Animal Classification Using Convolutional Neural Networks (CNN)

Developed a deep learning model for classifying animal images into 10 different categories using Convolutional Neural Networks (CNN).

Key Features

- Classified images into 10 animal categories.

- Preprocessed images by converting them to grayscale and resizing them to 50×50 pixels.

- Normalized pixel values to improve model training.

- Built a CNN architecture using TensorFlow and Keras, incorporating convolutional layers, batch normalization, max pooling, and dropout.

- Split the dataset into training and testing sets using Scikit-learn.

- Applied Early Stopping and ReduceLROnPlateau to improve the training process.

- Evaluated model performance using accuracy, a confusion matrix, and a classification report.

- Saved and reloaded the trained model for predictions on new images.

Technologies Used: Python, TensorFlow, Keras, OpenCV, NumPy, Matplotlib, Scikit-learn, and Seaborn.

Test Performance: Achieved approximately 61.76% test accuracy, providing a baseline for evaluating the model and identifying opportunities for further improvement.