Title

Blob Feature Extraction for Event Detection Cameras

Date of Award

1-1-2020

Degree Name

M.S. in Electrical Engineering

Department

Department of Electrical and Computer Engineering

Advisor/Chair

Advisor: Keigo Hirakawa

Abstract

Neuromorphic (a.k.a. event detection) cameras emerged out of biologically inspired visual perception. A key component of neuromorphic cameras is the dynamic vision sensor or DVS, which generates an asynchronous data stream reporting temporal log-intensity changes (or "events") of the pixel-sized photodiodes. In this thesis a novel blob feature extraction technique for neuromorphic cameras is proposed. Using asynchronous 3D distance transform, we are able to track a blob, describe its size/shape/orientation, efficiently match the feature descriptors, and perform image correspondence across multiple neuromorphic cameras.

Keywords

Electrical Engineering

Rights Statement

Copyright 2020, author

Share

COinS