Fast digital locally monotonic regression

Locally monotonic regression is the optimal counterpart of iterated median filtering. In a previous paper, Restrepo and Bovik (see ibid., vol.41, no.9, p.2796-2810, 1993) developed an elegant mathematical framework in which they studied locally monotonic regressions in R/sup N/. The drawback is that the complexity of their algorithms is exponential in N. We consider digital locally monotonic regressions, in which the output symbols are drawn from a finite alphabet and, by making a connection to Viterbi decoding, provide a fast O(|A|/sup 2//spl alpha/N) algorithm that computes any such regression, where |A| is the size of the digital output alphabet, a stands for lomo degree, and N is the sample size. This is linear in N, and it renders the technique applicable in practice.

[1]  H.-L. Lou,et al.  Implementing the Viterbi algorithm , 1995, IEEE Signal Process. Mag..

[2]  Jacobus P. H. Wessels,et al.  The art and theory of dynamic programming , 1977 .

[3]  Scott T. Acton,et al.  Nonlinear image estimation using piecewise and local image models , 1998, IEEE Trans. Image Process..

[4]  Thomas S. Huang,et al.  A generalization of median filtering using linear combinations of order statistics , 1983 .

[5]  Nikos D. Sidiropoulos,et al.  The Viterbi optimal runlength-constrained approximation nonlinear filter , 1996, IEEE Trans. Signal Process..

[6]  Keshab K. Parhi.  High-speed VLSI architectures for Huffman and Viterbi decoders , 1992 .

[7]  R. Bellman.  Dynamic programming. , 1957, Science.

[8]  Shuji Kubota,et al.  Novel Viterbi decoder VLSI implementation and its performance , 1993, IEEE Trans. Commun..

[9]  P. Glenn Gulak,et al.  A multiprocessor architecture for Viterbi decoders with linear speedup , 1993, IEEE Trans. Signal Process..

[10]  Andrew J. Viterbi,et al.  Error bounds for convolutional codes and an asymptotically optimum decoding algorithm , 1967, IEEE Trans. Inf. Theory.

[11]  N. Gallagher.  Median filters: a tutorial , 1988, 1988., IEEE International Symposium on Circuits and Systems.

[12]  P. B. Coaker,et al.  Applied Dynamic Programming , 1964 .

[13]  P. Glenn Gulak,et al.  VLSI Structures for Viterbi Receivers: Part I-General Theory and Applications , 1986, IEEE J. Sel. Areas Commun..

[14]  T. Nodes,et al.  Median filters: Some modifications and their properties , 1982 .

[15]  G. Wise,et al.  A theoretical analysis of the properties of median filters , 1981 .

[16]  J. Omura,et al.  On the Viterbi decoding algorithm , 1969, IEEE Trans. Inf. Theory.

[17]  H. Heijmans.  Morphological image operators , 1994 .

[18]  Alfredo Restrepo,et al.  On the statistical optimality of locally monotonic regression , 1994, IEEE Trans. Signal Process..

[19]  Thomas S. Huang,et al.  The Effect of Median Filtering on Edge Estimation and Detection , 1987, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[20]  B I Justusson,et al.  Median Filtering: Statistical Properties , 1981 .

[21]  S. G. Tyan,et al.  Median Filtering: Deterministic Properties , 1981 .

[22]  P.K. Rajan,et al.  Two-dimensional digital signal processing II: Transforms and median filters , 1982, Proceedings of the IEEE.

[23]  J.E. Mazo,et al.  Digital communications , 1985, Proceedings of the IEEE.

[24]  Trieu-Kien Truong,et al.  A VLSI design for a trace-back Viterbi decoder , 1992, IEEE Trans. Commun..

[25]  Alan C. Bovik,et al.  Statistical optimality of locally monotonic regression , 1990, Other Conferences.

[26]  Alfredo Restrepo,et al.  Locally monotonic regression , 1993, IEEE Trans. Signal Process..