X Ray Image Enhancement Technology for Steel Pipe Welding Based on Hopfield Neural Network

Abstract

Hopfield neural network is utilized to enhance x-ray image of thick steel pipe welding, and a gray mapping matrix is constructed to replace traditional gray transformation curves and functions in this paper. The maximum dimension of the gray mapping matrix is 256 256, so the calculation time has little relation with the size of the image. The criterion function of image quality is used to evaluate the quality of the transformed image. In proposed approach, the problem of image enhancement is transformed to an optimization problem, so the normalization of gray values for each pixel is not necessary. The energy function that improves the performance of image enhancement is also given for Hopfield neural network.

6 Figures and Tables

Cite this paper

@article{Weixin2009XRI, title={X Ray Image Enhancement Technology for Steel Pipe Welding Based on Hopfield Neural Network}, author={Gao Weixin and Sun Lianmin and Mu Xiangyang and Tang Nan and Wu Xiaomeng}, journal={2009 Second International Symposium on Computational Intelligence and Design}, year={2009}, volume={2}, pages={107-110} }