A Low-cost automatic compressing and noise-removing algorithm for satellite images.

Sreya (~PhiWhyyy)


1

Vote

Description:

Satellite imagery is a crucial source of data for numerous applications, from environmental monitoring to urban planning. However, these images face huge damage from noise, size and require accurate reconstruction. This work aims at exploring computationally intelligent techniques to make low cost algorithms for noise removal, denoising, and reconstructing satellite images using Python. We will focus on the efficacy of hybrid mathematical techniques in terms of image quality, compression ratio, and computational efficiency. This study also aims to provide insights into optimizing satellite image processing pipelines for better data storage. Here I am combining two very powerful numerical methods to make a hybrid algorithm. Singular Value Decomposition (SVD) is a great tool for denoising images because it can separate the noise from the underlying structure of the image. SVD might be a bit computationally expensive but measures can be taking to reduce the load and ease the code structure. after discarding the small, insignificant singular values and receiving the desired subbands. This denoised image is passed through Discrete Wavelet Transform to obtain the wavelet coefficient. quantifying the coefficient to achieve compression. Storing the significant coefficients and discarding the rest helps in have a good compression ratio. Future research can explore the integration of machine learning models to further enhance the denoising and compression processes. potentially leading to more adaptive and efficient satellite image processing system.

Prerequisites:

A basic idea on image processing and compression Linear Algebra Wavelet Transforms

Speaker Info:

I am Sreya Ghosh, an early career researcher, and I am aspiring to be a computer scientist. Out of my many interests, i love public astronomy and communicating science the most. I have been managing Phiwhyyy?!? to lucidly deliver the concepts of computational mathematics for STEM and non-STEM readers for the last 3 years and have recently started writing on Hashnode. More than a hobby, since 2016, I have been involved in Public Astronomy and have visited over 10 academic institutions, NGOs, and orphanages to share the joy of science across various parts of West Bengal, India.

I did my bachelors in mathematics with honours from Gokhale Memorial Girls College, affiliated with the University of Calcutta. I have done my masters in mathematics with a specialization in numerical methods from the Central University of Karnataka, Gulbarga, India.I try to actively work to convey gender equality through my blogs and outreach events, and I wish to do more in the coming years.

Speaker Links:

https://phiwhyyy.medium.com/ https://hashnode.com/@PhiWhyyy https://www.linkedin.com/in/sreyaghosh99/

Section: Python in Education and Research
Type: Talk
Target Audience: Beginner
Last Updated: