Computer-Vision-using-Python

Projects:

1. Lung Image Segmentation

This project is a Lung Segmentation deep learning pipeline using PyTorch Lightning and UNet for medical image segmentation. It trains a convolutional neural network (CNN) to segment lung regions from medical images, leveraging data augmentation, custom loss functions (Dice + BCE Loss), and pretrained weights to improve performance. The model is trained using Adam optimizer with a learning rate scheduler, and results are logged with TensorBoard. The training process includes dataset splitting (train/val/test), checkpoint saving, and evaluation on test data.