This repository contains the 8-week project offered by Stamatics Society at IIT Kanpur.
The project covers various topics in image processing, including sampling and quantization, image enhancement, spatial domain analysis, frequency domain analysis, edge detection, and image segmentation using ML algorithms.
Assignment 1 focused on basic Probability concepts as they are useful in image processing.
In this assignment, we worked on solving jigsaw puzzles using image processing techniques. The input image, 'jigsaw.jpg,' contained misaligned portions and sections. By rearranging the sub-matrices of the image, we solved the puzzle. The Google Colab file IP2_A2_Q1 contains the code and outputs the solved image.
We implemented a dot matrix display for numbers using circular discs. Each dot represented a pixel with a radius of 25 pixels. The dot matrix image had a size of 300x500 pixels and was grayscale. The Google Colab file IP2_A2_Q2 takes a 6-digit number as input and outputs the dot matrix representation of the last two digits.
In this assignment, we worked on cleaning the gutters of scanned books, magazines, or notebooks. Shadows in the gutter were removed using image processing techniques. The file IP_A3_Q1 takes the input as an integer (1, 2, or 3) corresponding to the provided images 'gutters1.jpg', 'gutters2.jpg', or 'gutters3.jpg' and outputs the cleaned image.
For this task, we enhanced night images captured by battery-powered CCTV cameras in remote villages. The goal was to discover possible traces of theft by processing and enhancing the images. The file IP_A3_Q2 takes the input as an integer (1, 2, 3, or 4) corresponding to the provided CCTV images 'cctv1.jpg', 'cctv2.jpg', 'cctv3.jpg', or 'cctv4.jpg' and outputs the respective enhanced images.
If you have any further questions or need additional information, please feel free to reach out!