We present a new scheme to increase the performance of edge-preserving imag
e smoothing algorithm from the parameter tuning of a Markov random field (M
RF) function. This method is based on an automatic control of the image-smo
othing strength in the MRF function in which the parameter control of the f
unction depends on edge magnitude resulted from discontinuities of image in
tensity. Without any binary decision for the edge magnitude, direct use of
full edge magnitude could improve the performance of discontinuity-preservi
ng image smoothing technique. (C) 2000 Pattern Recognition Society. Publish
ed by Elsevier Science Ltd. All rights reserved.