Medical image fusion method based on guided filter

被引:2
|
作者
Guo Pan [1 ,2 ]
He Wen-chao [1 ,2 ]
Liang Long-kai [1 ,2 ]
Zhang Meng [1 ,2 ]
Lyu Xu-hao [1 ,2 ]
Gong Xin [1 ,2 ]
机构
[1] Northeast Normal Univ, Coll Informat Technol & Engn, Coll Humanities & Sci, Changchun 130117, Jilin, Peoples R China
[2] Automobile Elect Technol & Engn Res Ctr Jilin Pro, Changchun 130117, Jilin, Peoples R China
关键词
medical image; image fusion; guiding filter; spatial continuity; neighborhood statistical characteristics;
D O I
10.3788/YJYXS20193406.0605
中图分类号
O7 [晶体学];
学科分类号
0702 ; 070205 ; 0703 ; 080501 ;
摘要
Aiming at the problem that artificial texture is easy to be produced in the process of medical image fusion, the guiding filter was applied to medical image fusion to effectively improve the spatial continuity of fusion image and reduce the generation of artificial texture. Firstly, the detail layer information of the image was obtained by using the detail enhancement characteristic of the guide filter. The basic layer image information was obtained by filtering the original image through the guide filter, and then the detail layer image information could be obtained by the difference operation between the original image and the guide filter. Then, the image fusion weight coefficient was obtained based on the neighborhood statistical characteristics fusion rule with variable parameter p. The fusion rule added variable parameter p to the original neighborhood statistical characteristics, which could effectively enhance the details of the fusion image. Finally, the original image was fused according to the obtained image fusion weight coefficient to obtain the final fusion image. The comparison and analysis of experimental simulation results indicate that the fusion effect is significantly improved with the change of variable parameter p. The fusion effect is basically stable after p > 10. This fusion algorithm can effectively realize medical image fusion and has certain advantages over other fusion methods.
引用
收藏
页码:605 / 612
页数:8
相关论文
共 18 条
  • [1] Fusion of MRI and CT images using guided image filter and image statistics
    Bavirisetti, Durga Prasad
    Kollu, Vijayakumar
    Gang, Xiao
    Dhuli, Ravindra
    [J]. INTERNATIONAL JOURNAL OF IMAGING SYSTEMS AND TECHNOLOGY, 2017, 27 (03) : 227 - 237
  • [2] The nonsubsampled contourlet transform: Theory, design, and applications
    da Cunha, Arthur L.
    Zhou, Jianping
    Do, Minh N.
    [J]. IEEE TRANSACTIONS ON IMAGE PROCESSING, 2006, 15 (10) : 3089 - 3101
  • [3] [方静 Fang Jing], 2017, [激光与红外, Laser and Infrared], V47, P914
  • [4] [葛雯 Ge Wen], 2016, [光电工程, Opto-Electronic Engineering], V43, P73
  • [5] Guided Image Filtering
    He, Kaiming
    Sun, Jian
    Tang, Xiaoou
    [J]. IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2013, 35 (06) : 1397 - 1409
  • [6] Learning the parts of objects by non-negative matrix factorization
    Lee, DD
    Seung, HS
    [J]. NATURE, 1999, 401 (6755) : 788 - 791
  • [7] Lee DD, 2001, ADV NEUR IN, V13, P556
  • [8] Image fusion based on complex-shearlet domain with guided filtering
    Liu, Shuaiqi
    Shi, Mingzhu
    Zhu, Zhihui
    Zhao, Jie
    [J]. MULTIDIMENSIONAL SYSTEMS AND SIGNAL PROCESSING, 2017, 28 (01) : 207 - 224
  • [9] [刘帅奇 Liu Shuaiqi], 2016, [信号处理, Journal of Signal Processing], V32, P276
  • [10] LIU SQ, 2018, ANAL APPL ALGORITHM, P4