A SIFT and SURF Comparison

• Elena I. Kostyukova

  English

  Author

• Youngsighn

  English

  Author

In many applications, accurate, reliable, and automatic image registration is a crucial task.
Feature detection, feature matching, derivation of transformation function based on related features in
images, and reconstruction of images based on derived transformation function are necessary phases
in image registration/alignment. Accurate feature detection and matching are necessary for a registered
image to be accurate. Therefore, in many picture applications, such as image registration, computer
vision, image mosaic, etc., these two intermediate processes are crucial. Scale Invariant Feature
Transform (SIFT) and Speed Up Robust Features (SURF) are two distinct approaches for scale and
rotation invariant interest point/feature detector and descriptor that are presented in this study.
Additionally, it offers a method for obtaining unique invariant features from pictures that can be
utilized to reliably match various perspectives of a scene or item.

• pdf