Exploiting color and topological features for region segmentation with recursive fuzzy C-means

• Autorzy: Cucchiara R. , Grana C. , Seidenari S. , Pellacani G.
• Języki publikacji: EN

Abstrakty

EN

In this paper we define a novel approach to images segmentation into regions which focuses on both visual and topologocal cues, namely color similarity, inclusion and spatial adjacency. Many color clustering algorithms have been proposed in the past for skin lesion images but mone exploits explicity the inclusion properties between regions. Our algorithm is based on a recursive version of fuzzy c-means (FCM) clustering algorithm in the 2D color histogram constructed by Principal Component Analysis (PCA) of the color space. The distinctive feature of the proposal is that recursion is guided by evaluation of adjacency and mutual inclusion properties of extracted regions; then, the recursive analysis addresses only included or regions with a non-negligible size. This approach allows a coarse-to-fine segmentation which focuses attention on the inner parts of the images, in order to highlight the internal structure of the object depiced in the image. This could be particulary useful in many applications, especially in biomedical image analysis. Inthis work we apply the technique to segmentation of skin lesions in dermatoscopic images. It could be a suitable support for diagnosis of skin melanoma, since dermatologists are interrested in analysis of spatial relations, symmetrical positions and inlusion of regions.

Słowa kluczowe

EN

PCA; fuzzy c-means; clustering; color segmentation; topological features; dermatoscopic images

• Wydawca: Institute of Information Technology of the Warsaw University of Life Sciences – SGGW
• Czasopismo: Machine Graphics and Vision
• Rocznik: 2002
• Tom: Vol. 11, No. 2/3
• Strony: 169--182
• Opis fizyczny: Bibliogr. 22 poz.

Twórcy

autor

Cucchiara R.

• Dipartamento di Scienze dell'Ingegneria, Univesita di Modena e Reggio Emilia, Italy

autor

Grana C.

• Dipartamento di Scienze dell'Ingegneria, Univesita di Modena e Reggio Emilia, Italy

autor

Seidenari S.

• Dipartamento di Scienze Neuropsicosensoriali, Univesita di Modena e Reggio Emilia, Italy

autor

Pellacani G.

• Dipartamento di Scienze Neuropsicosensoriali, Univesita di Modena e Reggio Emilia, Italy

Bibliografia

• [1] Edelsbrunner H.: Algorithms in Combinatorial Geometry. S-V, Hidelberg. 1987.
• [2] Lim Y. W ., Lee S. U.: On the color image segmentation algorithm based on the tresholding and the fuzzy c-means techniques. PR, 23 (9), 935-952. 1990.
• [3] Dave R. N.: Boundary Detection through Fuzzy Clustering. Invited Paper, IEEE Int. Conf. on Fuzzy System, San Diego, California, March 8-12, 127-134. 1992.
• [4] Gonzalez R. C., Wintz P. A.: Digital Image Processing. Reading, MA: Addison-Wesley. 1992.
• [5] Kasson J. K., Plouffe W.: An analysis of selected computer inter-change color spaces. ACM Trans. Graphics , 11(4), 373-405. 1992.
• [6] Umbaugh S. E., Moss R. H., Stoecker W. V., Hance G. A.: Automatic color segmentation algorithms: with application to skin tumor feature identification. IEEE Engineering in Medicine and Biology, 12(3), 75-82. 1993.
• [7] Wu Z., Leahy R.: An optimal graph theoretic approach to data clusering: Theory and its application to image segmentation. IEEE Trans. PAMI, 15(11), 1101-1113. 1993.
• [8] Green A., Martin N., Pfitzner J., O'Rourke M., Knight N.: Computer image analysis in the diagnosis of melanoma. J. of the American Academy f Dermatology, 31, December, 958-964. 1994.
• [9] Hall P. N., Claridge E., Morris Smith J. D.: Computer screening for early detection of melanoma- is there a future? Br. J. Dermatol., 132, 325-338. 1995.
• [10] Hance G. A., Umbaugh S. E., Moss R.H., Stocker W. V.: Unsupervised color image segmentation with application to skin tumor borders. IEEE Engineering in Medicine and Biology, 15(1), 104-111. 1996.
• [11] Argenyi Z. B.: Dermatoscopy (epiluminescenece microscopy) of pigmented skin lesions. Dermatologic, Clinics, 15(1), 79-95. 1997.
• [12] Gutkowicz-Krusin D., Elbaum M., Szwaykowski P., Kopf A. W.: Can early malignant melanoma be differentiated from atypical melanocytic nevus by in vivo techniques? Part II. Automatic machine vision classification. Skin Res Technol., 3(1), 15-22. 1997.
• [13] Schmid Ph., Fischer S.: Colour segmentation for the analysis of pigmented skin lesions". Proc. of the IPA'97, The Institution of Electrical Enginers, Dublin, July, vol. 2, 688-692. 1997.
• [14] Fleming M. G., Steger., Zhang J., Gao J., Cognetta A. B., Pollak I., Dyer C. R.: Techniques for a structural analysis of dermatoscopc imagery. Computerized Medical Imaging and Graphics, 22(5), 375-389. 1998.
• [15] Leeser M., Kitaryeva N., Crisman J.: Spatial ang color clustering on an FPGA-based computer system. Proc. SPIE, Vol. 3526, 25-33, Nov. Also available as Northeastern University TR ECE-CEG-98-024. 1998.
• [16] Schmid P.: Segmentation of digitized dermatoscopic images by two-dimensional color clustering. IEEE Trans. on Medical Imaging, 18(2), 164-171. 1999.
• [17] Day G. R.: How blurry i that border? An investigation into algorithmic reproduction of skin lesion border cut-off. Computerized Medical Imaging and Graphics, 24, 69-72. 2000.
• [18] Devaux J.-C., Gouton P., Truchet F.: Aerial colour image segmentation by karhunen-loeve transform. IEEE Int. Conf. on PR, Barcelona, Spain, Sept., 309-312. 2000.
• [19] Noordam J. C., van der Broek W. H. A. M., Buydens L. M. C.: Geometrically guided fuzzy c-meansclustering for multivariate image segmentation. Proc. of ICPR 2000, 1, 462-465. 2000.
• [20] Zhang C., Wang P.: A new method of color image segmentation based on intensity and hue clsustring. Proc. of ICPR 2000, 2, 617-620. 2000.
• [21] Berretti S., Del Bimbo A., Vicario E.: Efficient matching and indexing of graph models in contentbased retrieval. IEEE Trans. on PR&MI, 23(10), 1089-1104. 2001.
• [22] Cucchiara R., Grana C.: Using the topological tree for skin lesion structure description. Proc. of Sixth Int. Conf. on Knowledge-Based lntelligent lnformation and Engineering Systems, Italy, Sept. 2002.