Title :
On the edge preserving smoothing filter
Author :
Hung, Chih-Cheng
Author_Institution :
Dept. of Math. & Comput. Sci., Alabama A&M Univ., Normal, AL, USA
Abstract :
An improvement on the gradient inverse weighted (GIW) filter for digital image smoothing is proposed. The selection of the optimal homogeneous neighboring pixels for spatial smoothing is achieved through the integration of directional masks and the adaptation of the selected subregion based on local statistics of the image data. This adaptive approach is effective in smoothing images containing complex-shaped objects such as remotely sensed imagery. Experimental results and the comparison with Tomita´s (1977) filter, Nagao´s (1979) filter, and the GIW filter are also presented
Keywords :
adaptive filters; adaptive signal processing; edge detection; image enhancement; image segmentation; nonlinear filters; smoothing methods; Nagao´s filter; Tomita´s filter; adaptive approach; complex shaped objects; digital image smoothing; directional masks; edge preserving smoothing filter; experimental results; gradient inverse weighted filter; image data; image enhancement; local statistics; nonlinear filtering algorithm; optimal homogeneous neighboring pixels; remotely sensed imagery; spatial smoothing; subregion adaptation; Adaptive filters; Computer science; Digital filters; Digital images; Electronic mail; Mathematics; Noise shaping; Pixel; Shape; Smoothing methods;
Conference_Titel :
Southeastcon '97. Engineering new New Century., Proceedings. IEEE
Conference_Location :
Blacksburg, VA
Print_ISBN :
0-7803-3844-8
DOI :
10.1109/SECON.1997.598630
