Automatic Design and Synthesis of Automatic Target Recognition (ATR) Systems Using Learning Paradigms

Abstract

This report investigates evolutionary computational techniques such as genetic programming (GP), coevolutionary genetic programming (CGP), linear genetic programming (LGP) and genetic algorithms (GA) to automate the synthesis and analysis of object detection and recognition systems. It shows the efficacy of evolutionary computation in synthesizing effective composite operators and composite features from domain-independent primitive image processing operations and primitive features for object detection and recognition. Smart crossover, smart mutation and a new fitness function based on minimum description length (MDL) principle are designed to improve the efficiency of genetic programming. A new MDL-based fitness function is proposed to improve the genetic algorithm s performance on feature selection for object detection and recognition. Results are shown using MSTAR SAR imagery.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Oct 01, 2003
Accession Number
ADA424338

Entities

People

  • Bir Bhanu
  • Krzysztof Krawiec
  • Yingqiang Lin

Organizations

  • University of California, Riverside

Tags

Communities of Interest

  • Autonomy
  • Energy and Power Technologies
  • Human Systems
  • Sensors

DTIC Thesaurus Topics

  • Algorithms
  • Artificial Intelligence Software
  • Computer Programming
  • Computer Vision
  • Detection
  • Detectors
  • Evolutionary Algorithms
  • Genetic Algorithms
  • Image Processing
  • Information Processing
  • Information Science
  • Machine Learning
  • Pattern Recognition
  • Recognition
  • Software Development
  • Supervised Machine Learning
  • Target Recognition

Fields of Study

  • Computer science

Readers

  • Computer Vision.
  • Molecular and genetic basis of cancer.
  • Operations Research

Technology Areas

  • AI & ML
  • AI & ML - Machine Learning Algorithms
  • Biotechnology
  • Space
  • Space - Space Objects