Skip to main navigation Skip to search Skip to main content

Fast detection and localization algorithm for circular targets

  • Tsinghua University

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

Equipped with the mini camera, UAVs can be used as a surveillance device, which can be widely used in both civil fields as well as military fields. The real time video output by camera can be transferred down to the ground station and is processed to detect and localize the interested targets. A low-cost and real time circular targets detection and localization algorithm is proposed in this paper. The target is detected by image analysis based recognition method and the position is represented in UAV's coordinate frame. Further, the position is transferred to the world coordinate frame by using UAV's guidance data (position and attitude) to localize the targets. Experiment results are also given and show that the algorithm can be run in real time.

Original languageEnglish
Title of host publicationSeventh International Symposium on Instrumentation and Control Technology
Subtitle of host publicationSensors and Instruments, Computer Simulation, and Artificial Intelligence
DOIs
StatePublished - 2008
Event7th International Symposium on Instrumentation and Control Technology: Sensors and Instruments, Computer Simulation, and Artificial Intelligence - Beijing, China
Duration: 10 Oct 2008 → 13 Oct 2008

Publication series

NameProceedings of SPIE - The International Society for Optical Engineering
Volume7127
ISSN (Print)0277-786X

Conference

Conference7th International Symposium on Instrumentation and Control Technology: Sensors and Instruments, Computer Simulation, and Artificial Intelligence
Country/TerritoryChina
CityBeijing
Period10/10/08 → 13/10/08

Keywords

  • Image processing
  • Real time detection
  • Target detection
  • Target localization

Fingerprint

Dive into the research topics of 'Fast detection and localization algorithm for circular targets'. Together they form a unique fingerprint.

Cite this