Image Segmentation Based on Random Neural Network Model and Gabor Filters

2005 ◽  
Author(s):  
Rong Lu ◽  
Yi Shen
Keyword(s):  
Neural Network ◽  
Image Segmentation ◽  
Network Model ◽  
Neural Network Model ◽  
Gabor Filters ◽  
Random Neural Network

Research on Underwater Polarization Image Segmentation Inspired by Biological Optic Nerve

2013 ◽  
Vol 347-350 ◽  
pp. 2178-2184
Author(s):  
Hui Bin Wang ◽  
Yu Rong Wu ◽  
Jie Shen ◽  
Zhe Chen
Keyword(s):  
Neural Network ◽  
Image Segmentation ◽  
Network Model ◽  
Neural Network Model ◽  
Underwater Image ◽  
The Neural Network ◽  
Intensity Information ◽  
Segmentation Methods ◽  
Polarization Image ◽  
Feedback Neural Network

Due to effects of the light by water and other particles, the quality of underwater image will degrade. The traditional underwater image segmentation methods based on intensity and spectrum have difficulty in determining boundary. Inspired by the visual system of mantis shrimps, this paper constructed a feedback neural network model, in which the parameters were optimized using machine learning method. Based on this model, we combine the polarization and intensity information to achieve the underwater polarization image segmentation. The results of experiment prove that the neural network model designed in this paper can improve the accuracy of underwater image segmentation.

THE RANDOM NEURAL NETWORK MODEL FOR TEXTURE GENERATION

1992 ◽  
Vol 06 (01) ◽  
pp. 131-141 ◽  
Author(s):  
VOLKAN ATALAY ◽  
EROL GELENBE ◽  
NESE YALABIK
Keyword(s):  
Neural Network ◽  
Network Model ◽  
Neural Network Model ◽  
Markov Random Fields ◽  
Random Network ◽  
Computational Cost ◽  
Image Synthesis ◽  
Field Equations ◽  
Random Neural Network ◽  
Texture Generation

The generation of artifical textures is a useful function in image synthesis systems. The purpose of this paper is to describe the use of the random neural network (RN) model developed by Gelenbe to generate various textures having different characteristics. An eight parameter model, based on a choice of the local interaction parameters between neighbouring neurons in the plane, is proposed. Numerical iterations of the field equations of the neural network model, starting with a randomly generated gray-level image, are shown to produce textures having different desirable features such as granularity, inclination, and randomness. The experimental evaluation shows that the random network provides good results, at a computational cost less than that of other approaches such as Markov random fields. Various examples of textures generated by our method are presented.

Sign in / Sign up

Export Citation Format

Share Document