Topic
Hybrid neural network
About: Hybrid neural network is a research topic. Over the lifetime, 1305 publications have been published within this topic receiving 18223 citations.
Papers published on a yearly basis
Papers
More filters
••
20 Sep 2001
TL;DR: A hybrid neural network system for the recognition of handwritten character using SOFM,BP and Fuzzy network is presented and the recognition rate is improved visibly.
Abstract: A hybrid neural network system for the recognition of handwritten character using SOFM,BP and Fuzzy network is presented. The horizontal and vertical project of preprocessed character and 4_directional edge project are used as feature vectors. In order to improve the recognition effect, the GAT algorithm is applied. Through the hybrid neural network system, the recognition rate is improved visibly.
1 citations
01 Jan 1998
1 citations
••
12 Jul 1993TL;DR: A flexible system has been set up for image processing, with prototype liquid crystal input, BGO filter and photothermoplastic programmable memory, and the final system will implement a neural network classifier.
Abstract: A flexible system has been set up for image processing, with prototype liquid crystal input, BGO filter and photothermoplastic programmable memory. First experiments utilize fixed references for invariant pattern recognition. The final system will implement a neural network classifier.© (1993) COPYRIGHT SPIE--The International Society for Optical Engineering. Downloading of the abstract is permitted for personal use only.
1 citations
••
19 Apr 2004TL;DR: Experimental results illustrate the effectiveness of the new hybrid neural network OMART2-FAM; hence the justification for its implementation in a speech recognition system of Arabic commands is to maximize generalization and minimize misclassification error rates.
Abstract: A hybrid neural network architecture called OMART2-FAM is introduced. It consists of two neural networks connected by an intermediate memory. Optimized match adaptive resonance theory (OMART2) neural network and fuzzy associative memory (FAM) neural network are used for Arabic phoneme signals and Arabic word signals recognition respectively. The intermediate memory is a feedforward field, which retains and encodes the sequence of recognized phonemes at F2 output field of OMART2. Implementing complement coding normalizes connections between the intermediate memory and the input field of FAM. OMART2-FAM classifier of Arabic Internet navigator command signals is implemented. Experimental results show that the new algorithm of OMART2 generally exhibits faster learning and better clustering performance. Additionally, they illustrate the effectiveness of the new hybrid neural network OMART2-FAM; hence the justification for its implementation in a speech recognition system of Arabic commands is to maximize generalization and minimize misclassification error rates.
1 citations
•
28 Mar 2019
TL;DR: In this article, a three-dimensional convolutional self encoder and a hybrid neural network architecture of a total coupling recursive processing unit are adapted for planning an operation procedure for a deformable object.
Abstract: To provide an object operation method generation system whose load of calculation is small and results can be output fast.SOLUTION: A system is proposed for planning an operation procedure for a deformable object. As an approach, a three-dimensional convolutional self encoder in which objective behavior is learned and a hybrid neural network architecture of a total coupling recursive processing unit are adapted. When a start state and a target state are given to the system, a network generates an operation procedure using error backward propagation regarding input operation in the total coupling recursive processing unit.SELECTED DRAWING: Figure 2
1 citations