A Comparison of the Octave-band Directional Filter Bank and Gabor Filters for Texture Classification (TA-P4)

Author(s) : Paul Hong (Georgia Institute of Technology, USA) Lance Kaplan (Clark Atlanta University, USA) Mark Smith (Purdue University, USA)

Abstract : This paper investigates the similarities between Gabor filters and the octave-directional filter bank (OBDFB) with respect to texture classification. Both decompositions can discriminate angularly and radially in the 2-D frequency domain; however, the OBDFB has a computationally efficient implementation that makes it an attractive alternative to Gabor filters. An analysis of local energy features from Gabor filters and the OBDFB is presented showing similarities between their texture features. Texture classification experiments run using both decompositions show comparable classification accuracy.

Menu