计算机科学技术学报 ›› 2019,Vol. 34 ›› Issue (1): 35-46.doi: 10.1007/s11390-019-1897-9

所属专题： Artificial Intelligence and Pattern Recognition

• • 上一篇    下一篇

基于高阶衍生的均值CLBP甲状腺MR图像纹理特征提取

Zhe Liu1, Member, CCF, Cheng-Jian Qiu1, Yu-Qing Song1, Member, CCF, Xiao-Hong Liu1, Juan Wang1, and Victor S. Sheng2,*, Senior Member, IEEE, Member, ACM   

  1. 1 School of Computer Science and Telecommunication, Jiangsu University, Zhenjiang 212013, China;
    2 Department of Computer Science, University of Central Arkansas, Arkansas 72035, U.S.A.
  • 收稿日期:2018-07-12 修回日期:2018-12-18 出版日期:2019-01-05 发布日期:2019-01-12
  • 通讯作者: Victor S. Sheng E-mail:ssheng@uca.edu
  • 作者简介:Zhe Liu got her Ph.D. degree in computer science in 2012 from Jiangsu University, Zhenjiang. She is a visiting scholar of the Department of Radiology at University of Pittsburgh Medical Center, Pennsylvania, and also an associate professor at the School of Computer Science and Telecommunication, Jiangsu University, Zhenjiang. Her research interests include image processing, data mining and pattern recognition. She is a member of CCF.
  • 基金资助:
    This work was supported by the National Natural Science Foundation of China under Grant Nos. 61728205, 61772242, 61402204, and 61572239, and the Natural Science Foundation of Jiangsu Province of China under Grant No. BK20130529, the Research Fund for Advanced Talents of Jiangsu University of China under Grant No. 14JDG141, the Science and Technology Project of Zhenjiang City of China under Grant No. SH20140110, the Special Software Development Foundation of Zhenjiang City of China under Grant No. 201322, and the Science and Technology Support Foundation of Zhenjiang City (Industrial) under Grant No. GY2014013.

Texture Feature Extraction from Thyroid MR Imaging Using High-Order Derived Mean CLBP

Zhe Liu1, Member, CCF, Cheng-Jian Qiu1, Yu-Qing Song1, Member, CCF, Xiao-Hong Liu1, Juan Wang1, and Victor S. Sheng2,*, Senior Member, IEEE, Member, ACM   

  1. 1 School of Computer Science and Telecommunication, Jiangsu University, Zhenjiang 212013, China;
    2 Department of Computer Science, University of Central Arkansas, Arkansas 72035, U.S.A.
  • Received:2018-07-12 Revised:2018-12-18 Online:2019-01-05 Published:2019-01-12
  • Contact: Victor S. Sheng E-mail:ssheng@uca.edu
  • About author:Zhe Liu got her Ph.D. degree in computer science in 2012 from Jiangsu University, Zhenjiang. She is a visiting scholar of the Department of Radiology at University of Pittsburgh Medical Center, Pennsylvania, and also an associate professor at the School of Computer Science and Telecommunication, Jiangsu University, Zhenjiang. Her research interests include image processing, data mining and pattern recognition. She is a member of CCF.
  • Supported by:
    This work was supported by the National Natural Science Foundation of China under Grant Nos. 61728205, 61772242, 61402204, and 61572239, and the Natural Science Foundation of Jiangsu Province of China under Grant No. BK20130529, the Research Fund for Advanced Talents of Jiangsu University of China under Grant No. 14JDG141, the Science and Technology Project of Zhenjiang City of China under Grant No. SH20140110, the Special Software Development Foundation of Zhenjiang City of China under Grant No. 201322, and the Science and Technology Support Foundation of Zhenjiang City (Industrial) under Grant No. GY2014013.

PDF (PC)

465

补充材料

1

在医学图像领域，针对传统局部二值模式（LBP）及其改进算法对噪声敏感，仅利用单一的局部差分符号信息，且其二值量化方法过于简化了局部纹理信息，忽视了高阶方向邻域像素和邻域采样点之间的凹凸信息，从而导致提取纹理信息不充分等多种问题。为此，提出一种基于高阶衍生的均值完全局部二值模式（DM_CLBP）改进算法。首先利用矩形区域块的均值灰度值代替单个像素点的灰度值，然后采用二阶差分法获得邻域的像素值差。参照完全局部二值模式（CLBP）的算法思想，级联符号和幅值两个分量并采用均匀模式进行编码和重组。实验结果表明，所提方法高阶DM_CLBM描述子对数据集的分类准确率达到94.4%，与LBP及其改进算法相比，本研究提出的算法有效地区分甲状腺MR图像的病变区域和正常区域，提高了分类的精确率。

关键词: 甲状腺MR图像, 局部二值模式, 纹理特征, DM_CLBP

Abstract: In the field of medical imaging, the traditional local binary pattern (LBP) and its improved algorithms are often sensitive to noise. Traditional LBPs are solely based on the signal information from local differences, and the binary quantization method oversimplifies the local texture features while disregarding the imaging information from the concaveconvex regions between the high-order pixels and the neighboring sampling points. Therefore, we propose an improved Derived Mean Complete Local Binary Pattern (DM_CLBP) algorithm based on high-order derivatives. In the DM_CLBP method, the grey value of a single pixel is replaced by the mean grey value of the rectangular area block, and the difference between pixel values in the area is obtained using the second-order differentiation method. Based on the calculation concept of the complete local binary pattern (CLBP) algorithm, the cascade signs and magnitudes of the two components are encoded and recombined in DM_CLBP using a uniform pattern. The results from the experiments showed that the proposed DM_CLBP descriptors achieved a classification accuracy of 94.4%. Compared with LBP and other improved algorithms, the DM_CLBP algorithm presented in this study can effectively differentiate between lesion areas and normal areas in thyroid MR (magnetic resonance) images and shows the improved accuracy of area classification.

Key words: thyroid magnetic resonance imaging (MRI), local binary pattern, texture feature, complete local binary pattern (CLBP)

[1] Paes J E, Hua K D, Nagy R et al. The relationship between body mass index and thyroid cancer pathology features and outcomes:A clinicopathological cohort study. The Journal of Clinical Endocrinology & Metabolism, 2010, 95(9):4244-4250.
[2] Zhang W, Edwards A, Flemington E K et al. Inferring polymorphism-induced regulatory gene networks active in human lymphocyte cell lines by weighted linear mixed model analysis of multiple RNA-Seq datasets. PLOS One, 2013, 8(10):Article No. e78868.
[3] Wu J, Mazur T R, Ruan S et al. A deep Boltzmann machine-driven level set method for heart motion tracking using cine MRI images. Medical Image Analysis, 2018, 47:68-80.
[4] Shin Y G, Yoo J, Kwon H J et al. Histogram and gray level co-occurrence matrix on gray-scale ultrasound images for diagnosing lymphocytic thyroiditis. Computers in Biology & Medicine, 2016, 75:257-266.
[5] Nugroho H A, Rahmawaty M, Triyani Y et al. Texture analysis for classification of thyroid ultrasound images. In Proc. International Electronics Symposium, September 2016, pp.476-480.
[6] Liang K B, Guan Y H, Luo Y T. A brain MR image segmentation method based on Gaussian model and Markov random field. In Proc. IEEE Advanced Information Management, Communicates, Electronic and Automation Control, Sept. 2017, pp.2042-2048.
[7] Wang Q Y, Li Q N, Mi R et al. Radiomics nomogram building from multiparametric MRI to predict grade in patients with glioma:A cohort study. Journal of Magnetic Resonance Imaging, 2018. https://onlinelibrary.wiley.com/doi/10.1002/jmri.26265, Dec. 2018.
[8] Zhou Y X, Wu Q, Liang J Y et al. Facial expression recognition based on discriminative CLBP. Computer Engineering and Applications, 2017, 53(4):163-169. (in Chinese)
[9] Guo Z, Zhang L, Zhang D. A completed modeling of local binary pattern operator for texture classification. IEEE Transactions on Image Processing, 2010, 19(6):1657-1663.
[10] Liu L, Xie Y X, Wei Y M et al. A review of local binary pattern methods. Journal of Image and Graphics, 2014, 19(12):1696-1720. (in Chinese)
[11] Ojala T, Pietikäinen M, Harwood D. Performance evaluation of texture measures with classification based on Kullback discrimination of distributions. In Proc. the 12th International Conference on Pattern Recognition, October 1994, pp.582-585.
[12] Liao S, Chung A C S. Face recognition by using elongated local binary patterns with average maximum distance gradient magnitude. In Proc. the 8th Asian Conference on Computer Vision, November 2007, pp.672-679.
[13] Sun X, Wang J, Chen R et al. Multi-scale local pattern cooccurrence matrix for textural image classification. In Proc. International Joint Conference on Neural Networks, June 2012, Article No. 12.
[14] Juliano A F, Cunnane M B. Benign conditions of the thyroid gland. Seminars Ultrasound, CT and MRI, 2012, 33(2):130-137.
[15] Peng W, Liu C, Xia S et al. A novel computational CT image analysis method for classifying nodules from normal thyroid tissue. In Proc. IET International Conference on Biomedical Image and Signal Processing, November 2015, Article No. 3.
[16] Nugroho H A, Rahmawaty M, Triyani Y et al. Texture analysis for classification of thyroid ultrasound images. In Proc. International Electronics Symposium, September 2016, pp.476-480.
[17] Yuan F. Rotation and scale invariant local binary pattern based on high order directional derivatives for texture classification. Digital Signal Processing, 2014, 26:142-152.
[18] Tabatabaei S M, Chalechale A. Using DLBP texture descriptors and SVM for Down syndrome recognition. In Proc. the 4th International Conference on Computer and Knowledge Engineering, October 2014, pp.554-558.
[19] Guo Y, Zhao G, Pietikäinen M. Discriminative features for texture description. Pattern Recognition, 2012, 45(10):3834-3843.
[20] Bragesteiro L D, Oliveira L F, Weingaertner D. Blockwise classification of lung patterns in unsegmented CT images. In Proc. the 25th IEEE International Symposium on Computer-Based Medical Systems, June 2015, pp.177-182.
[21] Guo Z, Zhang L, Zhang D. A completed modeling of local binary pattern operator for texture classification. IEEE Trans. Image Process, 2010, 19(6):1657-1663.
[22] Zhao Q, Okada K, Rosenbaum K et al. Digital facial dysmorphology for genetic screening:Hierarchical constrained local model using ICA. Medical Image Analysis, 2014, 18(5):699-710.
No related articles found!
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
No Suggested Reading articles found!
版权所有 © 《计算机科学技术学报》编辑部
本系统由北京玛格泰克科技发展有限公司设计开发 技术支持：support@magtech.com.cn
总访问量：