Title :
Image segmentation using multilayer neural network
Author :
Malki, Heidar A.
Author_Institution :
Houston Univ., TX, USA
Abstract :
An image segmentation method based on multilayer neural networks is introduced. This method was applied to both the aerial and texture images. The performance of this method on a synthetic image is compared to a model based image segmentation technique. Simulation results on both aerial and texture images are provided. A method to extract important features of the image before the training is proposed
Keywords :
feedforward neural nets; image segmentation; image texture; aerial images; backpropagation; feature extraction; image segmentation; multilayer neural networks; synthetic image; texture images; Costs; Feature extraction; Formal languages; Image segmentation; Image texture analysis; Multi-layer neural network; Neural networks; Parameter estimation; Robustness; Signal detection;
Conference_Titel :
Neural Networks, 1992. IJCNN., International Joint Conference on
Conference_Location :
Baltimore, MD
Print_ISBN :
0-7803-0559-0
DOI :
10.1109/IJCNN.1992.227319
