Metadata Learning of Non-Visual Features: Co-Occurrence Overlap Function for Rectangular Regions and Ground Truth Data

Abstract

To incorporate object locations in a multi-target detection model, we assume that a close duplicate cannot be learned by the model efficiently. So, we use a region-based approach which uses more object location compared to the ground truth locations to localize the targets. The proposed model is able to learn a similarity metric with respect to the ground truth locations which is robust (low false positives) enough forvarying images conditions, small aerial target sizes and using few training samples.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Apr 01, 2020
Accession Number
AD1097283

Entities

People

  • Asif Mehmood
  • Vasanth Iyer

Tags

Communities of Interest

  • Air Platforms
  • Autonomy
  • Space

DTIC Thesaurus Topics

  • Accuracy
  • Aerial Photography
  • Aerial Targets
  • Air Force
  • Air Force Research Laboratories
  • Aspect Ratio
  • Computer Vision
  • Deep Learning
  • Detection
  • Detectors
  • Learning
  • Machine Learning
  • Metadata
  • Satellite Imaging
  • Training
  • United States
  • Unmanned Vehicles

Fields of Study

  • Computer science

Readers

  • Computational Modeling and Simulation
  • Neural Network Machine Learning.
  • Sensor Fusion and Tracking Systems.