DocumentCode :
1422236
Title :
Scale- and Affine-Invariant Fan Feature
Author :
Cui, Chunhui ; Ngan, King Ngi
Author_Institution :
Dept. of Electron. Eng., Chinese Univ. of Hong Kong, Shatin, China
Volume :
20
Issue :
6
fYear :
2011
fDate :
6/1/2011 12:00:00 AM
Firstpage :
1627
Lastpage :
1640
Abstract :
Most existing feature detectors assume no surface discontinuity within the keypoints´ support regions and, hence, have little chance to match the keypoints located on or near the surface boundaries. These keypoints, though not many, are salient and representative. In this paper, we show that they can be successfully matched by using the proposed scale- and affine-invariant Fan features. Specifically, the image neighborhood of a keypoint is depicted by multiple fan subregions, namely Fan features, to provide robustness to surface discontinuity and background change. These Fan features are made scale-invariant by using the automatic scale selection method based on the Fan Laplacian of Gaussian (FLOG). Affine invariance is further introduced to the Fan features based on the affine shape diagnosis of the mirror-predicted surface patch. The Fan features are then described by Fan-SIFT, which is an extension of the famous scale-invariant feature transform (SIFT) descriptor. Experimental results of quantitative comparisons show that the proposed Fan feature has good repeatability that is comparable to the state-of-the-art features for general structured scenes. Moreover, by using Fan features, we can successfully match image structures near surface discontinuities despite significant scale, viewpoint, and background changes. These structures are complementary to those found by the traditional methods and are especially useful for describing weakly textured scenes, which is demonstrated in our experiments on image matching and object rendering.
Keywords :
Gaussian processes; Laplace transforms; affine transforms; feature extraction; image matching; affine-invariant fan feature; afflne shape diagnosis; automatic scale selection method; fan Laplacian of Gaussian; feature detection; image matching; image neighborhood; mirror-predicted surface patch; scale-invariant fan feature transform descriptor; surface discontinuity; Clutter; Detectors; Feature extraction; Image edge detection; Kernel; Shape; Three dimensional displays; Affine invariance; background invariance; feature description; feature detection; scale invariance; viewpoint invariance; Algorithms; Image Enhancement; Image Interpretation, Computer-Assisted; Pattern Recognition, Automated; Reproducibility of Results; Sensitivity and Specificity; Subtraction Technique;
fLanguage :
English
Journal_Title :
Image Processing, IEEE Transactions on
Publisher :
ieee
ISSN :
1057-7149
Type :
jour
DOI :
10.1109/TIP.2010.2103948
Filename :
5682405
Link To Document :
بازگشت