计算机工程 ›› 2012, Vol. 38 ›› Issue (21): 229-231,244.doi: 10.3969/j.issn.1000-3428.2012.21.061

• 图形图像处理 • 上一篇    下一篇

一种新的毫米波图像去噪方法?

苏品刚1,2, 尚 丽1,3,陈 杰1,颜廷秦1,4   

  1. (1. 苏州市职业大学电子信息工程系,江苏 苏州 215104;2. 东南大学毫米波国家重点实验室,南京 210096; 3. 中国科学技术大学信息科学与技术学院自动化系,合肥 230026;4. 苏州市数字化设计与制造技术重点实验室,江苏 苏州 215104)
  • 收稿日期:2011-12-07 出版日期:2012-11-05 发布日期:2012-11-02
  • 作者简介:苏品刚(1971-),男,副教授、硕士,主研方向:毫米波焦平面成像技术;尚 丽,副教授、博士;陈 杰,讲师、硕士; 颜廷秦,副教授、硕士
  • 基金项目:
    国家自然科学基金资助项目(60970058);江苏省“青蓝工程”基金资助项目;苏州市科技基础设施建设计划基金资助项目(SZS201 009);苏州市职业大学创新团队基金资助项目(3100125)

A Novel Millimeter Wave Image Denoising Method

SU Pin-gang 1,2, SHANG Li 1,3, CHEN Jie 1, YAN Ting-qin 1,4   

  1. (1. Department of Electronic Information Engineering, Suzhou Vocational University, Suzhou 215104, China; 2. State Key Laboratory of Millimeter Wave, Southeast University, Nanjing 210096, China; 3. Department of Automation, Institute of Information Science & Technology, University of Science and Technology of China, Hefei 230026, 4. Suzhou Key Lab of Digital Design & Manufacturing Technology, Suzhou 215104, China)
  • Received:2011-12-07 Online:2012-11-05 Published:2012-11-02

摘要: 毫米波(MMW)焦平面成像系统得到的图像质量较差,为此,结合基于人类视觉系统(HVS)的自适应中值滤波和轮廓波变换,提出一种新的MMW图像去噪方法。利用基于HVS的自适应中值滤波方法实现噪声的自适应检测和滤波,滤除脉冲噪声。根据轮廓波分解的方向性和能量变化特性对中值滤波结果进行变换,并对得到高频子带做阈值去噪处理,以最大限度保留图像轮廓。实验结果表明,与传统方法相比,该方法所得图像的峰值信噪比较高。

关键词: 毫米波图像, 人类视觉系统, 自适应中值滤波, 轮廓波变换, 图像消噪

Abstract: Aiming at the problem that the image generated by Millimeter Wave(MMW) focal plane imaging system is contaminated by much unknown noise and has lower resolution, a novel image denoising method, combined with Human Visual System(HVS) based self-adaptive median filtering and contourlet transformation, is proposed in this paper. The noise is self-adaptively detected and filtered by using the method of HVS-based adaptive median filtering, and pulse noise is filtered efficiently. Utilizing the contourlet decomposition orientation and the energy variation to transform the median filtering result, the high frequency sub-bands are obtained, and they are further denoised by the threshold, which retains more image contour in degree. Experimental results show that that the proposed method has better denoising performance than other methods.

Key words: Millimeter Wave(MMW) image, Human Visual System(HVS), self-adaptive median filtering, contourlet transformation, image denoising

中图分类号: