Image Processing

This repository provides a collection of assignments, experiments, and implementations in Image Processing and Deep Learning.
It covers classical techniques (augmentation, filtering, morphology, compression) as well as modern neural network–based methods (CNNs, GANs, UNet, etc.).

📂 Repository Structure

• 1. Augmentation, Morphology and Restoration
  Basic image operations including augmentation, morphological transformations, and restoration techniques.

• 2. Smoothing and Edge Detection
  Image filtering with smoothing operations and edge detection algorithms.

• 3. Classification Using LeNet
  Implementation of LeNet-5 CNN for image classification tasks.

• 4. Segmentation Using UNet
  Deep learning–based image segmentation using U-Net architecture with Dice coefficient evaluation.

• 5. DCGAN Faces
  Implementation of Deep Convolutional GAN to generate realistic human faces.

• 5. GAN
  General GAN experiments exploring adversarial training for image generation.

• 6. JPEG Compression
  Implementation of JPEG compression pipeline, including DCT and quantization.

• Assignments
  Collection of smaller image processing exercises and lab work.

⚙️ Requirements

Install dependencies with:

pip install -r requirements.txt

## Main Libraries

- `numpy`
- `matplotlib`
- `opencv-python`
- `scikit-learn`
- `pytorch` (for GAN/UNet parts, if included)


## Getting Started

Clone the repository:

```bash
git clone https://github.com/Abhijeet-KC/Image_Processing.git
cd Image_Processing

Highlights

• Classical image processing operations (augmentation, morphology, filtering, edge detection)
• Compression implementation with JPEG pipeline
• CNN for classification (LeNet)
• Semantic segmentation with UNet
• Generative Adversarial Networks (GAN, DCGAN) for synthetic image generation

👥 Contributors

1. Abhijeet K.C.
2. Aaditya Joshi
3. Ankit Neupane
4. Shreejay Shakya