Abstract

Three-dimensional models of urban objects are widely used in geographic information systems, telecommunications, or defense applications. The classic technique for obtaining such models is stereoscopy. Images are densely matched, and images of above-ground structures are delineated. We propose two semiautomatic methods based on the Hough transform and statistically active models to delineate buildings. The first one delineates rectangular shapes; the second one deals with more-complex buildings. Each one is based on a criterion optimization that takes both photometric and altimetric information into account. Results based on real data show that the first method is robust and that the second one, which deals with a broad range of buildings, seems to be a good compromise between robustness and applicability.

© 2004 Optical Society of America

Full Article  |  PDF Article

References

  • View by:
  • |
  • |
  • |

  1. W. Forstner, “3D-city models: automatic and semiautomatic acquisition methods,” in Proceedings of Photogrammetric Week’99, D. Fritsh, R. Spiller, eds. (Wichmann, Karlsruhe, Germany, 1999), http://www.ipb.uni-bonn.de/Publications/papers99/foerstner99_city.ps.gz .
  2. C. Baillard, A. Zisserman, “A plane-sweep strategy for the 3D reconstruction of buildings from multiple images,” in 19th ISPRS Congress and Exhibition (n.p., Amsterdam, 2000), Vol. 32, Part B2, pp. 56–62, http://citeseer.nj.nec.com/baillard00planesweep.html .
  3. W. Weidner, W. Forstner, “Towards automatic building extraction from high resolution digital elevation models,” ISPRS J. Photogram. Remote Sens. 50(4), 38–49 (1995).
    [CrossRef]
  4. N. Paparoditis, M. Cord, M. Jordan, J. P. Cocquerez, “Building detection and reconstruction from mid- and high-resolution aerial imagery,” Comput. Vis. Image Underst. 72, 122–142 (1998).
    [CrossRef]
  5. A. Michel, H. Oriot, O. Goretta, “Extraction of rectangular roofs on stereoscopic images—an interactive approach,” in Proceedings of ISPRS ’98 (Ohio State U., Columbus, 1998), Vol. 32, Part III/I, pp. 367–373.
  6. S. Scholze, T. Moons, L. Van Gool, “A probabilistic approach to roof extraction and reconstruction,” in Photogrammetric Computer Vision (ISPRS, Graz, Austria, 2002), Commission III, Vol. B, pp. 231–237.
  7. N. Haala, C. Brenner, “Extraction of buildings and trees in urban environments,” J. Photogramm. Remote Sens. 54, 130–137 (1999).
    [CrossRef]
  8. A. Fisher, T. H. Kolbe, F. Lang, “Integration of 2D and 3D reasoning for building reconstruction using a generic hierarchical model,” in Proceedings of the Workshop on Semantic Modeling for the Acquisition of Topographic Information from Images and Maps SMATI’97 (Birkhauser Verlag, Berlin, 1997), pp. 159–180.
  9. M. Ortner, X. Descombes, J. Zerubia, “Building extraction from digital elevation model,” Rapport INRIA 4517 (INRIA, n.p., 2002), http://www-sop.inria.fr/rapports/sophia/RR-4517.html .
  10. O. Faugeras, Three-Dimensional Computer Vision: a Geometric Viewpoint (MIT Press, Cambridge, Mass., 1993).
  11. H. Oriot, G. Le Besnerais, “Matching aerial stereo images using graduated non convexity techniques,” in 19th ISPRS Congress and Exhibition (Ohio State U., Columbus, 1998), Vol. 32, Part III/I, pp. 146–152.
  12. P. V. C. Hough, “Method and means for recognizing complex patterns,” U.S. patent3,069,654 (18December1962).
  13. C. Chesnaud, “Techniques statistiques de segmentation par contour actif et mise en oeuvre rapide,” (“Snake region based segmentation and fast algorithm implementation”), Ph.D. dissertation (Université de Droit, d’Économie et des Sciences, Aix-Marseille, France, 2000).
  14. C. Chesnaud, Ph. Réfrégier, “Statistical region snake based segmentation adapted to different physical noise models,” IEEE Trans. Pattern Anal. Mach. Intell. 21, 1137–1157 (1999).
    [CrossRef]
  15. C. Chesnaud, Ph. Réfrégier, “Snake region based segmentation for different physical noise models and fast algorithm implementation,” presented at “Physics in Signal and Image Processing, PSIP’99 First International Symposium,” Paris, France, 18–19 January 1999.
  16. H. Oriot, “Statistical snakes for building extraction from stereoscopic aerial images,” in ISPRS Conference on Photographic Image Analyses (Institute of Photogrammetry and Geoinformation, U. Hannover, Hannover, Germany, 2003), pp. 65–70.
  17. http://isprs.ign.fr .
  18. C. Baillard, “Analyse d’images aériennes stéréoscopiques pour la restitution 3D des milieux urbains” (“Analysis of stereoscopic aerial images for urban 3D extraction”), Ph.D. dissertation, École Nationale Supériure des Télécommunications, Paris, France, 1997).
  19. C. Baillard, O. Dissard, O. Jamet, H. Maitre, “Detection of above-ground in urban areas: application to DTM generation,” in Image and Signal Processing for Remote Sensing III, J. Desachy, P. Sabatier, eds., Proc. SPIE2955, 129–140 (1996).
    [CrossRef]

1999

N. Haala, C. Brenner, “Extraction of buildings and trees in urban environments,” J. Photogramm. Remote Sens. 54, 130–137 (1999).
[CrossRef]

C. Chesnaud, Ph. Réfrégier, “Statistical region snake based segmentation adapted to different physical noise models,” IEEE Trans. Pattern Anal. Mach. Intell. 21, 1137–1157 (1999).
[CrossRef]

1998

N. Paparoditis, M. Cord, M. Jordan, J. P. Cocquerez, “Building detection and reconstruction from mid- and high-resolution aerial imagery,” Comput. Vis. Image Underst. 72, 122–142 (1998).
[CrossRef]

1995

W. Weidner, W. Forstner, “Towards automatic building extraction from high resolution digital elevation models,” ISPRS J. Photogram. Remote Sens. 50(4), 38–49 (1995).
[CrossRef]

Baillard, C.

C. Baillard, “Analyse d’images aériennes stéréoscopiques pour la restitution 3D des milieux urbains” (“Analysis of stereoscopic aerial images for urban 3D extraction”), Ph.D. dissertation, École Nationale Supériure des Télécommunications, Paris, France, 1997).

C. Baillard, O. Dissard, O. Jamet, H. Maitre, “Detection of above-ground in urban areas: application to DTM generation,” in Image and Signal Processing for Remote Sensing III, J. Desachy, P. Sabatier, eds., Proc. SPIE2955, 129–140 (1996).
[CrossRef]

Brenner, C.

N. Haala, C. Brenner, “Extraction of buildings and trees in urban environments,” J. Photogramm. Remote Sens. 54, 130–137 (1999).
[CrossRef]

Chesnaud, C.

C. Chesnaud, Ph. Réfrégier, “Statistical region snake based segmentation adapted to different physical noise models,” IEEE Trans. Pattern Anal. Mach. Intell. 21, 1137–1157 (1999).
[CrossRef]

C. Chesnaud, “Techniques statistiques de segmentation par contour actif et mise en oeuvre rapide,” (“Snake region based segmentation and fast algorithm implementation”), Ph.D. dissertation (Université de Droit, d’Économie et des Sciences, Aix-Marseille, France, 2000).

C. Chesnaud, Ph. Réfrégier, “Snake region based segmentation for different physical noise models and fast algorithm implementation,” presented at “Physics in Signal and Image Processing, PSIP’99 First International Symposium,” Paris, France, 18–19 January 1999.

Cocquerez, J. P.

N. Paparoditis, M. Cord, M. Jordan, J. P. Cocquerez, “Building detection and reconstruction from mid- and high-resolution aerial imagery,” Comput. Vis. Image Underst. 72, 122–142 (1998).
[CrossRef]

Cord, M.

N. Paparoditis, M. Cord, M. Jordan, J. P. Cocquerez, “Building detection and reconstruction from mid- and high-resolution aerial imagery,” Comput. Vis. Image Underst. 72, 122–142 (1998).
[CrossRef]

Dissard, O.

C. Baillard, O. Dissard, O. Jamet, H. Maitre, “Detection of above-ground in urban areas: application to DTM generation,” in Image and Signal Processing for Remote Sensing III, J. Desachy, P. Sabatier, eds., Proc. SPIE2955, 129–140 (1996).
[CrossRef]

Faugeras, O.

O. Faugeras, Three-Dimensional Computer Vision: a Geometric Viewpoint (MIT Press, Cambridge, Mass., 1993).

Fisher, A.

A. Fisher, T. H. Kolbe, F. Lang, “Integration of 2D and 3D reasoning for building reconstruction using a generic hierarchical model,” in Proceedings of the Workshop on Semantic Modeling for the Acquisition of Topographic Information from Images and Maps SMATI’97 (Birkhauser Verlag, Berlin, 1997), pp. 159–180.

Forstner, W.

W. Weidner, W. Forstner, “Towards automatic building extraction from high resolution digital elevation models,” ISPRS J. Photogram. Remote Sens. 50(4), 38–49 (1995).
[CrossRef]

W. Forstner, “3D-city models: automatic and semiautomatic acquisition methods,” in Proceedings of Photogrammetric Week’99, D. Fritsh, R. Spiller, eds. (Wichmann, Karlsruhe, Germany, 1999), http://www.ipb.uni-bonn.de/Publications/papers99/foerstner99_city.ps.gz .

Goretta, O.

A. Michel, H. Oriot, O. Goretta, “Extraction of rectangular roofs on stereoscopic images—an interactive approach,” in Proceedings of ISPRS ’98 (Ohio State U., Columbus, 1998), Vol. 32, Part III/I, pp. 367–373.

Haala, N.

N. Haala, C. Brenner, “Extraction of buildings and trees in urban environments,” J. Photogramm. Remote Sens. 54, 130–137 (1999).
[CrossRef]

Hough, P. V. C.

P. V. C. Hough, “Method and means for recognizing complex patterns,” U.S. patent3,069,654 (18December1962).

Jamet, O.

C. Baillard, O. Dissard, O. Jamet, H. Maitre, “Detection of above-ground in urban areas: application to DTM generation,” in Image and Signal Processing for Remote Sensing III, J. Desachy, P. Sabatier, eds., Proc. SPIE2955, 129–140 (1996).
[CrossRef]

Jordan, M.

N. Paparoditis, M. Cord, M. Jordan, J. P. Cocquerez, “Building detection and reconstruction from mid- and high-resolution aerial imagery,” Comput. Vis. Image Underst. 72, 122–142 (1998).
[CrossRef]

Kolbe, T. H.

A. Fisher, T. H. Kolbe, F. Lang, “Integration of 2D and 3D reasoning for building reconstruction using a generic hierarchical model,” in Proceedings of the Workshop on Semantic Modeling for the Acquisition of Topographic Information from Images and Maps SMATI’97 (Birkhauser Verlag, Berlin, 1997), pp. 159–180.

Lang, F.

A. Fisher, T. H. Kolbe, F. Lang, “Integration of 2D and 3D reasoning for building reconstruction using a generic hierarchical model,” in Proceedings of the Workshop on Semantic Modeling for the Acquisition of Topographic Information from Images and Maps SMATI’97 (Birkhauser Verlag, Berlin, 1997), pp. 159–180.

Le Besnerais, G.

H. Oriot, G. Le Besnerais, “Matching aerial stereo images using graduated non convexity techniques,” in 19th ISPRS Congress and Exhibition (Ohio State U., Columbus, 1998), Vol. 32, Part III/I, pp. 146–152.

Maitre, H.

C. Baillard, O. Dissard, O. Jamet, H. Maitre, “Detection of above-ground in urban areas: application to DTM generation,” in Image and Signal Processing for Remote Sensing III, J. Desachy, P. Sabatier, eds., Proc. SPIE2955, 129–140 (1996).
[CrossRef]

Michel, A.

A. Michel, H. Oriot, O. Goretta, “Extraction of rectangular roofs on stereoscopic images—an interactive approach,” in Proceedings of ISPRS ’98 (Ohio State U., Columbus, 1998), Vol. 32, Part III/I, pp. 367–373.

Moons, T.

S. Scholze, T. Moons, L. Van Gool, “A probabilistic approach to roof extraction and reconstruction,” in Photogrammetric Computer Vision (ISPRS, Graz, Austria, 2002), Commission III, Vol. B, pp. 231–237.

Oriot, H.

A. Michel, H. Oriot, O. Goretta, “Extraction of rectangular roofs on stereoscopic images—an interactive approach,” in Proceedings of ISPRS ’98 (Ohio State U., Columbus, 1998), Vol. 32, Part III/I, pp. 367–373.

H. Oriot, “Statistical snakes for building extraction from stereoscopic aerial images,” in ISPRS Conference on Photographic Image Analyses (Institute of Photogrammetry and Geoinformation, U. Hannover, Hannover, Germany, 2003), pp. 65–70.

H. Oriot, G. Le Besnerais, “Matching aerial stereo images using graduated non convexity techniques,” in 19th ISPRS Congress and Exhibition (Ohio State U., Columbus, 1998), Vol. 32, Part III/I, pp. 146–152.

Paparoditis, N.

N. Paparoditis, M. Cord, M. Jordan, J. P. Cocquerez, “Building detection and reconstruction from mid- and high-resolution aerial imagery,” Comput. Vis. Image Underst. 72, 122–142 (1998).
[CrossRef]

Réfrégier, Ph.

C. Chesnaud, Ph. Réfrégier, “Statistical region snake based segmentation adapted to different physical noise models,” IEEE Trans. Pattern Anal. Mach. Intell. 21, 1137–1157 (1999).
[CrossRef]

C. Chesnaud, Ph. Réfrégier, “Snake region based segmentation for different physical noise models and fast algorithm implementation,” presented at “Physics in Signal and Image Processing, PSIP’99 First International Symposium,” Paris, France, 18–19 January 1999.

Scholze, S.

S. Scholze, T. Moons, L. Van Gool, “A probabilistic approach to roof extraction and reconstruction,” in Photogrammetric Computer Vision (ISPRS, Graz, Austria, 2002), Commission III, Vol. B, pp. 231–237.

Van Gool, L.

S. Scholze, T. Moons, L. Van Gool, “A probabilistic approach to roof extraction and reconstruction,” in Photogrammetric Computer Vision (ISPRS, Graz, Austria, 2002), Commission III, Vol. B, pp. 231–237.

Weidner, W.

W. Weidner, W. Forstner, “Towards automatic building extraction from high resolution digital elevation models,” ISPRS J. Photogram. Remote Sens. 50(4), 38–49 (1995).
[CrossRef]

Comput. Vis. Image Underst.

N. Paparoditis, M. Cord, M. Jordan, J. P. Cocquerez, “Building detection and reconstruction from mid- and high-resolution aerial imagery,” Comput. Vis. Image Underst. 72, 122–142 (1998).
[CrossRef]

IEEE Trans. Pattern Anal. Mach. Intell.

C. Chesnaud, Ph. Réfrégier, “Statistical region snake based segmentation adapted to different physical noise models,” IEEE Trans. Pattern Anal. Mach. Intell. 21, 1137–1157 (1999).
[CrossRef]

ISPRS J. Photogram. Remote Sens.

W. Weidner, W. Forstner, “Towards automatic building extraction from high resolution digital elevation models,” ISPRS J. Photogram. Remote Sens. 50(4), 38–49 (1995).
[CrossRef]

J. Photogramm. Remote Sens.

N. Haala, C. Brenner, “Extraction of buildings and trees in urban environments,” J. Photogramm. Remote Sens. 54, 130–137 (1999).
[CrossRef]

Other

A. Fisher, T. H. Kolbe, F. Lang, “Integration of 2D and 3D reasoning for building reconstruction using a generic hierarchical model,” in Proceedings of the Workshop on Semantic Modeling for the Acquisition of Topographic Information from Images and Maps SMATI’97 (Birkhauser Verlag, Berlin, 1997), pp. 159–180.

M. Ortner, X. Descombes, J. Zerubia, “Building extraction from digital elevation model,” Rapport INRIA 4517 (INRIA, n.p., 2002), http://www-sop.inria.fr/rapports/sophia/RR-4517.html .

O. Faugeras, Three-Dimensional Computer Vision: a Geometric Viewpoint (MIT Press, Cambridge, Mass., 1993).

H. Oriot, G. Le Besnerais, “Matching aerial stereo images using graduated non convexity techniques,” in 19th ISPRS Congress and Exhibition (Ohio State U., Columbus, 1998), Vol. 32, Part III/I, pp. 146–152.

P. V. C. Hough, “Method and means for recognizing complex patterns,” U.S. patent3,069,654 (18December1962).

C. Chesnaud, “Techniques statistiques de segmentation par contour actif et mise en oeuvre rapide,” (“Snake region based segmentation and fast algorithm implementation”), Ph.D. dissertation (Université de Droit, d’Économie et des Sciences, Aix-Marseille, France, 2000).

A. Michel, H. Oriot, O. Goretta, “Extraction of rectangular roofs on stereoscopic images—an interactive approach,” in Proceedings of ISPRS ’98 (Ohio State U., Columbus, 1998), Vol. 32, Part III/I, pp. 367–373.

S. Scholze, T. Moons, L. Van Gool, “A probabilistic approach to roof extraction and reconstruction,” in Photogrammetric Computer Vision (ISPRS, Graz, Austria, 2002), Commission III, Vol. B, pp. 231–237.

W. Forstner, “3D-city models: automatic and semiautomatic acquisition methods,” in Proceedings of Photogrammetric Week’99, D. Fritsh, R. Spiller, eds. (Wichmann, Karlsruhe, Germany, 1999), http://www.ipb.uni-bonn.de/Publications/papers99/foerstner99_city.ps.gz .

C. Baillard, A. Zisserman, “A plane-sweep strategy for the 3D reconstruction of buildings from multiple images,” in 19th ISPRS Congress and Exhibition (n.p., Amsterdam, 2000), Vol. 32, Part B2, pp. 56–62, http://citeseer.nj.nec.com/baillard00planesweep.html .

C. Chesnaud, Ph. Réfrégier, “Snake region based segmentation for different physical noise models and fast algorithm implementation,” presented at “Physics in Signal and Image Processing, PSIP’99 First International Symposium,” Paris, France, 18–19 January 1999.

H. Oriot, “Statistical snakes for building extraction from stereoscopic aerial images,” in ISPRS Conference on Photographic Image Analyses (Institute of Photogrammetry and Geoinformation, U. Hannover, Hannover, Germany, 2003), pp. 65–70.

http://isprs.ign.fr .

C. Baillard, “Analyse d’images aériennes stéréoscopiques pour la restitution 3D des milieux urbains” (“Analysis of stereoscopic aerial images for urban 3D extraction”), Ph.D. dissertation, École Nationale Supériure des Télécommunications, Paris, France, 1997).

C. Baillard, O. Dissard, O. Jamet, H. Maitre, “Detection of above-ground in urban areas: application to DTM generation,” in Image and Signal Processing for Remote Sensing III, J. Desachy, P. Sabatier, eds., Proc. SPIE2955, 129–140 (1996).
[CrossRef]

Cited By

OSA participates in CrossRef's Cited-By Linking service. Citing articles from OSA journals and other participating publishers are listed here.

Alert me when this article is cited.


Figures (16)

Fig. 1
Fig. 1

Initialization of the building dilineation algorithm on the (a) left- and (b) right-hand image of a stereoscopic pair in La Defense district of Paris.

Fig. 2
Fig. 2

Straight line in image space (left) and in Hough space (right).

Fig. 3
Fig. 3

Hough transform of a rectangle.

Fig. 4
Fig. 4

(a), (b) Left- and right-hand images. White lines show the initialization. (c), (d) Extracted buildings from the left- and right-hand images, respectively. Photos taken in Montreuil by IGN (French National Geographic Institute).

Fig. 5
Fig. 5

Extraction of left- and right-hand images of a low-contrast building in Montreuil.

Fig. 6
Fig. 6

Extraction of a sloping roof in Montreuil.

Fig. 7
Fig. 7

Mean chamfer distance for values of the initialization threshold for the three buildings shown in Figs. 46.

Fig. 8
Fig. 8

Contours and areas associated with a given polygon.

Fig. 9
Fig. 9

Regularizing term for angles that vary from 0 to 180°.

Fig. 10
Fig. 10

Extraction of a low-contrast building (the original image is of an area in Amiens). See Ref. 17.

Fig. 11
Fig. 11

Extraction of a heterogeneous complex building in Amiens. See Ref. 17.

Fig. 12
Fig. 12

Extraction of two buildings in Amiens. See Ref. 17.

Fig. 13
Fig. 13

Mean chamfer distances for values of (a) the percentage of nonoccluded pixels, (b) the rigidity parameter, and (c) the edge parameters for the buildings shown in Figs. 1012.

Fig. 14
Fig. 14

Stereoscopic pair of aerial images of La Defense district in Paris.

Fig. 15
Fig. 15

Buildings extracted by the two proposed methods. Rectangular buildings are delineated with algorithm explained in Section 4; other buildings are delineated with the algorithm explained in Section 5.

Fig. 16
Fig. 16

Enlargement of three buildings from Fig. 15 whose extraction presents problems (see text).

Tables (1)

Tables Icon

Table 1 Analysis of Parameters Used for Complex-Building Extractions

Equations (4)

Equations on this page are rendered with MathJax. Learn more.

Hρ, θ=i,jDρ,θIi,jNθ1+disti,j,
ρˆ1, ρˆ2, ρˆ3, ρ̑4, θ̑=argmaxθ,ρ1,ρ2,ρ3,ρ4Hρ1ρaccessθρ1, θ+Hρ2ρaccessθρ2, θ+Hρ3ρaccessθρ3, θ+π2+Hρ4ρaccessθρ4, θ+π2,
ρ1, ρ2, ρ3, ρ4, θ=argmaxθmaxρρaccessθHρ, θ+maxρρaccessθHρ, θ+maxρρaccessθHρ, θ+π2+maxρρaccessθHρ, θ+π2.
F=x,yTcminddbCorrIlx, y, Irx+d, y term 1 +x,yEcminddgCorrIlx, y, Irx+d, y term 2 +x,yOcNo-matching-cost term 3 -λ x,yCgradIlx, ynx,y term 4 -λ x,yCgradIrx+dx,y, ynx,y term 5 +β siSC Rsi-1, si, si+1. term 6

Metrics