This page displays some of the projects I worked on as a part of my academic interests and also some of the summer and winter schools I attended. These projects and summer and winter schools have been stepping stones in my career as they imparted several multi-disciplinary skills in Electrical Engineering, Computer Sciences and Physics. I have written a brief description of each project which can be found by clicking the titles of the project.
- Using Deep Learning methods to automatically detect artifacts in Polarimetric Images (May 2019 to August 2019):
I am currently working as an intern at the California Institute of Technology in Pasadena on using Convolutional Neural Networks to automatically detect artifacts in astronomical images obtained using an imaging polarimeter. At Caltech I am working with Dr. Ashish Mahabal (Computational Scientist at the Center for Data Driven Discovery), Dr. Gina Panapoulou, Prof. Kieran Cleary and Prof. Tony Readhead. We presented a poster on our work so far at the Astroinformatics conference in June 2019. The poster can be found ere: Poster on using Deep Learning Methods for detecting artifacts in Polarimetric Images
On June 24th 2019, I also gave an online seminar on my work at a conference happening in Oslo, Norway and its video is available on YouTube at this link: YouTube link of my talk at the 3rd PASIPHAE meeting in Oslo, Norway.
A paper on the same is being written and will be submitted to the Monthly Notices of the Royal Astronomical Society very soon.
In Winter 2019, I took the EECS 598 Special Topics course on Deep Learning taught by Prof. Honglak Lee at the University of Michigan. I worked on a project during the course to use Graph Based and Time Series based classification models to classify the brain scans (MEG data) of Musicians and Non-Musicians taken in resting state. This is the first known attempt at doing so and I am writing a paper with Dr. Kanad Mandke, a research associate at the University of Cambridge on this work. We achieved excellent training accuracies and testing accuracies and also found the regions in the brain responsible for music actions. The image below shows the regions in red which more found to have the most significant classification features.
During Fall 2018, I worked on a team project in the EECS 504 (Computer Vision course) to automatically identify parking spots using Computer Vision and Machine Learning Techniques. The project was inspired by an idea to develop a live app showing the locations of available parking spots in your location so that you don’t spend time searching for one. This idea is also useful for autonomous vehicles.
In December 2017 and January 2018 I further worked with Prof. Anthony Readhead at the California Institute of Technology in USA on developing a lab for the WALOP project at Caltech. From instrument design testing to developing Python scripts to automatically analyze the lab data, we worked on testing the accuracy of the instrument design to measure the degree of linear polarization of light and angle of linear polarization.
Between December 2016 and June 2017, I worked on the characterization of the readout electronics and noise characterization of a Charge Coupled Device (CCD) controller at IUCAA, Pune. I co-authored a paper on the same which can be found at this link.