🐶🐱 Image Classification using CNN A deep learning project that builds and trains a Convolutional Neural Network (CNN) using TensorFlow/Keras to classify images of cats and dogs. The model is trained on augmented image data, evaluated with performance metrics, and tested on new images with visualization support.

🚀 Features CNN built using TensorFlow/Keras Sequential API

Real-time image data augmentation

Training/Validation accuracy and loss visualization

Test-time prediction with image enhancement

Evaluation metrics: Accuracy, Precision, Recall, F1 Score

Confusion matrix visualization

Learning rate scheduler (ReduceLROnPlateau)

Dropout regularization for better generalization

🛠️ Tech Stack & Tools Language: Python

Framework: TensorFlow 2.x, Keras

Libraries: NumPy, Matplotlib, PIL, Scikit-learn

Data Preprocessing: ImageDataGenerator

Image Enhancement: ImageEnhance, ImageFilter (PIL)

Visualization: Matplotlib, ConfusionMatrixDisplay

📁 Dataset Structure Copy Edit dataset/ ├── training_set/ │ ├── cats/ │ └── dogs/ ├── test_set/ │ ├── cats/ │ └── dogs/ └── single_prediction/ └── dog.jpg 📊 Model Summary Conv2D → MaxPool → Conv2D → MaxPool → Dropout

Flatten → Dense → Dropout → Output Layer (Sigmoid)

Optimizer: Adam with learning rate 0.0001

Loss: binary_crossentropy

Epochs: 25

🧪 Evaluation Metrics Accuracy, Precision, Recall, F1 Score

Confusion matrix plot

Epoch-wise training/validation accuracy and loss plots

🖼️ Prediction Output Example Upload a custom image for single prediction

Image is preprocessed, enhanced, and passed through the trained model

The model outputs whether it’s a Cat or a Dog with visualization

📷 Sample Output

🙌 Acknowledgements TensorFlow/Keras

Scikit-learn for metrics

PIL for image enhancement