Multi-Scale Edge Extraction Based Stereo Matching Algorithm


Article Preview

In the field of robot vision, edge feature based stereo matching algorithm can reconstruct the targets with clear contours, which needs accurate and continuous target edges been extracted. In the paper, the smoothing filter operator was designed based on the discrete criteria of edge extraction and its correspondence optimal linear filter. Edge extraction was carried out incorporated the nonmaximum suppression and two thresholds techniques. The discrete criteria based multi-scale edge extraction method was studied with full use of the multi-scale character of the edge information. The detected multi-scale edges were synthesized to obtain the accurate and continuous single pixel wide edge. Then an edge feature based stereo matching algorithm was proposed to obtain 3D information of target. The experimental results demonstrate that the method can effectively suppress disturbance in outdoor environment and reconstruct target contour clearly.



Edited by:

Ran Chen




J. Lian et al., "Multi-Scale Edge Extraction Based Stereo Matching Algorithm", Applied Mechanics and Materials, Vols. 44-47, pp. 4162-4166, 2011

Online since:

December 2010




[1] Lazaros N, Sirakoulis G C, Gasteratos A. Review of stereo vision algorithms: From software to hardware. International Journal of Optomechatronics, Vol. 2, No. 4(2008), pp.435-462.


[2] Houben Q, Diaz J C T, Warzee N, Debeir O, Czyz J. Multi-feature stereo vision system for road traffic analysis. Proceedings of the 4th International Conference on Computer Vision Theory and Applications, Vol. 2(2009), pp.554-559.


[3] Sadeghi H, Moallem P, Monadjemi S A. Feature based color stereo matching algorithm using restricted search. Lecture Notes in Electrical Engineering, Proceedings of the European Computing Conference, Vol. 27, No. 1(2009), pp.105-111.


[4] Sach L T, Atsuta K, Hamamoto K, Kondo S. A robust road profile estimation method for low texture stereo images. International Conference on Image Processing, (2009), pp.4273-4276.


[5] Demigny D. On optimal linear filtering for edge detection. IEEE Transaction on Image Processing, Vol. 11, No. 7(2002), pp.728-737.