• DocumentCode
    3350408
  • Title

    High quality panoramic image generation using multiple panoramic annular lens images

  • Author

    Araki, Satoshi ; Maeda, Kei ; Shibata, Keiji ; Horita, Yuukou

  • Author_Institution
    Grad. Sch. of Sci. & Eng., Univ. of Toyama, Toyama, Japan
  • fYear
    2010
  • fDate
    26-29 Sept. 2010
  • Firstpage
    1213
  • Lastpage
    1216
  • Abstract
    Surveillance with intelligent camera systems is widely used in traffic control, driving assistance and surgery. One of the most essential specifications for surveillance systems is compactness. Therefore, in these applications with large field of view, we can avoid the use of multiple camera systems or mechanically controlled cameras. It is well known that omnidirectional camera systems using panoramic annular lens (PAL) have many advantages. Nevertheless, because of low spatial resolution of angle of elevation for the vertical direction of captured image by using PAL, the image after panoramic expansion usually has a disadvantage such as image quality degradation. To overcome this disadvantage, high quality panoramic image generation method is essential. Consequently, we propose the high quality panoramic image generation method using multiple PAL images. By synthesizing multiple PAL images for different zoom ratio, we can obtain fine quality panoramic image that improve image quality deterioration.
  • Keywords
    image resolution; lenses; surveillance; PAL image; compactness; high quality panoramic image generation; image quality degradation; image quality deterioration; intelligent camera system; low spatial resolution; omnidirectional camera system; panoramic annular lens image; panoramic expansion; surveillance system; zoom ratio; Cameras; Integrated optics; Lenses; Optical imaging; Pixel; Spatial resolution; expansion; high-quality; multiple image; omnidirectional camera system; panorama;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Image Processing (ICIP), 2010 17th IEEE International Conference on
  • Conference_Location
    Hong Kong
  • ISSN
    1522-4880
  • Print_ISBN
    978-1-4244-7992-4
  • Electronic_ISBN
    1522-4880
  • Type

    conf

  • DOI
    10.1109/ICIP.2010.5652454
  • Filename
    5652454