Back to search

Rochester Institute of Technology

An Analysis of multimodal sensor fusion for target detection in an urban environment

Abstract

dc:description.abstract

This work makes a compelling case for simulation as an attractive tool in designing cutting-edge remote sensing systems to generate the sheer volume of data required for a reasonable trade study. The generalized approach presented here allows multimodal system designers to tailor target and sensor parameters for their particular scenarios of interest via synthetic image generation tools, ensuring that resources are best allocated while sensors are still in the design phase. Additionally, sensor operators can use the customizable process showcased here to optimize image collection parameters for existing sensors. In the remote sensing community, polarimetric capabilities are often seen as a tool without a widely accepted mission. This study proposes incorporating a polarimetric and spectral sensor in a multimodal architecture to improve target detection performance in an urban environment. Two novel multimodal fusion algorithms are proposed--one for the pixel level, and another for the decision level. A synthetic urban scene is rendered for 355 unique combinations of illumination condition and sensor viewing geometry with the Digital Imaging and Remote Sensing Image Generation (DIRSIG) model, and then validated to ensure the presence of enough background clutter. The utility of polarimetric information is shown to vary with the sun-target-sensor geometry, and the decision fusion algorithm is shown to generally outperform the pixel fusion algorithm. The results essentially suggest that polarimetric information may be leveraged to restore the capabilities of a spectral sensor if forced to image under less than ideal circumstances.

Degree

thesis:*
Level thesis:degree_level
Dissertation
Discipline thesis:degree_discipline
Chester F. Carlson Center for Imaging Science (COS)

Author and committee

dc:creator, dc:contributor.*
Author dc:creator
  • Flusche, Brian M.
Contributors dc:contributor
  • Schott, John

Subjects

dc:subject × 4

Identifiers

dc:identifier.*
Repository record dc:identifier
https://repository.rit.edu/theses/2972

Chain of custody

source
Harvested from
Rochester Institute of Technology
Base URL
repository.rit.edu/do/oai/
Last updated
2026-11-19
Source record
OAI-PMH GetRecord
citation

Flusche, Brian M.. An Analysis of multimodal sensor fusion for target detection in an urban environment. Dissertation thesis, https://repository.rit.edu/theses/2972