Title :
A remanent noise problem with the median filter
Author_Institution :
Dept. of Phys., R. Holloway & Bedford New Coll., Egham, UK
fDate :
30 Aug-3 Sep 1992
Abstract :
This paper is concerned with the noise suppression capabilities of the median filter. It finds that though the median filter is highly successful at removing noise from both 1-D and 2-D signals without causing undue blurring effects, it is bound to leave some noise on the final signal. In certain cases this noise can be substantial, when compared with the noise left by a mean or Gaussian filter. However, the effect is turned to good use when the median filter is employed in a particular type of (2-D) corner detector. Indeed, the remanent noise of the median filter provides a sound explanation of the excellent properties of this type of corner detector
Keywords :
edge detection; filtering and prediction theory; interference suppression; signal processing; 1D signals; 2D signals; corner detector; median filter; noise suppression capabilities; remanent noise problem; Acoustic noise; Detectors; Educational institutions; Filtering; Gaussian noise; Low pass filters; Machine vision; Noise level; Physics; Signal design;
Conference_Titel :
Pattern Recognition, 1992. Vol.III. Conference C: Image, Speech and Signal Analysis, Proceedings., 11th IAPR International Conference on
Conference_Location :
The Hague
Print_ISBN :
0-8186-2920-7
DOI :
10.1109/ICPR.1992.202035