Title :
Properties of multichannel texture analysis filters
Author_Institution :
Dept. of Electr. & Comput. Eng., Texas Univ., Austin, TX, USA
Abstract :
Texture analysis algorithms that decompose images into oriented spatial frequency channels are studied. Optimality properties for texture segmentation filters are considered using idealized (narrowband) image texture models. The functional uncertainty of the channel filters is shown to define a tradeoff between spectral selectivity and accuracy in boundary localization that is optimized by the 2-D Gabor functions. The idealized texture model is then relaxed to analyze the effects of textural perturbations interpreted as localized amplitude and phase variations on the segmentation. The effects of these perturbations are found to be effectively ameliorated with postdetection smoothing
Keywords :
digital filters; pattern recognition; picture processing; 2-D Gabor functions; amplitude variations; boundary localization; functional uncertainty; image decomposition; image texture models; multichannel texture analysis filters; narrowband models; optimality properties; phase variations; postdetection smoothing; spatial frequency channels; spectral selectivity; textural perturbations; texture analysis algorithms; texture segmentation filters; Algorithm design and analysis; Frequency; Gabor filters; Image analysis; Image segmentation; Image texture; Image texture analysis; Narrowband; Smoothing methods; Uncertainty;
Conference_Titel :
Acoustics, Speech, and Signal Processing, 1990. ICASSP-90., 1990 International Conference on
Conference_Location :
Albuquerque, NM
DOI :
10.1109/ICASSP.1990.115958
