Skip to main content
Log in

Comparaison de deux familles complètes de descripteurs de formes pour l’indexation de bases d’objets 2D à niveaux de gris

Comparison between two complete sets of shape descriptors for 2D grey-level object content-based retrieval

  • Published:
Annals of Telecommunications Aims and scope Submit manuscript

Résumé

Cet article présente deux familles de descripteurs de formes pour l’indexation et la recherche d’objets à niveaux de gris dans des bases d’images. Ces familles sont invariantes par similitudes planes et complètes. La propriété de complétude garantit qu ’un objet est identifié de manière unique à une similitude près et assure une discrimination totale des formes, ce qui se révèle intéressant lorsque les collections d’images sont soit fortement homogènes, soit très volumineuses. La première famille complète de descripteurs invariants est issue du prolongement analytique de la transformée de Fourier-Mellin alors que la seconde est extraite des moments complexes de l’image. La similarité entre les objets est estimée grâce aux distances induites par les deux représentations invariantes. Ces familles ont été testées et évaluées ces familles sur deux bases d’objets à niveaux de gris. Les descripteurs se montrent discriminants et robustes aux faibles déformations de formes et les résultats de recherche sont encourageants. La propriété de complétude apparaît comme une propriété intéressante qui offre une représentation riche et flexible des formes.

Abstract

In this work, we propose the use of two sets of 2D shape descriptors for grey-level object content-based retrieval and image indexing. These families are invariant under planar similarities and complete. The completeness property ensures that an object is identified in a unique way up to a similarity transformation and guarantees a perfect discrimination between shapes. This reveals to be interesting when image databases are either homogeneous or large. The first set of invariant descriptors is extracted from the analytical Fourier-Mellin transform, while the second one comes from the complex moment image representation. The similarity between shapes is estimated according to the distances induced by the two invariant descriptions. We tested and compared these families on two real grey-level object databases for content-based image retrieval. Retrieval results show that the invariant descriptors produce a reliable shape description, discriminating and stable to small shape variations. The completeness is an interesting feature that offers an extensive and flexible representation of shapes.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Bibliographie

  1. Abu-Mostafa (Y.), Psaltis (D.), Image normalisation by complex moments,IEEE trans, on Pattern Analysis and Machine Intelligence, (1985), 7, n° 1, pp. 46–55.

    Article  Google Scholar 

  2. Aigrain (P.) et al, Content-based representation and retrieval of visual media: a state-of-the-art review,Multimedia Tools and Applications, (1996), 3, n° 3, pp 179–202.

    Article  Google Scholar 

  3. Bach (J.R.) et al, The Virage image search engine: An open framework for image management, dansSymposium on Electronic Imaging: Science and Technology — Storage & Retrieval for still image and video databases IV, (février 1996),2670, pp. 76–87.

    Google Scholar 

  4. Crimmins (T.R.), A complete set of Fourier descriptors for two-dimensional shapes,IEEE Trans. on Systems, Man, and Cybernetics, (1982),12, n° 6, pp. 848–855.

    Article  MathSciNet  Google Scholar 

  5. Derrode (S.), Représentation de formes planes à niveaux de gris par différentes approximations de Fourier-Mellin analytique en vue d’indexation de bases d’images,Thèse de l’université de Rennes, décembre 1999.

  6. Eggleston (P.), Content-based feature indexing and retrieval for image databases, inSPIE, (1992),1819.

  7. Flickner (M.) et al, Query by image and video content: the qbic system,IEEE Trans. on Computers, (1995),28, n° 9, pp. 23–32.

    Google Scholar 

  8. Gary (E.), Mehrotra (R.), Shape similarity-based retrieval in image databases, inSPIE, (1992),1662.

  9. Ghorbel (F.), Stability of invariant Fourier descriptors and its inference in the shape classification, dansProc. of the IEEE Int. Conf. on Pattern Recognition,20 aoû,t 1992, The Hague (Nl.), pp. 130–134.

    Google Scholar 

  10. Ghorbel (F.), A complete invariant description for gray-level images by the harmonic analysis approach,Pattern Recognition Letters, (1994),15, pp. 1043–1051.

    Article  Google Scholar 

  11. Ghorbel (F.), Toward a unitary formulation for invariant pattern recognition; application to image coding,Ann. Telecommunic, (1998),53, n° 5-6, pp. 242–260.

    Article  Google Scholar 

  12. Grace (A.E.) etSpann (M.), A comparison between Fourier-Mellin descriptors and moment based features for invariant object recognition using neural networks, Pattern Recognition Letters, 1991, Vol. 12, pp. 635–643.

    Article  Google Scholar 

  13. Kamoun (M.), Chakchouk (M.), Boulila Triki (N.) et Ghorbel (F.), Indexation des bases d’images par le contenu en faisant coopérer deux types de descripteurs de formes,Compression et REprésentation des Signaux Audiovisuels (CORESA’99), 14-15 juin 1999, Sophia- Antipolis (Fr.).

  14. Liao (S.X.) etPawlak (M.), On image analysis by moments,IEEE Trans. on Pattern Analysis and Machine Intelligence, 1996, Vol. 18(3), pp. 254–266.

    Article  Google Scholar 

  15. Mezhoud (R.), Bannour (T.), Derrode (S.) et Ghorbel (F.), A complete and invariant description of gray level images from geometric moments, dans1 er atelier de Traitement et d’Analyse d’Images : Méthodes et Applications (TA1MA’99), Hammamet (Tn), 1999.

  16. Milanese (R.) etCherbuliez (M.), A rotation-, translation-, and scale-invariant approach to content-based image retrieval,Journal of Visual Communication and Image Representation, 1999, Vol. 10, pp 186–196.

    Article  Google Scholar 

  17. Minka (T.) etPicard (R.), Interactive learning using a society of models,Pattern Recognition, 1997, Vol. 30(4), pp. 565–681.

    Article  Google Scholar 

  18. Ming-Kuel (H.), Visual pattern recognition by moment invariants, IRE Trans. on Information Theory, 1962, Vol. 8, pp. 179–187.

    Article  Google Scholar 

  19. Mokhtarian (F.), Abbasi (F.) et Kittler (J.), Robust and efficient shape indexing through curvature scale space, dansProc. of the 1996 British Machine and Vision Conference BMVC’96, septembre 1996, Edinburgh, Scotland.

  20. Pentland (A.), Picard (R.) etSclaroff (S.), Photo- book: Tools for content-based manipulation of image databases,International Journal of Computer Vision, 1996, Vol. 18(3), pp. 233–254.

    Article  Google Scholar 

  21. Rui (Y), Huang (T.S.), Ortega (M.) etMehrotra (S.), Relevance feedback: a power tool in interactive content-based image retrieval,IEEE trans. on Circuits and Systems for Video Technology, special issue on Interactive Multimedia Systems for the Internet, 1998, Vol. 8(5), pp. 644–655.

    Google Scholar 

  22. Rui (Y.), Huang (T.S.) etChang (S.), Image retrieval: current techniques, promising directions, and open issues,Journal of Visual Communication and Image Representation, 1999, Vol. 10, pp. 39–62.

    Article  Google Scholar 

  23. Schmid (C.) etMohr (R.), Local grey-value invariants for image retrieval,IEEE Trans. on Pattern Analysis and Machine Intelligence, 1997, Vol. 19(5), pp. 530–535.

    Article  Google Scholar 

  24. Sheng (Y), Duvernoy (J.), Circular-Fourier radial- Mellin transform descriptors for pattern recognition,Journal of the Optical Society of America A, 1986, Vol. 3(6), pp. 885–888.

    Article  Google Scholar 

  25. Teh (C.H.) etChin (R.T)., On image analysis by the method of moments,IEEE Trans. Pattern Analysis Machine Intelligence, 1988, Vol. 10(4), pp. 496–513.

    Article  Google Scholar 

Download references

Author information

Authors and Affiliations

Authors

Corresponding authors

Correspondence to Stéphane Derrode, Rim Mezhoud or Faouzi Ghorbel.

Rights and permissions

Reprints and permissions

About this article

Cite this article

Derrode, S., Mezhoud, R. & Ghorbel, F. Comparaison de deux familles complètes de descripteurs de formes pour l’indexation de bases d’objets 2D à niveaux de gris. Ann. Telecommun. 55, 184–193 (2000). https://doi.org/10.1007/BF03001911

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/BF03001911

Mots-clés

Key words

Navigation