Center for Automatic Target Recognition Research. Delivery Order 0005: Image Georegistration, Camera Calibration, and Dismount Categorization in Support of DEBU from Layered Sensing

Abstract

Aerial multi-head camera systems can provide a single synthetic image with large coverage from a set of images acquired simultaneously at particular time instant from each camera. To generate precise synthetic images, it is important to know the geometry between camera heads. In this project, DSM (Digital Surface Model) was generated from images acquired from aerial multi-head camera system. The process for generating DSM can be divided into two parts; Mosaic image generation and DSM generation. Relative position and orientation of physical cameras (six cameras) are given in terms of CAHVOR models. Mosaic images were generated projecting images acquired from physical cameras to a synthetic camera model via the reference plane. The synthetic camera for mosaic image is precisely designed to minimize resampling error. The EOP (exterior orientation parameters) of generated mosaic images can be calculated from navigation solutions which are given in pos file. 3D coordinates of a point can be calculated by space intersection of conjugate points of a pair of images with known IOP and EOP. Conjugate points can be automatically obtained by using image matching methods. Result of the image matching is 3D coordinates of a point. DSM can be generated by interpolating matching results.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Jul 01, 2011
Accession Number
ADA550532

Entities

People

  • Alper Yılmaz

Organizations

  • Ohio State University

Tags

Communities of Interest

  • Air Platforms
  • Materials and Manufacturing Processes

DTIC Thesaurus Topics

  • Air Force
  • Air Force Research Laboratories
  • Algorithms
  • Coordinate Systems
  • Data Sets
  • Evolutionary Algorithms
  • Genetic Algorithms
  • Geometry
  • Layered Sensing
  • Machine Learning
  • Measurement
  • Navigation
  • Pattern Recognition
  • Recognition
  • Supervised Machine Learning
  • Target Recognition
  • Three Dimensional

Readers

  • Computer Vision.

Technology Areas

  • Space