A hybrid neuro-fuzzy classifier was developed for sea-bottom identification from acoustic echoes acquired at normal incidence by a single beam echosounder. A multistage fuzzy neural network (MSFNN) structure was constructed and tested on the data collected on 38 kHz and 120 kHz echosounde's frequencies. Two MSFNN models termed as incremental fuzzy neural network (IFNN) and aggregated fuzzy neural network (AFNN), were considered. Firstly, in IFNN, an approximate decision is undertaken based only on the one set of input variables. The decision is then fine-tuned by considering more and more factors in following stages until the final decision, corresponding to the output class, is undertaken. In AFNN, the input variables are divided into M subsets, where each of them is fed to one sub-stage. The final output is derived by the reasoning with all intermediate variables, which work as the outputs of sub-stages in the preceding stage. The proposed approach improves classification results twofold. Firstly, it increases generalisation ability of the system by multifrequency observations. Secondly, it reduces requirements on computation power and memory by applying a rnultistage solution.
Sea bottom recognition using multistage fuzzy neural network operating on multi-frequency data
Meeresbodenerkennung mit einem im Mehrfrequenzmodus betriebenen neuronalen Multistufen-Fuzzy-Netzwerk
Acustica ; 86 , 5 ; 830-837
2000
8 Seiten, 15 Bilder, 6 Tabellen, 24 Quellen
Article (Journal)
English
Object Recognition Using Multi-Layer Hopfield Neural Network
British Library Conference Proceedings | 1994
|Traffic sign recognition using weighted multi‐convolutional neural network
Wiley | 2018
|Routing in Optical Multistage Interconnection Networks: A Neural Network Solution
British Library Online Contents | 1995
|