Recognizing Planar Symbols with Severe Perspective Deformation

Abstract—A common problem encountered in recognizing symbols in real-scene images is the perspective deformation. In this paper, a recognition method resistant to perspective deformation is proposed. Particularly, a symbol is represented by a sequence of cross ratio spectra, in which perspective deformation is modeled as an uneven stretching deformation. Dynamic Time Warping algorithm is employed to compare the similarity of spectra and to find the pixel level correspondence between two symbols. The proposed method shows not only good resistance to severe perspective deformation and other common noise in real scene images, but also good discriminating ability to similar symbols.

• Type: Journal
• Year: 2010
• Where: PAMI
• Authors: Linlin Li, Chew Lim Tan
• Keywords: PAMI 2010, Perspective Deformation, Severe Perspective Deformation, Uneven Stretching Deformation