• DocumentCode
    1790201
  • Title

    Lens distortion correction and geometrical alignment for Around View Monitoring system

  • Author

    Santhanam, S.M. ; Balisavira, V. ; Roh, S.H. ; Pandey, V.K.

  • Author_Institution
    Mando Softtech India Pvt. Ltd., Bangalore, India
  • fYear
    2014
  • fDate
    22-25 June 2014
  • Firstpage
    1
  • Lastpage
    2
  • Abstract
    Around View Monitoring (AVM) system assists drivers by providing a more natural perspective of scenes (around the vehicle) to the driver, thus enhancing road safety. The AVM system usually employs several ultra-wide angle lens cameras to cover a large area around the vehicle. These camera lenses are usually spherical in nature and they suffer from severe distortions. Conventional distortion correction techniques have their own limitations and hence, a novel distortion correction technique is proposed to correct the distortion introduced by such lenses. To generate the bird´s eye view of the surrounding environment, a homography based geometrical alignment technique is also proposed. A unique pattern is used in conjunction with a feature based refinement technique to align the multi-camera views precisely, and to generate an accurate bird´s eye view of the surroundings. Further, colour temperature balancing and blending techniques are applied to enhance the appearance of the bird´s eye view.
  • Keywords
    cameras; geometrical optics; image sensors; lenses; optical distortion; road safety; AVM; around view monitoring system; blending techniques; camera lenses; colour temperature balancing; feature-based refinement technique; geometrical alignment; homography; lens distortion correction; multicamera views; road safety; Calibration; Cameras; Image color analysis; Lenses; Monitoring; Transmission line matrix methods; Vehicles; Bird´s eye view; automotive; geometrical alignment;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Consumer Electronics (ISCE 2014), The 18th IEEE International Symposium on
  • Conference_Location
    JeJu Island
  • Type

    conf

  • DOI
    10.1109/ISCE.2014.6884316
  • Filename
    6884316