3.2 RGB image segmentation and removal
The second experimental data which is the focus in this study is
close-range building RGB image under the ground platform.
Because this is visible light RGB image shot by ordinary digital
camera, NDVI approach is no longer applicable. CIE L*a*b
method is used for segment. Under the threshold a<-12,
removal results is Fig 2 5). The sky and windows are segmented
in error. The segment method of satellite image with CIE L*a*b
International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Volume XXXIX-B3, 2012
XXII ISPRS Congress, 25 August — 01 September 2012, Melbourne, Australia
does not apply to ground image, because there are some noise
by foreground and background. Vegetation extraction based on
SVM is utilized under human supervision, whose result can be
seen in Fig 2 c). The green areas in the image are vegetation
ones. And the correct recognition rate of vegetation occlusion is
not only better than Fig 2 5), but also the wrong recognition rate
is reduced greatly.
c)
Figure 3. a) True color close-range image. 5) Result by CIE L*a*b with a threshold. c) Result by feature (L, a, b) with SVM.
The third experimental data is close-range building RGB image
where there are yellow or brown vegetations. Under the
threshold a«-12, there are not only many error segment areas in
Fig 3. b), but also some withered and yellow vegetation isn't
recognized by this method. Thus, SVM method based on
selecting (L, a, b) as feature is implemented for removal
vegetation occlusion. The experimental results are shown in Fig
3 c). We can find that almost all the occlusion of vegetation is
removed.
3.3 Conclusions and future work
An effective approach for extraction and segment the vegetation
occlusion of building by RGB close-range imagery is presented.
The proposed image segmentation approach by CIE L*a*b and
SVM has good potential for the removal of vegetation occlusion
from imagery since it works on both CIR and standard RGB
imagery. Especially, no matter whether vegetation is green or
yellow, it adapts to the segment and removal of ground
vegetation occlusion. We can draw the conclusion that RGB
image can substitute for CIR image to recognize vegetation
occlusion.
Tree that is a familiar occlusion has branches, leaves and trunk,
which leads to spectrum difference, texture difference and
geometry difference (edge). Hence, in order to get better result
of occlusion removal, more information and feature (spectrum,
texture and geometry) should be implemented and synthesized.
This work should pave the way for texture reconstruction and
repair for future 3D reconstruction.
3.4 Acknowledgements and Appendix (optional)
This work is supported by National Natural Science Foundation
of China (NSFC) (Grant No. 41101407, 41001204 and
41001260), the Natural Science Foundation of Hubei Province,
China (Grant No. 2010CDZ005) and self-determined research
funds of CCNU from the colleges’ basic research and operation
of MOE (Grant No. CCNU10A01001). Heartfelt thanks are also
given for the comments and contributions of anonymous
reviewers and members of the Editorial team.
References
[1] BURGES, CJ.C., 1998. A Tutorial on Support Vector
Machines for Pattern Recognition. Data Mining and
Knowledge Discovery(2): 121—167.
[2] Ford, A. and Roberts, A., 1998. Colour Space
Conversions -
http://www.poynton.com/PDFs/coloureg.pdf.
[3] HunterLab, 2008. CIE L*a*b* Color Scale -
www.hunterlab.com/appnotes/an07_96a.pdf.
[4] LI, C. and Zhou, Y., 2010. 3D Auto-Reconstruction for
Street Elevation Based on Line and Plane Feature, The
2nd International Conference on Computer and
Automation Engineering, Singapore, pp. 460-466.
[5] LIU, Y. and GUAN, Z., 2010. A GRID-BASE LINE
ANALYSIS FOR STREET OCCLUSION REMOVAL
AND BUILDING FACADE TEXTURING, ASPRS
Annual Conference , San Diego,Califonia.
[6] LYON, J.G., YUAN, D., LUNETTA, R.S. and ELVIDGE,
C.D., 1998. A change detection experiment using
vegetation indices. Photogrammetric Engineering and
Remote Sensing, 66: 143—150.
[7] NASA, Normalized Difference Vegetation Index (NDVI)
http://earthobservatory.nasa.gov/Features/MeasuringVege
tation/measuring vegetation 2.php.
[8] Ortin, D. and Remondino, F., 2005. OCCLUSION-FREE
IMAGE GENERATION FOR REALISTIC TEXTURE
MAPPING, International Archives of Photogrammetry,
Remote Sensing and Spatial Information Sciences,
Mestre-Venice, Italy.
[9] SEII, E., 2002. Theory perceive the color, introduction to
graphic of computer - http://semmix.pl/color/default.htm.
[10] ZHANG, Y., Heipke, C., Butenuth, M. and HU, X., 2006.
Automatic Extraction of Wind Erosion Obstacles by
Integration of GIS Data, DSM and Stereo Images.
International Journal of Remote Sensing, 27(8): 1677-
1690.
Revised July 2011