Morphological Associative Memories Based on Fuzzy Sets
-
- Fukumoto Shinya
- Faculty of Engineering, Kagoshima University
-
- Miyajima Hiromi
- Faculty of Engineering, Kagoshima University
-
- Murashima Sadayuki
- Faculty of Engineering, Kagoshima University
-
- Otsuka Sakuichi
- Faculty of Engineering, Kagoshima University
Bibliographic Information
- Other Title
-
- ファジィ集合を用いた形態学的連想記憶について
Abstract
Research on associative memory is attracting attention. However, we were beginning to feel the limits of the performance of the conventional associative memory that is typified by Hopfield's model. A morphological associative memory is expected to be superior in performance, because there is no limit to the memory capacity as the associative memory. Ritter proposed morphological associative memories represented by W and M. The memory W is robust under erosive noise and M is robust under dilative noise. However, these memories are not robust under general noise that includes both erosive and dilative noises. We present a morphological associative memory that is applied the idea of the fuzzy sets to the calculation process. In this method, we need to find nucleuses of the learning patterns called kernel. Through numerical experiments, we examine the relationships between the number of kernel bits and recalling rate of the learning patterns.
Journal
-
- Proceedings of the Fuzzy System Symposium
-
Proceedings of the Fuzzy System Symposium 24 (0), 51-51, 2008
Japan Society for Fuzzy Theory and Intelligent Informatics
- Tweet
Details 詳細情報について
-
- CRID
- 1390282680648238976
-
- NII Article ID
- 130005035284
-
- Text Lang
- ja
-
- Data Source
-
- JaLC
- CiNii Articles
-
- Abstract License Flag
- Disallowed