计算机工程 ›› 2013, Vol. 39 ›› Issue (6): 272-276.doi: 10.3969/j.issn.1000-3428.2013.06.060

• 图形图像处理 • 上一篇    下一篇

基于多方向复支持度变换的多聚焦图像融合

胡德恒,张 澜,尚赵伟   

  1. (重庆大学计算机学院,重庆 400030)
  • 收稿日期:2012-07-23 出版日期:2013-06-15 发布日期:2013-06-14
  • 作者简介:胡德恒(1987-),男,硕士研究生,主研方向:图像处理,模式识别;张 澜,硕士研究生;尚赵伟,副教授
  • 基金项目:
    国家自然科学基金资助项目(61173130);重庆市自然科学基金资助项目(CSTC2010BB2217)

Multi-focus Image Fusion Based on Multi-directional Complex Support Value Transform

HU De-heng, ZHANG Lan, SHANG Zhao-wei   

  1. (College of Computer Science, University of Chongqing, Chongqing 400030, China)
  • Received:2012-07-23 Online:2013-06-15 Published:2013-06-14

摘要: 针对支持度变换应用于多聚焦图像融合存在的不足,提出复方向滤波器组与支持度变换相结合的图像融合方法。采用支持度变换对图像进行多尺度分解以提取高频和低频分量,运用复方向滤波器组对各级高频分量进行分解,提取各方向图像信息,对源图像各对应频带的分解信息按照规则融合,并重构得到融合图像。实验结果表明,该方法不仅可取得较好的主观融合效果,而且客观融合指标也优于支持度变换方法。

关键词: 支持度变换, 复方向滤波器组, 图像融合, 多聚焦, 多尺度, 支持向量机

Abstract: With drawbacks of multi-focus image fusion method using Support Value Transform(SVT), a new image fusion method using the combination of SVT and Complex Directional Filter Banks(CDFB) is proposed. It obtains both high and low components of source image by SVT. CDFB is applied to all high components to extract directional information of image. Then a fusion operation, incorporating certain rules, is processed with corresponding directional information on every subbands respectively. The fusion image is reconstructed with the fused high and low components. Experimental result shows that this method not only gains pleasing fusion result, but also reserves better objective evaluation indexes when compared with SVT method.

Key words: Support Value Transform(SVT), Complex Directional Filter Banks(CDFB), image fusion, multi-focus, muti-scale, Support Vector Machine(SVM)

中图分类号: