Using Immune Genetic Algorithm to Optimize BP Neural Network and Its Application

DOI

10.2991/cnct-16.2017.86How to use a DOI?

Keywords

Immune genetic algorithm, Neural network, Image recognition

Abstract

It proposes a target classification and recognition method based on immune genetic neural network algorithm. Immune genetic algorithm is adopted to optimize the initial weights and thresholds of BP(Back Propagation)network. Respectively use traditional BP neural network algorithm and immune genetic neural network algorithm to train network, until the convergence error precision of neural network reaches a pre-set requirement. Simulation results show that with the same hidden layer nodes and error precision request, it proposes immune genetic neural network algorithm for image recognition, its effect is better than that of traditional BP algorithm, improves the convergence speed of BP neural network, and reduces the training time. The recognition rate improves to some extent.

Copyright

© 2017, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Peng-fei LIU
AU  - Qun-tai SHEN
AU  - Jun ZHI
PY  - 2016/12
DA  - 2016/12
TI  - Using Immune Genetic Algorithm to Optimize BP Neural Network and Its Application
BT  - Proceedings of the International Conference on Computer Networks and Communication Technology (CNCT 2016)
PB  - Atlantis Press
SP  - 628
EP  - 632
SN  - 2352-538X
UR  - https://doi.org/10.2991/cnct-16.2017.86
DO  - 10.2991/cnct-16.2017.86
ID  - LIU2016/12
ER  -