Open Access
Review
Table 1
Advantages and disadvantages of feature point detection algorithms.
| Feature point detection algorithm | References | Advantages | Disadvantages |
|---|---|---|---|
| Harris | Moravec [51], Harris and Stephens [52], Harris [53] | Rotational invariance, luminance invariance | Lack of scale invariant properties and affine invariance properties |
| SIFT | Lowe [54, 57], Brown and Lowe [56], Cruz-Mota et al. [61], Al-khafaji et al. [63], Li and Yuan [64] | Rotational invariance, scale invariance, luminance invariance, good robustness | The calculated dimension is too large and the operation speed is slow |
| Harris-Laplacian | Mikolajczyk and Schmid [55] | Rotational invariance, scale invariance | Redundancy point, anti-interference ability is not strong, real-time performance is poor |
| SURF | Bay et al. [58] | Compared with SIFT, the calculation is smaller and the speed is faster, Rotational invariance, scale invariance, good robustness | The computational speed is an order of magnitude faster than SIFT and an order of magnitude slower than ORB |
| FAST | Rosten and Drummond [59] | Fast operation speed | Easy to be affected by noise, poor robustness, Lack of rotation invariance |
| ORB | Rublee et al. [60] | Fast operation speed, Rotational invariance | Lack of scale invariant properties |
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.