Intelligent surveillance system design based on independent component analysis and wireless sensor network

Long Xie, Masatoshi Ogawa, Youichi Kigawa, Harutoshi Ogai

Research output: Contribution to journalArticle

1 Citation (Scopus)


This paper explores the development of a real time intelligent surveillance system using the technology of pattern recognition based on independent component analysis (ICA) and a novel matching method as a reaction to perceptions of insecurity in sensitive spaces. An array of motion images of people are caught by micro digital cameras on board and transferred through wireless network to FPGA board. The feature points of the shot image and the image in database are extracted out using ICA algorithm in embedded PowerPC. The most similar images are picked up from the image database, which is classified to different clusters, and the potential insecurity level of invaders is detected. Furthermore the respective locations are connected by wireless network. The system of hardware and software co-design is implemented on Xilinx FPGA with the performance of high efficiency, low power consumption and easy integration with other devices.

Original languageEnglish
Pages (from-to)1543-1550+17
JournalIEEJ Transactions on Electronics, Information and Systems
Issue number12
Publication statusPublished - 2006 Jan 1



  • FPGA
  • ICA
  • Intelligent system
  • Pattern recognition
  • Wireless sensor network

ASJC Scopus subject areas

  • Electrical and Electronic Engineering

Cite this