Skip to content

SOCIS 2017 Ian Harper

Nabil Freij edited this page Feb 22, 2024 · 3 revisions

Profile

Bio

I'm a final year theoretical physics student at Trinity College Dublin. I have just completed my final year. I have worked extensively in Python, particularly in the packages Numpy, Scipy and Matplotlib. I have worked with image reconstruction in the past using singular value decomposition to deconstruct and re-assemble an image to take up less space with minimal loss of image quality. I've worked in other languages in the past such as C++ and Processing.js, and I have knowledge of Java, Javascript and HTML. During the summer of 2016 I worked on a physics project at University College Dublin called Social Physics: A Model of Opinion Dynamics. I used processing.js to write a web applet to simulate the effects of opinion dynamics. I also used Matlab to run simulations of a fortran source file to extrapolate data from the opinion system

Proposal

I want to work on developing the sunkit-image package. I hope to achieve the following goals:

  1. Port the Multi-Scale Gaussian Normalisation code from #1899.
  2. Convert the differential rotation code in SunPy to use sunpy.coordinates.
  3. Implement image warping for solar differential rotation. #1876.
  4. Implement the OCCULT-2 algorithm for coronal loop tracing.
  5. Implement running and base difference functionality and the persistence transform. And if time permits, I want to
  6. Refactor and write a Python wrapper for FLCT code.
  7. Implement image alignment using feature detection and tracking. Example

Timeline

  • Week 1: Study the image processing features of sunpy and plan how to implement the relevant code
  • Week 2/3: Study the MGN code and code it.
  • Week 4/5: Study solar differential rotation and the differential rotation code. Change to sunpy coordinates.
  • Week 6: Start Implementing the image warping algorithm for solar differential rotation. Test and review.
  • Week 7/8: Study coronal loop tracing and implement the OCCULT-2 algorithm
  • Week 9/10: Study how to calculate running and base difference functionality and implement it with the persistence transform.
  • Week 11: Test, debug, clean and optimise code, along with documentation.
  • Week 12: Work on refactoring and writing the Python wrapper. Try implement image alignment.

Extra info

  • Availability: From 9:00 to 23:00. (UTC-4).
Clone this wiki locally