Quantitative Comparison of Tensorial Image Descriptions for the Application to Perceptual Grouping by the Tensor Voting Technique

A. Massad and B. Mertsching (Germany)


Computer Vision, Perceptual Grouping, Tensor Voting


We have applied the perceptual grouping method known as tensor voting to grey-level images by using local orien tation tensors. For that purpose, three tensorial descrip tion approaches are compared: Two methods are based on quadrature filters, namely Gabor filters and polar sep arable lognormal filters as introduced by Granlund and Knutsson. The third method employs structure tensors. The approaches are quantitatively evaluated on a set of test images with regards to their accuracy and noise robustness. According to the results, the tensors computed from a Ga bor filter set show the highest angular precision at edges while simultaneously representing junctions correctly.

Important Links:

Go Back