Display and Enhancement of Infrared Images

Abstract

We review how IR images differ from visible images, emphasize the importance of using a variety of IR images for assessing the utility of various algorithms for displaying them, and develop a gamma function suitable for 8-bit display on Sun workstation monitors. For globally monotonic mapping of wide dynamic range IR imagery to 8 bits for display, we describe and compare direct scaling, histogram equalization, histogram projection, and several hybrid algorithms, including undersampled projection and plateau equalization. For locally adaptive contrast enhancement, which is not globally monotonic, we describe local histogram projection, modular techniques (sawtoothing), and high frequency sharpening in the spatial and spatial frequency domains. We stress tradeoffs among these algorithms as to degree of enhancement, artifacts, and computational costs. All topics are illustrated by an extensive set of IR images taken with state-of-the-art PtSi Schottky barrier cameras. We also briefly address issues germane to real-time implementation.... Histogram projection, Infrared imagery, Local contrast enhancement, PtSi Cameras, Histogram equalization.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Sep 01, 1992
Accession Number
ADA264280

Entities

People

  • Jerry Silverman
  • Virgil E. Vickers

Organizations

  • Rome Laboratory

Tags

Communities of Interest

  • Advanced Electronics
  • Air Platforms
  • Biomedical
  • Sensors

DTIC Thesaurus Topics

  • Abstracts
  • Aircrafts
  • Algorithms
  • Artifacts
  • Cameras
  • Contrast
  • Detectors
  • Digital Images
  • Dynamic Range
  • Equalization
  • Frequency
  • Gray Scale
  • Image Processing
  • Images
  • Infrared Images
  • Optical Phenomena
  • Signal Processing

Readers

  • Computer Vision.
  • Image Processing and Computer Vision.