internship Provider: CodeTech IT Solutions

Intern ID: CITS1618

Full Name: Nuka Aravindh

Duration: 4 Weeks

📌 Project Overview

This project is a Deep Learning-based Image Classification system that predicts whether an uploaded image is a Cat or a Dog. The model is built using Convolutional Neural Networks (CNN) with TensorFlow and Keras and deployed using Streamlit and Render.

🚀 Live Demo

https://image-classifier-cnn-1.onrender.com

✨ Features

• Upload JPG, JPEG, and PNG images
• Real-time Cat vs Dog prediction
• Prediction confidence score
• Interactive Streamlit web application
• Deep Learning model built using CNN

🛠️ Technologies Used

• Python
• TensorFlow
• Keras
• NumPy
• Pillow
• Streamlit
• Git & GitHub
• Render

🧠 Model Architecture

• Convolution Layer
• Max Pooling Layer
• Convolution Layer
• Max Pooling Layer
• Flatten Layer
• Dense Layer
• Output Layer (Sigmoid)

📊 Performance

• Validation Accuracy: 86.31%
• Validation Loss: 0.4953

📂 Dataset

Dataset: Microsoft Cats vs Dogs Dataset

Source: https://www.microsoft.com/en-us/download/details.aspx?id=54765

📁 Project Structure

Image_Classifier_CNN/

├── models/

│ └── cat_dog_classifier.keras

├── Notebook/

│ └── cat_dog_classifier.ipynb

├── Screenshots/

├── app.py

├── requirements.txt

├── runtime.txt

├── .python-version

└── README.md

⚙️ Installation

Clone Repository

git clone https://github.com/aravindh-nuka/Image_Classifier_CNN.git

Install Dependencies

pip install -r requirements.txt

Run Application

streamlit run app.py

🎯 Future Improvements

• Transfer Learning using VGG16 and ResNet50
• Multi-Class Animal Classification
• Webcam Prediction
• Mobile Deployment

👨‍💻 Author

Aravindh Nuka