Scale-invariant feature transform

from Wikipedia, the free encyclopedia

Scale-invariant feature transform ( English , " scale -invariant feature transformation ", SIFT for short ) is an algorithm for the detection and description of local features in images. The detector and the feature descriptions are, within certain limits, invariant to coordinate transformations such as translation, rotation and scaling. They are also robust against lighting variations, image noise and lower geometric deformation of a higher order, such as those caused by projective imaging of an object from different points of view in space. Originally developed for object recognition , the algorithm is now used in other areas of image processing . Including z. B. image registration or the creation of panorama images from individual images ( stitching ), object tracking in videos, navigation by robots or gesture recognition .

The algorithm was published in 1999 by David G. Lowe and is considered the reference method in its field of application. The University of British Columbia held a US patent on the algorithm . The basic concepts of SIFT can be found in various further developments by other authors. The calculation effort could be reduced and the robustness of the detector and feature descriptions improved (e.g. SURF , Affine-SIFT, PCA-SIFT).

literature

  • David G. Lowe: Object Recognition from Local Scale-Invariant Features . In: ICCV '99 Proceedings of the International Conference on Computer Vision . Volume 2, pages 1150–1157 ( online )
  • David G. Lowe: Distinctive Image Features from Scale-Invariant Keypoints In: International Journal of Computer Vision . Volume 60, No. 2, pages 91–110, 2004 ( online )
  • Matthew Brown, David Lowe: Automatic Panoramic Image Stitching using Invariant Features. In: International Journal of Computer Vision . Volume 74, No. 1, August 2007 ( online ; PDF; 3.7 MB)
  • Thomas Läb, Wolfgang Förstner : Automatic relative orientation of images. In: Proceedings of the 5th Turkish-German Joint Geodetic Days, March 29th - 31st, 2006 . Berlin, ISBN 3-9809030-4-4 ( online )
  • David G. Lowe: Method and Apparatus for Identifying Scale Invariant Features in an image and use of same for locating an object in an image Patent US 6,711,293 B1, published March 23, 2004

Web links

Individual evidence

  1. ↑ US 6711293 B1: Method and apparatus for Identifying scale invariant features in an image and use of same for locating of object in an image (Engl.)