Research

Research Interests

Machine Learning, Deep Learning, Pattern Recognition, Computational Intelligence, Computer Vision, Image Processing

Summary

I conduct research in the Machine Learning and Sensing Lab on seafloor segmentation using deep learning methodologies for the Office of Naval Research (ONR). During my PhD studies, I plan to develop and refine novel deep learning methods for texture characterization, segmentation and classification. These methods could then be applied toward automated image understanding, object detection and classification. Our goal on the application of these methods to sonar data is to develop an environmentally aware automatic target recognition system. The first step towards this goal is to segment the synthetic aperture sonar (SAS) images into different environments (seagrass, sand ripple, etc.) using the novel texture characterization approaches and then, develop target detectors that are optimized for each environment. My research can be applied to numerous difficult real-world applications affecting human life such as self-driving cars, autonomous robotics, augmented reality, automated medical diagnosis, and explosive hazard detection for the military.

Current Research Projects

Deep Learning for Texture Analysis

Seafloor Segmentation for Office of Naval Research (ONR)

Fractal Analysis of Root Architectures

Bias in Machine Learning