CPP_ImageProcessing_Library

This project is a C++ library for image processing that evolves through three versions: V1.0, V1.1, and V1.2. Each version introduces new features and improvements in image manipulation, filtering, and segmentation.

Project Structure

• TER_lib_V1.0/: Basic image processing functions using dynamic arrays (std::vector) for image representation.
• TER_lib_V1.1/: Introduction of object-oriented programming (OOP) with classes and templates for better flexibility and modularity.
• TER_lib_V1.2/: Advanced image processing techniques, including convolution filters, Sobel edge detection, multi-thresholding, Otsu's thresholding, and K-means clustering for image segmentation.

---

Version 1.0 - Basic Image Processing

This version focuses on basic image manipulation using dynamic arrays. It includes:

• Image creation: Functions to create white images, checkerboard patterns, and sinusoidal patterns.
• File I/O: Read and write raw images (.raw format), with support for big-endian and little-endian formats.
• Basic type conversions: Convert images between different types.
• LUT Application: Apply Look-Up Tables (LUTs) to grayscale images for false color representation.

---

Version 1.1 - Object-Oriented Programming (OOP)

This version refactors the code to use object-oriented programming principles, enhancing modularity and flexibility:

• Image class: Encapsulates image data and operations into a template class Image<T>, improving reusability and clarity.
• RGB Images: Support for RGB image manipulation with conversion from grayscale to RGB using LUTs.
• Overloaded operators: The () operator is overloaded to simplify pixel access.

---

Version 1.2 - Advanced Image Processing

In this version, advanced techniques are introduced:

• Convolution filters: Generic convolution filters to apply various effects like blurring or edge enhancement.
• Sobel edge detection: Detect edges in an image using the Sobel operator.
• Thresholding: Implement multi-thresholding and Otsu's thresholding methods for image segmentation.
• K-means clustering: Cluster pixel data into groups using K-means for image classification.
• Fourier transforms: Perform frequency domain filtering using Fast Fourier Transform (FFT).

---

How to Compile

Each version contains a Makefile that automates the compilation process. Navigate to the specific version folder (e.g., TER_lib_V1.0/) and run:

make

This will generate the executable, which can be run with:

./image_processor

Requirements

• C++17
• A compiler that supports C++17 (e.g., GCC, Clang).

Generated Images

Output images will be saved in the generated_images/ directory for each version. These images can be visualized using tools like Photopea, an online image editor.