模式识别与人工智能
2025年4月3日 星期四   首 页     期刊简介     编委会     投稿指南     伦理声明     联系我们                                                                English
模式识别与人工智能  2018, Vol. 31 Issue (11): 1028-1039    DOI: 10.16451/j.cnki.issn1003-6059.201811007
研究与应用 最新目录| 下期目录| 过刊浏览| 高级检索 |
结合深度学习与特征多尺度融合的微钙化簇检测
张新生1, 王哲1
1.西安建筑科技大学管理学院 西学710055
Microcalcification Clusters Detection Method Based on Deep Learning and Multi-scale Feature Fusion
ZHANG Xinsheng1, WANG Zhe1
1.School of Management]Xi′an University of Architecture and Technology, Xi′an 710055

全文: PDF (1066 KB)   HTML (1 KB) 
输出: BibTeX | EndNote (RIS)      
摘要 为了准确识别X线图像中的微钙化簇以进行乳腺癌的辅助诊断与早期预防,结合细粒度级联增强网络(FCE-Net)与多尺度特征融合算法(MFF),提出微钙化簇目标检测方法.首先构建FCE-Net累加卷积模块层级权重,并增强多分支结构,得到细粒度卷积特征图.然后构建MFF候选检测网络,通过二倍上采样融合多尺度特征,得到目标置信度和区域坐标.最后在感兴趣区域池化层分类目标并调整边界框.在MIAS数据集上实验表明,结合FCE-Net与MFF可以提升微小目标的深层特征提取能力,同时增强目标分类与定位的准确度.
服务
把本文推荐给朋友
加入我的书架
加入引用管理器
E-mail Alert
RSS
作者相关文章
张新生
王哲
关键词 目标检测深度学习卷积神经网络多尺度特征融合(MFF)微钙化簇    
Abstract:To accurately identify the microcalcification in X-ray images for diagnosis and early prevention of breast cancer, a microcalcification object detection method combining fine-grained cascading enhancement-network(FCE-Net) and multi-scale feature fusion(MFF]is proposed. Firstly, the residual weights of FCE-Net are accumulated and the multi-branch structure is enhanced within the convolution module to obtain hierarchical and fine-grained convolution plots. Then, a candidate detection network based on MFF is constructed and multi-scale features are merged by double-upsampling. Thus, the confidence and regional coordinates of the object of the microcalcification clusters are acquired. Finally, the object region is classified and the bounding box is adjusted in the pooling layer of the region of interest. The experimental results on MIAS breast cancer dataset show that the method combining FCE-Net and MFF has a better ability to extract deep feature with enhanced classification and positioning accuracies.
Key wordsObject Detection    Deep Learning    Convolutional Neural Network    Multi-scale Feature Fusion(MFF)    Microcalcification Clusters   
收稿日期: 2018-05-30     
ZTFLH: TP 391  
基金资助:国家自然科学基金面上项目(No.41877527)、陕西省自然科学基金项目(No.2016JM6023)资助
通讯作者: 张新生,博士,教授,主要研究方向为模式识别、智能信息处理等.E-mail:xinsheng.zh@outlook.com.   
作者简介: 王哲,硕士研究生,主要研究方向为模式识别、智能信息处理等.E-mail:xauatWZ@163.com.
引用本文:   
张新生, 王哲. 结合深度学习与特征多尺度融合的微钙化簇检测[J]. 模式识别与人工智能, 2018, 31(11): 1028-1039. ZHANG Xinsheng, WANG Zhe. Microcalcification Clusters Detection Method Based on Deep Learning and Multi-scale Feature Fusion. , 2018, 31(11): 1028-1039.
链接本文:  
http://manu46.magtech.com.cn/Jweb_prai/CN/10.16451/j.cnki.issn1003-6059.201811007      或     http://manu46.magtech.com.cn/Jweb_prai/CN/Y2018/V31/I11/1028
版权所有 © 《模式识别与人工智能》编辑部
地址:安微省合肥市蜀山湖路350号 电话:0551-65591176 传真:0551-65591176 Email:bjb@iim.ac.cn
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