Silhouette Processing Via Mathematical Morphology with Pose-Aware Structuring Elements Based on 3D Model

• Autorzy: Jabłoński, M.
• Języki publikacji: EN

Abstrakty

EN

In the paper, the method of poseaware silhouette processing is presented. Morphological closing is proposed to enhance segmented silhouette object. The contribution of the work is adaptation of structuring element used for mathematical morphology erosions and dilations. It is proposed to use camera parameters, 3D model of the scene, model of the silhouette and its position to compute structuring element adequate to the individual projected to the camera image. Structuring element computation and basic morphology operators were implemented in OpenCL environment and tested on parallel GPU platform. Comparison with utility software packages is provided and results are briefly discussed.

Słowa kluczowe

EN

OpenCL; GPU; image processing algorithms

• Czasopismo: Image Processing & Communications
• Rocznik: 2012
• Tom: Vol. 17, no. 4
• Strony: 71--78
• Opis fizyczny: Bibliogr. 11 poz., rys., tab.

Twórcy

autor

Jabłoński, M.

• AGH University of Science and Technology, Institute of Automatics, al. A. Mickiewicza 30, 30-059 Kraków,

Bibliografia

• [1] Jabłoński M., Bubliński Z., Acceleration of image processing algorithms using graphics card resources (in Polish), Automatyka UWND, Vol. 13, No. 3/1, pp. 849-853, , 2009
• [2] M. Jabłoński, D. Pociecha, Automatic annotation of pin markers in tube pilger cold rolling process (in Polish), Automatyka, AGH, UWND, Vol. 14, No. 3/1, pp 289-299, 2010
• [3] M. Jabłoński, Scene configuration in stationary camera vision system (in polish), Automatyka, AGH, UWND, Vol. 15 No. 3, pp. 119-129, 2011
• [4] V. Lepetit, F. Moreno-Noguer, P. Fua, EPnP: An Accurate O(n) Solution to the PnP Problem, International Journal of Computer Vision, Vol. 81, No. 2, pp. 155-166, February 2009
• [5] L.D.F.D. Costa, R.M. Cesar, Shape Analysis and Classification: Theory and Practice (1st ed.), CRC Press, USA. 2000
• [6] The OpenCL Specification , Editor: Aaftab Munshi, Khronos OpenCLWorking Group, Version: 1.1, Document Revision: 4, 2011
• [7] J.R. Parker, Algorithms for Image Processing and Computer Vision, John Wiley & Sons, Inc., New York, USA, pp. 68-102, 1997
• [8] S. Pan, M. Kudo, Segmentation of pores in wood microscopic images based on mathematical morphology with a variable structuring element, Computers and Electronics in Agriculture, Vol. 75, No. 2, pp. 250-260, February 2011
• [9] S. Huet, V. Boulos, V. Fristot, L. Salvo, DFG implementation on multi GPU cluster with computation-communication overlap, Conference on Design and Architectures for Signal and Image Processing (DASIP), pp.1-8, Nov., 2011
• [10] F.Y. Shih, V. Gaddipati, General sweep mathematical morphology, Pattern Recognition, Vol. 36, No. 7, pp. 1489-1500, July 2003
• [11] R. Tadeusiewicz, M. Ogiela, Medical Image Understanding Technology: Studies in Fuzziness and Soft Computing, Vol. 156, Springer, 2004