计算机与现代化 ›› 2009, Vol. 1 ›› Issue (12): 92-94.doi: 10.3969/j.issn.1006-2475.2009.12.025

• 图像处理 • 上一篇    下一篇

基于BP神经网络的畸变图像校正方法研究

左 欣1,钟 诚2,徐 丹1
  

  1. 1.江苏科技大学计算机科学与工程学院,江苏 镇江 212003; 2.镇江舰艇学院,江苏 镇江 212003
  • 收稿日期:2009-07-15 修回日期:1900-01-01 出版日期:2009-11-27 发布日期:2009-11-27

Research on Image Distortion Correction Method Based on BP Neural Networks

ZUO Xin1, ZHONG Cheng2, XU Dan1   

  1. 1. School of Computer Science and Engineering, Jiangsu University of Science and Technology, Zhenjiang 212003, China;2. Zhenjiang Watercraft College, Zhenjiang 212003, China
  • Received:2009-07-15 Revised:1900-01-01 Online:2009-11-27 Published:2009-11-27

摘要: 针对传统的非线性畸变图像校正方法存在模型建立较难等缺点,本文采用基于神经网络畸变图像校正方法,首先将免疫学中的克隆选择算法引入量子遗传算法,然后将这种混合的量子遗传算法和BP算法有机结合,优势互补作为神经网络的学习算法用于畸变图像的校正。实验结果表明该方法具有很高的精度。

关键词: BP神经网络, 畸变图像, 畸变校正, 量子遗传算法, 克隆选择

Abstract: Due to the shortcomings of traditional methods, such as building model difficultly, the paper uses the distortion correction method based on neural networks. Firstly, it introduces the clonal selection algorithm of immunology into quantum genetic algorithm, then proposes a method which combines the HQGA and BP algorithm to supplement mutually as neural networks study algorithm applied in distorted image correction. The experiment results show that this method has high accuracy.

Key words: BP neural network, distorted image, distortion correction, quantum genetic algorithm, clonal selection

中图分类号: