期刊文献+

基于偏微分方程和Gabor算法的图像细节特征保护方法 被引量:2

Methods of image detail characteristics protection based on partitial differential equation and Gabor algorithm
下载PDF
导出
摘要 传统的数字图像恢复算法在图像恢复过程中会抹掉图像的局部特征如纹理或者细节特征,从而可能降低图像的可读性和可理解性。利用偏微分方程(PDE)和Gabor(PG算法)结合运用于图像恢复,不但能保护图像的边缘细节,而且还能有效地保护图像的局部特征,如纹理、有意义的小细节等反映图像重要信息的特征。将Gabor过滤器的图像特征识别特性融入PDE的图像恢复算法之中,对图像细节特征进行加权保护,再利用较简洁的基于局部能量残余的图像分区机制来控制图像光滑的强度,从而有效地解决传统PDE算法的"盲目性"。 Traditional digital image restoration algorithm in image restoration process would erase the local features of the image such as texture or detail characteristics, which may reduce the image”s readability and understandability. This article introduced the image restoration based on PDE and Gabor ( PG algorithm ) , which not only could protect the image edge details, but also could effectively protect the local characteristics of the image, and reflected the characteristics of important image information such as texture and meaningful small details. This article adopted image feature recognition characteristics of the Gabor filter into the PDE image restoration algorithm, protecting the characteristics of image detail weightedly, recycling more concise image partition mechanism based on local energy residual to control the intensity of the image smoothness, and solving the traditional "blindness" of the PDE algorithm effectively.
出处 《计算机应用》 CSCD 北大核心 2015年第A01期255-257,261,共4页 journal of Computer Applications
关键词 图像恢复 细节保护 偏微分方程变分 GABOR小波 image restoration detail protection Partial Differential Equation ( PDE) variation Gabor wavelet
  • 相关文献

参考文献8

  • 1BOV1K A C, CLARK M, GEISLER W S. Multichannel texture a- nalysis using localized spatial filters[ J]. IEEE Transactions on Pat- tern Analysis and Machine Intelligence, 1990, 12(1) : 55 - 73.
  • 2DAUGMAN J G. Two-dimensional spectral analysis of cortical re- ceptive field profiles[ M]. Vision Research, 1980, 20(10) : 847 - 856.
  • 3DAUGMAN J. Uncertainty relation for resolution in space, spatial frequency andorientation optimized by two-dimensional visual corti- cal filters[ J]. Journal of the Optical Society of America: A, 1985, 2(7): 1160-1169.
  • 4DILEEP A D, SEKHAR C C. Selection of non-uniibrmly spaced ori- entations for Gabor fihers using multiple kernel learning[ C]//MLSP 2010: Proceedings of the 2010 IEEE International Workshop on Ma- chine Learning for Signal Processing. Piscataway: IEEE, 2010:415 - 420.
  • 5RUDIN L, OSHER S, FATERNI E. Non|inear total variation based noise removal algorithms[ J]. Physica D, 1992, 60:259 -268.
  • 6AUBERT G, VESE L. A variational method in image recovery[ J]. SIAM Journal of Numerical Analysis, 1997, 34(5) : 1948 - 1979.
  • 7G1LBOA G, SOCHEN N, ZEEVI Y. Texture presetwing variational denoising using an adaptive fidelity term [ C]// ProcVLSM 2003. Piscataway: 1EEE, 2003:137 - 144.
  • 8JAIN A K. Fundamentals of digital image processing[ M]. Upper Saddle River: Prentice-Hall, 1989:23-30.

同被引文献11

引证文献2

二级引证文献6

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部