DocumentCode :
64677
Title :
Spatial Statistics of Image Features for Performance Comparison
Author :
Bostanci, E. ; Kanwal, Navdeep ; Clark, Adrian F.
Author_Institution :
Sch. of Comput. Sci. & Electron. Eng., Univ. of Essex, Colchester, UK
Volume :
23
Issue :
1
fYear :
2014
fDate :
Jan. 2014
Firstpage :
153
Lastpage :
162
Abstract :
When matching images for applications such as mosaicking and homography estimation, the distribution of features across the overlap region affects the accuracy of the result. This paper uses the spatial statistics of these features, measured by Ripley´s K-function, to assess whether feature matches are clustered together or spread around the overlap region. A comparison of the performances of a dozen state-of-the-art feature detectors is then carried out using analysis of variance and a large image database. Results show that SFOP introduces significantly less aggregation than the other detectors tested. When the detectors are rank-ordered by this performance measure, the order is broadly similar to those obtained by other means, suggesting that the ordering reflects genuine performance differences. Experiments on stitching images into mosaics confirm that better coverage values yield better quality outputs.
Keywords :
feature extraction; image matching; image segmentation; statistical analysis; visual databases; Ripley K-function; feature detectors; homography estimation; image database; image features; image matching; mosaicking estimation; performance comparison; spatial statistics; Algorithm design and analysis; Detectors; Feature extraction; Image processing; Performance evaluation; Robustness; Spatial analysis; Spatial statistics; evaluation; image feature coverage;
fLanguage :
English
Journal_Title :
Image Processing, IEEE Transactions on
Publisher :
ieee
ISSN :
1057-7149
Type :
jour
DOI :
10.1109/TIP.2013.2286907
Filename :
6645433
Link To Document :
بازگشت