Computer Engineering and Applications ›› 2010, Vol. 46 ›› Issue (9): 166-168.DOI: 10.3778/j.issn.1002-8331.2010.09.047

• 图形、图像、模式识别 • Previous Articles     Next Articles

Appling region variance and weighted mean to image fusion of curvelet transform

QIU Xuan1,ZHOU Ze-ming2,HU You-bin2   

  1. 1.China Satellite Marinetime Tracking & Control Department,Jiangyin,Jiangsu 214431,China
    2.Meteorology College of PLA University of Science and Technology,Nanjing 211101,China
  • Received:2009-02-16 Revised:2009-04-17 Online:2010-03-21 Published:2010-03-21
  • Contact: QIU Xuan

应用邻域方差加权平均的curvelet变换融合

邱 宣1,周则明2,胡友彬2   

  1. 1.中国卫星海上测控部,江苏 江阴 214431
    2.解放军理工大学 气象学院,南京 211101
  • 通讯作者: 邱 宣

Abstract: An efficient method of image fusion based on weighted mean and neighboring region variance of curvelet transform is proposed.Firstly,the multi-spectral and high resolution image is decomposed by curvelet transform.Then the curvelet coefficients of the fused image can be obtained by using different rules.The approximate coefficients come from the multispectral image,and the detail coefficient is based on the neighboring region variance.Finally,the fused image is obtained by inverse curvelet transform.The experiment results show that the method can improve the average gradient and preserve spectrum index commendably.

Key words: image fusion, curvelet transform, neighboring region variance

摘要: 提出一种基于邻域方差加权平均的多源遥感图像曲波变换融合方法。将低分辨率的多光谱图像和高分辨率的全色图像作曲波变换,融合图像的曲波系数中的低频分量取多光谱图像的低频分量,中高频系数与高频系数采用邻域方差加权平均的方法由两幅图像曲波变换系数共同决定。逆曲波变换得到融合图像。实验表明算法有效提高了图像的清晰度和保持了光谱特性。

关键词: 图像融合, 曲波变换, 邻域方差

CLC Number: