SIA OpenIR  > 工业控制网络与系统研究室
患者肺部图像潜在病变区域纹理优化识别仿真
Alternative TitleTexture Optimization Simulation of the Potential Lesion Area of the Patient's Lung Image
李杨1,2; 梁炜1; 谈金东3
Department工业控制网络与系统研究室
Source Publication计算机仿真
ISSN1006-9348
2018
Volume35Issue:9Pages:417-420
Contribution Rank1
Funding Organization国家自然科学基金重点项目(61333019)
Keyword肺部图像 潜在病变区域 图像纹理识别
Abstract对患者肺部图像潜在病变区域纹理识别,能够有效提高图像诊断结果精度。对肺部潜在病变区域图像纹理进行优化识别,需要通过图像灰度全局信息初始化水平集,利用拟合函数定义局部能量函数,完成潜在病变区域图像纹理优化识别。传统方法对肺部潜在病变区域图像求解获得图像纹理裂纹长度,提取出纹理灰度等值线,但忽略了对纹理能量函数的定义,导致纹理识别精度偏低。提出基于水平集分割的患者肺部图像潜在病变区域纹理优化识别方法。对患者肺部潜在病变区域图像进行预处理,采用图像梯度方差加权信息熵算法自适应改变滤波器参数。采用图像灰度全局信息初始化水平集,局部能量函数由图像局部灰度拟合函数定义。融合多种纹理特征,将融合结果输入到神经网络的Softmax层进行潜在病变区域图像纹理识别。实验结果表明,所提方法具有准确和鲁棒性好的特点。
Other AbstractTraditional methods ignore the definition of texture energy function, which leads to low accuracy of texture recognition. In this article, we present a method to optimize and recognize the texture in potential lesion region of lung image based on level set segmentation. Firstly, we preprocessed the lung image in potential lesion region of patient and used the image gradient variance weighting information entropy algorithm to change filter parameters adaptively. Then, we used grayscale global information to initialize the level set. Meanwhile, we defined a local energy function by local grayscale fitting function of the image. Integrated with various texture features, we inputted fusion results to the Softmax layer of neural network to recognize the image texture in the potential lesion region. Simulation results prove that the proposed method has good accuracy and robustness.
Language中文
Document Type期刊论文
Identifierhttp://ir.sia.cn/handle/173321/23420
Collection工业控制网络与系统研究室
Corresponding Author李杨
Affiliation1.中国科学院沈阳自动化研究所
2.中国科学院大学
3.田纳西大学
Recommended Citation
GB/T 7714
李杨,梁炜,谈金东. 患者肺部图像潜在病变区域纹理优化识别仿真[J]. 计算机仿真,2018,35(9):417-420.
APA 李杨,梁炜,&谈金东.(2018).患者肺部图像潜在病变区域纹理优化识别仿真.计算机仿真,35(9),417-420.
MLA 李杨,et al."患者肺部图像潜在病变区域纹理优化识别仿真".计算机仿真 35.9(2018):417-420.
Files in This Item: Download All
File Name/Size DocType Version Access License
患者肺部图像潜在病变区域纹理优化识别仿真(515KB)期刊论文出版稿开放获取CC BY-NC-SAView Download
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[李杨]'s Articles
[梁炜]'s Articles
[谈金东]'s Articles
Baidu academic
Similar articles in Baidu academic
[李杨]'s Articles
[梁炜]'s Articles
[谈金东]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[李杨]'s Articles
[梁炜]'s Articles
[谈金东]'s Articles
Terms of Use
No data!
Social Bookmark/Share
File name: 患者肺部图像潜在病变区域纹理优化识别仿真.pdf
Format: Adobe PDF
All comments (0)
No comment.
 

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.