MAP image restoration and segmentation by constrained optimization

被引：12
|
作者：
Li, SZ [1 ]
机构：
[1] Nanyang Technol Univ, Sch Elect & Elect Engn, Singapore 639798, Singapore
来源：
IEEE TRANSACTIONS ON IMAGE PROCESSING | 1998年 / 7卷 / 12期
关键词：
augmented Lagrange method; combinatorial optimization; constrained optimization; graded Hopfield networks; Markov random fields (MRF's); maximum a posteriori;
D O I：
10.1109/83.730386
中图分类号：
TP18 [人工智能理论];
学科分类号：
081104 ; 0812 ; 0835 ; 1405 ;
摘要：
The combinatorial optimization problem of MAP estimation is converted to one of constrained real optimization and then solved by using the proposed augmented Lagrange-Hopfield (ALH) method. The ALH effectively overcomes instabilities that are inherent in the penalty method or the Lagrange multiplier method in constrained optimization. It produces good solutions,vith reasonable costs.
引用
收藏
页码：1730 / 1735
页数：6
相关论文
共 50 条
12345