Feature extraction assumes a number of forms in a number of applications. In this paper, we improve feature extraction by not only increasing the number of quality features that one can extract but also ensuring that the features we do extract are, indeed, representative high-quality features instead of false, minute, or noise features. We show that higher frequencies do not, for the purposes of feature extraction, necessarily represent human-salient features and that the combination of contrast enhancement, decimation, and lowpass filtering achieve more robust feature extraction than simple high-frequency boosting. Our ideal feature extractor therefore incorporates a decimator for reduction to an idealized size, contrast enhancement through stretched dynamic range, and frequency-domain filtering with a Gaussian lowpass filter.
[1]
G LoweDavid,et al.
Distinctive Image Features from Scale-Invariant Keypoints
,
2004
.
[2]
David G. Lowe,et al.
Object recognition from local scale-invariant features
,
1999,
Proceedings of the Seventh IEEE International Conference on Computer Vision.
[3]
Luc Van Gool,et al.
SURF: Speeded Up Robust Features
,
2006,
ECCV.
[4]
Rafael C. González,et al.
Digital image processing using MATLAB
,
2006
.
[5]
B. Wandell.
Foundations of vision
,
1995
.
[6]
Barry R. Masters,et al.
Digital Image Processing, Third Edition
,
2009
.