Distinctive Image Features from Scale-Invariant Keypoints
Citations
332 citations
Cites methods from "Distinctive Image Features from Sca..."
...However, this system has the following limitations: 1) SIFT features cannot be consistently detected in latents and full prints, 2) the minutiae extractor and matcher (VeriFinger 4.2) used in [16] are not suitable for latent palmprint matching, and 3) latent images were not used to evaluate the algorithms....
[...]
...The first stage of SIFT algorithm, as outlined in [18], constructs the scale space by obtaining Difference of Gaussian (DOG) images at various scales....
[...]
...texture) information, to improve the identification performance, an additional feature set - Scale Invariant Feature Transformation (SIFT) [18] - utilizing the texture information is used in conjunction with minutiae....
[...]
...In our experiments, we utilized the SIFT implementation by Lowe in [18], where the parameters, number of octaves, number of scale samples per octave are set to 5 and 3, respectively....
[...]
...A partial-to-full palmprint matching system was proposed in [16] that used both SIFT [17] and minutiae features in matching....
[...]
331 citations
Cites methods from "Distinctive Image Features from Sca..."
...In all the experiments, we used two types of low level descriptors: SIFT-like gradient orientation histograms [20] and simple local color statistics (RGB mean and variations)....
[...]
329 citations
328 citations
Cites background or methods from "Distinctive Image Features from Sca..."
...Lowe [19] proposed the SIFT for object detection based on its template image....
[...]
...SIFT is a local descriptor extraction method having valuable properties such as invariance to illumination and viewpoint [19]....
[...]
...More details on SIFT can be found in [19]....
[...]
327 citations
References
46,906 citations
16,989 citations
13,993 citations
7,057 citations
3,422 citations