IEEE Geoscience and Remote Sensing Magazine - December 2015 - 26

Table 1. Confusion maTrix and Common performanCe meTriCs CalCulaTed from roof maTChing.
CorreCT maTChes

CorreCT non-maTChes prediCTed posiTive/negaTive

Predicted matches

TP = 43

FP = 0

P' = 43

Predicted non-matches

FN = 7

TN = 12

N' = 19

P = 50

N = 12

Total = 62

TPR = TP/P = 0.86

FPR = FP/N = 0

Actual positive/negative

PPV = TP/ P' = 1

ACC = (TP + TN)/(P+N) = 0.89

FIgUre 8. The examples of matching failure. The first and second is limited of inconsistent contour detection; The third results from the

inconsistency of classidication, and the forth from too many broken blocks making the Euler Number uncertain.
of incorrect matching. The missing match is derived from
contour deviation or classification uncertainty (Fig. 8).
3.3 Image mosaIc Instance
After regional feature matching, a series of applications
can be performed. In this paper, image mosaic is taken
as an instance. The centroids of the matched roofs, eliminated as outliers by RANSAC algorithm, serve as the tiepoints. Fig. 9 shows the conditions of tie-points and the
ultimate mosaicked image. The images are mosaicked with
a less obvious seam line.
3.4 comparatIve study
The repetitive and similar building roofs bring the biggest challenge for one-to-one correspondence of feature
description. The proposed method, as a feature-level
matching, is expected to exclude incorrect matches
completely so as to be a basis in image mosaic, image
retrieval, and stereo matching.
Image matching is relatively mature in natural image processing. SURF [20] is a robust feature detection method and

matches the keypoints with
minimum Euclidean distance
The prOpOSed apprOaCh
for the invariant descriptor
CaN aLSO be appLIed FOr
vector. We tested the repetitive
Image reTrIeVaL, Image
parts as the green box in Fig.
4. The result (Fig. 10) shows
mOSaIC, aNd geOmeTrIC
obvious mismatch. FLANN
meaSUremeNT IN COmpLex
[21] is a library for performUrbaN SCeNeS.
ing fast approximate nearest
neighbor searches in high
dimensional spaces. FLANN
gains 20 matching points without scattered distribution in
the test area (Fig. 11), whereas 43 tie-points, obtained by roofdetection based method, are nearly randomly distributed in
the overlapping area. Obviously, it is more feasible to obtain
scatted tie-points so as to construct optimal transformation
matrix in image matching applications.
IV. CONCLUSION
A fast and automatic image matching method is presented
for  urban aerial images. Roof features are produced by

FIgUre 9. The tie-points linked by lines consisting of centroids (left) and the mosaicked image based on the tie-points (right).

26

ieee Geoscience and remote sensing magazine

december 2015



Table of Contents for the Digital Edition of IEEE Geoscience and Remote Sensing Magazine - December 2015

IEEE Geoscience and Remote Sensing Magazine - December 2015 - Cover1
IEEE Geoscience and Remote Sensing Magazine - December 2015 - Cover2
IEEE Geoscience and Remote Sensing Magazine - December 2015 - 1
IEEE Geoscience and Remote Sensing Magazine - December 2015 - 2
IEEE Geoscience and Remote Sensing Magazine - December 2015 - 3
IEEE Geoscience and Remote Sensing Magazine - December 2015 - 4
IEEE Geoscience and Remote Sensing Magazine - December 2015 - 5
IEEE Geoscience and Remote Sensing Magazine - December 2015 - 6
IEEE Geoscience and Remote Sensing Magazine - December 2015 - 7
IEEE Geoscience and Remote Sensing Magazine - December 2015 - 8
IEEE Geoscience and Remote Sensing Magazine - December 2015 - 9
IEEE Geoscience and Remote Sensing Magazine - December 2015 - 10
IEEE Geoscience and Remote Sensing Magazine - December 2015 - 11
IEEE Geoscience and Remote Sensing Magazine - December 2015 - 12
IEEE Geoscience and Remote Sensing Magazine - December 2015 - 13
IEEE Geoscience and Remote Sensing Magazine - December 2015 - 14
IEEE Geoscience and Remote Sensing Magazine - December 2015 - 15
IEEE Geoscience and Remote Sensing Magazine - December 2015 - 16
IEEE Geoscience and Remote Sensing Magazine - December 2015 - 17
IEEE Geoscience and Remote Sensing Magazine - December 2015 - 18
IEEE Geoscience and Remote Sensing Magazine - December 2015 - 19
IEEE Geoscience and Remote Sensing Magazine - December 2015 - 20
IEEE Geoscience and Remote Sensing Magazine - December 2015 - 21
IEEE Geoscience and Remote Sensing Magazine - December 2015 - 22
IEEE Geoscience and Remote Sensing Magazine - December 2015 - 23
IEEE Geoscience and Remote Sensing Magazine - December 2015 - 24
IEEE Geoscience and Remote Sensing Magazine - December 2015 - 25
IEEE Geoscience and Remote Sensing Magazine - December 2015 - 26
IEEE Geoscience and Remote Sensing Magazine - December 2015 - 27
IEEE Geoscience and Remote Sensing Magazine - December 2015 - 28
IEEE Geoscience and Remote Sensing Magazine - December 2015 - 29
IEEE Geoscience and Remote Sensing Magazine - December 2015 - 30
IEEE Geoscience and Remote Sensing Magazine - December 2015 - 31
IEEE Geoscience and Remote Sensing Magazine - December 2015 - 32
IEEE Geoscience and Remote Sensing Magazine - December 2015 - 33
IEEE Geoscience and Remote Sensing Magazine - December 2015 - 34
IEEE Geoscience and Remote Sensing Magazine - December 2015 - 35
IEEE Geoscience and Remote Sensing Magazine - December 2015 - 36
IEEE Geoscience and Remote Sensing Magazine - December 2015 - 37
IEEE Geoscience and Remote Sensing Magazine - December 2015 - 38
IEEE Geoscience and Remote Sensing Magazine - December 2015 - 39
IEEE Geoscience and Remote Sensing Magazine - December 2015 - 40
IEEE Geoscience and Remote Sensing Magazine - December 2015 - 41
IEEE Geoscience and Remote Sensing Magazine - December 2015 - 42
IEEE Geoscience and Remote Sensing Magazine - December 2015 - 43
IEEE Geoscience and Remote Sensing Magazine - December 2015 - 44
IEEE Geoscience and Remote Sensing Magazine - December 2015 - 45
IEEE Geoscience and Remote Sensing Magazine - December 2015 - 46
IEEE Geoscience and Remote Sensing Magazine - December 2015 - 47
IEEE Geoscience and Remote Sensing Magazine - December 2015 - 48
IEEE Geoscience and Remote Sensing Magazine - December 2015 - 49
IEEE Geoscience and Remote Sensing Magazine - December 2015 - 50
IEEE Geoscience and Remote Sensing Magazine - December 2015 - 51
IEEE Geoscience and Remote Sensing Magazine - December 2015 - 52
IEEE Geoscience and Remote Sensing Magazine - December 2015 - 53
IEEE Geoscience and Remote Sensing Magazine - December 2015 - 54
IEEE Geoscience and Remote Sensing Magazine - December 2015 - 55
IEEE Geoscience and Remote Sensing Magazine - December 2015 - 56
IEEE Geoscience and Remote Sensing Magazine - December 2015 - 57
IEEE Geoscience and Remote Sensing Magazine - December 2015 - 58
IEEE Geoscience and Remote Sensing Magazine - December 2015 - 59
IEEE Geoscience and Remote Sensing Magazine - December 2015 - 60
IEEE Geoscience and Remote Sensing Magazine - December 2015 - 61
IEEE Geoscience and Remote Sensing Magazine - December 2015 - 62
IEEE Geoscience and Remote Sensing Magazine - December 2015 - Cover3
IEEE Geoscience and Remote Sensing Magazine - December 2015 - Cover4
https://www.nxtbook.com/nxtbooks/ieee/geoscience_december2023
https://www.nxtbook.com/nxtbooks/ieee/geoscience_september2023
https://www.nxtbook.com/nxtbooks/ieee/geoscience_june2023
https://www.nxtbook.com/nxtbooks/ieee/geoscience_march2023
https://www.nxtbook.com/nxtbooks/ieee/geoscience_december2022
https://www.nxtbook.com/nxtbooks/ieee/geoscience_september2022
https://www.nxtbook.com/nxtbooks/ieee/geoscience_june2022
https://www.nxtbook.com/nxtbooks/ieee/geoscience_march2022
https://www.nxtbook.com/nxtbooks/ieee/geoscience_december2021
https://www.nxtbook.com/nxtbooks/ieee/geoscience_september2021
https://www.nxtbook.com/nxtbooks/ieee/geoscience_june2021
https://www.nxtbook.com/nxtbooks/ieee/geoscience_march2021
https://www.nxtbook.com/nxtbooks/ieee/geoscience_december2020
https://www.nxtbook.com/nxtbooks/ieee/geoscience_september2020
https://www.nxtbook.com/nxtbooks/ieee/geoscience_june2020
https://www.nxtbook.com/nxtbooks/ieee/geoscience_march2020
https://www.nxtbook.com/nxtbooks/ieee/geoscience_december2019
https://www.nxtbook.com/nxtbooks/ieee/geoscience_september2019
https://www.nxtbook.com/nxtbooks/ieee/geoscience_june2019
https://www.nxtbook.com/nxtbooks/ieee/geoscience_march2019
https://www.nxtbook.com/nxtbooks/ieee/geoscience_december2018
https://www.nxtbook.com/nxtbooks/ieee/geoscience_september2018
https://www.nxtbook.com/nxtbooks/ieee/geoscience_june2018
https://www.nxtbook.com/nxtbooks/ieee/geoscience_march2018
https://www.nxtbook.com/nxtbooks/ieee/geoscience_december2017
https://www.nxtbook.com/nxtbooks/ieee/geoscience_september2017
https://www.nxtbook.com/nxtbooks/ieee/geoscience_june2017
https://www.nxtbook.com/nxtbooks/ieee/geoscience_march2017
https://www.nxtbook.com/nxtbooks/ieee/geoscience_december2016
https://www.nxtbook.com/nxtbooks/ieee/geoscience_september2016
https://www.nxtbook.com/nxtbooks/ieee/geoscience_june2016
https://www.nxtbook.com/nxtbooks/ieee/geoscience_march2016
https://www.nxtbook.com/nxtbooks/ieee/geoscience_december2015
https://www.nxtbook.com/nxtbooks/ieee/geoscience_september2015
https://www.nxtbook.com/nxtbooks/ieee/geoscience_june2015
https://www.nxtbook.com/nxtbooks/ieee/geoscience_march2015
https://www.nxtbook.com/nxtbooks/ieee/geoscience_december2014
https://www.nxtbook.com/nxtbooks/ieee/geoscience_september2014
https://www.nxtbook.com/nxtbooks/ieee/geoscience_june2014
https://www.nxtbook.com/nxtbooks/ieee/geoscience_march2014
https://www.nxtbook.com/nxtbooks/ieee/geoscience_december2013
https://www.nxtbook.com/nxtbooks/ieee/geoscience_september2013
https://www.nxtbook.com/nxtbooks/ieee/geoscience_june2013
https://www.nxtbook.com/nxtbooks/ieee/geoscience_march2013
https://www.nxtbookmedia.com