Current Work

This project mainly focus towards creating a low power EEG amplifier which can be made in home at low cost. The main goal of this project is to enable long term brain monitoring which will enhance the screening and diagnostics of epilepsy in infants.

The currently available solutions for this task in the market work only on good conductive gel- electrode interfaces and are quite expensive. This will be an issue when it comes to long term monitoring, as the gel dries of the data collected will be have issues regarding the accuracy. Basically, the commercially available EEG amplifiers do not handle high impedance electrode to skin interfaces very well. Hence our project is aimed at solving this/these? issues by designing an amplifier to (complete the sentence).

Introduction

During the facial feature detection, I noticed that traditional methods were not working when trying to identify hair, and extraction of hair. The task was even harder when the picture quality was very poor (eg - CCTV footage ). Hence the Superpixel concept and GLOC model was implemented to extract hair region.

Overview

The conditional random field (CRF) is a powerful tool for building models to label segments in images. They are particularly appropriate for modeling local interactions among labels for regions (e.g., superpixels). Complementary to this, the restricted Boltzmann machine (RBM) has been used to model global shapes produced by segmentation models. Here, we present a new model that uses the combined power of these two types of networks to build a state-of-the-art labeler(check spellings), and demonstrate its labeling performance for the parts of complex face images. Specifically, we address the problem of labeling the Labeled Faces in the Wild data set into hair, skin and background regions. The CRF is a good baseline labeler, but we show how an RBM can be added to the architecture to provide a global shape bias that complements the local modeling provided by the CRF. This hybrid model produces results that are both quantitatively and qualitatively better than the CRF alone. In addition to high quality segmentation results, we demonstrate that the hidden units in the RBM portion of our model can be interpreted as face attributes which have been learned without any attribute-specific training data.

IFEC Introduction

IFEC is an international student competition for innovation, conservation, and effective use of electrical energy, which is open to college and university student teams from recognized undergraduate engineering programs in any location.

The competition is sponsored by the Institute of Electrical and Electronics Engineers (IEEE) Power Electronics Society (PELS), Power & Energy Society (PES), Industry Application Society (IAS) and Power Sources Manufacturers Association (PSMA).

Topic: Power Supply for Nano Satellites

Preliminary Specifications:

- 2x DC inputs: Max power < 60 W, Max current < 3 A, voltage 20 V

- 3x DC outputs: 3.3 V @ 5 A, 5 V @ 4 A, and 8 V @ 2600 mAh

- Functionalities: MPPT control (uniform and non-uniform solar irradiance)

- Functionalities: Battery charging and discharging

- Judgment criteria: Functionality, efficiency, power density, and cost.

Final Competition Location : Aalborg University, Aalborg, Denmark