✨ 22AIE313 - COMPUTER VISION & IMAGE PROCESSING ✨ ASSIGNMENT 1

Roll : CH.EN.U4AIE22015 | CH.EN.U4AIE22050
Name : Guruprasath M R | Shree prasad M

Overview

This assignment involves various image processing techniques using Python libraries such as OpenCV, NumPy, and scikit-image. The tasks include:

1. Loading and Filtering Noisy Images: Applying different filters to noisy images and evaluating their performance using PSNR and SSIM metrics.
2. Binary Segmentation: Implementing binary segmentation algorithms like K-Means and Otsu's Thresholding.
3. Multi-Region Segmentation: Applying multi-region segmentation algorithms such as Felzenszwalb's algorithm, SLIC, and Mean Shift.
4. Combining Segmentations: Combining binary and multi-region segmentations and refining the results.
5. Evaluation: Evaluating the segmentation results using metrics like Variation of Information (VI), Adapted Rand Error (ARE), and Dice Coefficient.

Problem Statement

The objective of this assignment is to explore and implement various image processing techniques to enhance and segment noisy images. The key challenges include:

• Effectively reducing noise in images while preserving important details.
• Accurately segmenting images into meaningful regions using both binary and multi-region segmentation techniques.
• Combining different segmentation methods to improve overall segmentation quality.
• Evaluating the performance of the implemented techniques using appropriate metrics.

Dependencies

• Python 3.x
• OpenCV
• NumPy
• Matplotlib
• scikit-image
• scikit-learn
• pandas
• tqdm

Instructions

1. Setup: Ensure all dependencies are installed. You can use the following command to install the required packages:

   pip install opencv-python-headless numpy matplotlib scikit-image scikit-learn pandas tqdm

2. Run the Notebook: Open the assignment.ipynb file in Jupyter Notebook or Jupyter Lab and run the cells sequentially.

3. Data: Place your noisy images in the noisy_images folder.

Results

The results of the segmentation and evaluation are printed in tabular format and visualized using Matplotlib. The best filters and segmentation methods are determined based on the evaluation metrics.

Conclusion

This assignment demonstrates the application of various image processing techniques and their evaluation. By combining binary and multi-region segmentation methods, we can achieve more detailed segmentation results, especially in challenging conditions such as low-light night photos. The combination and refinement of different segmentation methods significantly improve the overall segmentation quality.