Road scene object detection using pre-trained RGB neural networks on linear Stokes images

Research output: Chapter in Book/Report/Conference proceedingConference contribution

2 Scopus citations

Abstract

Neural networks trained on RGB and monochromatic images are tested on images augmented by polarimetry for recognition of road-based objects. The goal of this work is to understand the scene conditions for which object detection and recognition can be improved by linear Stokes measurements. Shadows, windows, low albedo, and other object features which reduce RGB image contrast also decrease neural network detection performance. This work demonstrates specific cases for which linear Stokes images increase image contrast and therefore increase object detection by a neural network. Linear Stokes videos for five difference scenes are collected at three times of day and two driving directions. Although limited in scope, this work demonstrates some enhancement to object detection by adding polarimetry to neural networks trained on RGB images.

Original languageEnglish (US)
Title of host publicationPolarization
Subtitle of host publicationMeasurement, Analysis, and Remote Sensing XIV
EditorsDavid B. Chenault, Dennis H. Goldstein
PublisherSPIE
ISBN (Electronic)9781510636019
DOIs
StatePublished - 2020
EventPolarization: Measurement, Analysis, and Remote Sensing XIV 2020 - None, United States
Duration: Apr 27 2020May 8 2020

Publication series

NameProceedings of SPIE - The International Society for Optical Engineering
Volume11412
ISSN (Print)0277-786X
ISSN (Electronic)1996-756X

Conference

ConferencePolarization: Measurement, Analysis, and Remote Sensing XIV 2020
Country/TerritoryUnited States
CityNone
Period4/27/205/8/20

Keywords

  • Autonomous Driving
  • Image Fusion
  • Linear Stokes
  • Object Detection
  • Polarization
  • Road-Based Neural Networks

ASJC Scopus subject areas

  • Electronic, Optical and Magnetic Materials
  • Condensed Matter Physics
  • Computer Science Applications
  • Applied Mathematics
  • Electrical and Electronic Engineering

Fingerprint

Dive into the research topics of 'Road scene object detection using pre-trained RGB neural networks on linear Stokes images'. Together they form a unique fingerprint.

Cite this