Dr Paul Hill

Machine Learning Lead Engineer


			        
                            

About

I am a Machine Learning Professional currently working as Senior Research Fellow at the University of Bristol. Also, I was previously at the University of Bristol (before a couple years in industry) researching into many different fields related to signal and image processing. I am also a lecturer delivering the whole of the masters course Speech and Audio Processing. I'm also a member of the Visual Information Laboratory and the Bristol Vision Institute, which are led by Prof. David Bull within the Department of Electrical and Electronic Engineering, University of Bristol.

Research Interests

  • Compression
  • Image Fusion
  • Machine Learning
  • Audio Processing
  • Remote Sensing
  • Signal Processing
  • Perceptual Processing
  • 2D Transforms

Book


News & Activities

  • Jul- 2020: Research Audience Immersion on Local Creative Cluster: CIC
  • Mar 2020: Supervision of Lines and Waves / Comms Labs
  • Feb-Apr 2020: Lecturing on Optimal Signal Processing
  • Jan- 2020: Local organising committee PCS 2021

Recent Projects

HABNet: Machine Learning, Remote Sensing Based Detection of Harmful Algal Blooms

This project is concerned with the application of machine learning techniques to develop a state-of-the-art detection and prediction system for spatiotemporal events found within remote sensing data; specifically, Harmful Algal Bloom events (HABs). https://github.com/csprh/HABNetDatacubeExtraction https://github.com/csprh/HABNetModelling

Automated Alert System for Volcanic Unrest

Satellite radar (InSAR) can be employed to observe volcanic ground deformation, which has shown a significant statistical link to eruptions. The explosion in data has however brought major challenges associated with timely dissemination of information and distinguishing volcano deformation patterns from noise, which currently relies on manual inspection. Here, we present a novel approach to detect volcanic ground deformation automatically from InSAR images.

Perceptual Image Fusion

The effective fusion of two or more visual sources can provide significant benefits for visualisation, scene understanding, target recognition and situational awareness in multi-sensor applications such as medicine, surveillance and remote sensing. Our work has focused on the use of image fusion within the wavelet domain leading to perceptually and region based fusion methods.

Wavelet Contrast Sensitivity

This project developed specific conventional contrast sensitivity functions (CSFs) for a variety of wavelet transforms (Conventional Wavelet, Dual Tree Complex Wavelet, Curvelet, and Steerable Pyramid Transforms). These functions have been subsequently used in our Perceptual Fusion work (linked above)


Papers

For all my papers see both google scholar