Semantic Segmentation for Aerial Imagery Using U-Nets

Abstract

In situations where global positioning systems are unavailable, alternative methods of localization must be implemented. A potential step to achieving this is semantic segmentation, or the ability for a model to output class labels by pixel. This research aims to utilize datasets of varying spatial resolutions and locations to train a fully convolutional neural network architecture called the U-Net to perform segmentations of aerial images. Variations of the U-Net architecture are implemented and compared to other existing models in order to determine the best in detecting buildings and roads. A final dataset will also be created combining two datasets to determine the ability of the U-Net to segment classes regardless of location. The final segmentation results will demonstrate the overall efficacy of semantic segmentation for different datasets for potential localization applications.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Mar 19, 2020
Accession Number
AD1104208

Entities

People

  • Terrence J. Yi

Organizations

  • Air Force Institute of Technology

Tags

Communities of Interest

  • Space

DTIC Thesaurus Topics

  • Air Force
  • Artificial Intelligence
  • Artificial Intelligence Software
  • Computer Vision
  • Computers
  • Convolutional Neural Networks
  • Data Processing
  • Deep Learning
  • Global Navigation Satellite Systems
  • Global Positioning Systems
  • Image Classification
  • Machine Learning
  • Navigation
  • Neural Networks
  • Two Dimensional
  • United States
  • United States Government

Readers

  • Computer Vision.

Technology Areas

  • AI & ML
  • AI & ML - Neural Networks
  • Space