《计算机应用》唯一官方网站 ›› 2020, Vol. 40 ›› Issue (2): 561-565.DOI: 10.11772/j.issn.1001-9081.2019091641

• 第七届CCF大数据学术会议 • 上一篇    下一篇

基于多尺度卷积特征融合的肺结节图像检索方法

顾军华1,2,3, 王锋2,3, 戚永军1,4(), 孙哲然2,3, 田泽培2,3, 张亚娟2,3   

  1. 1.电工设备可靠性与智能化国家重点实验室 (河北工业大学),天津 300401
    2.河北省大数据计算重点实验室 (河北工业大学),天津 300401
    3.河北工业大学 人工智能与数据科学学院,天津 300401
    4.北华航天工业学院 信息技术中心,河北 廊坊 065000
  • 收稿日期:2019-08-20 修回日期:2019-09-26 接受日期:2019-09-26 发布日期:2019-10-14 出版日期:2020-02-10
  • 通讯作者: 戚永军
  • 作者简介:顾军华(1966—),男,天津人,教授,博士,CCF会员,主要研究方向:数据挖掘、智能信息处理
    王锋(1995—),男,河北保定人,硕士研究生,主要研究方向:医学图像处理、图像检索
    孙哲然(1996—),女,河北张家口人,硕士研究生,主要研究方向:医学图像处理、图像分类
    田泽培(1994—),女,河北石家庄人,硕士研究生,主要研究方向:医学图像处理、目标检测
    张亚娟(1984—),女,河北廊坊人,硕士,主要研究方向:数据挖掘、图像处理。
  • 基金资助:
    国家自然科学基金资助项目(61702157);河北省自然科学基金重点项目(F2016202144)

Retrieval method of pulmonary nodule images based on multi-scale convolution feature fusion

Junhua GU1,2,3, Feng WANG2,3, Yongjun QI1,4(), Zheran SUN2,3, Zepei TIAN2,3, Yajuan ZHANG2,3   

  1. 1.State Key Laboratory of Reliability and Intelligence of Electrical Equipment (Hebei University of Technology),Tianjin 300401,China
    2.Hebei Province Key Laboratory of Big Data Calculation (Hebei University of Technology),Tianjin 300401,China
    3.School of Artificial Intelligence,Hebei University of Technology,Tianjin 300401,China
    4.Information Technology Center,North China Institute of Aerospace Engineering,Langfang Hebei 065000,China
  • Received:2019-08-20 Revised:2019-09-26 Accepted:2019-09-26 Online:2019-10-14 Published:2020-02-10
  • Contact: Yongjun QI
  • About author:GU Junhua, born in 1966, Ph. D., professor. His research interests include data mining, intelligent information processing.
    WANG Feng, born in 1995, M. S. candidate. His research interests include medical image processing, image retrieval.
    SUN Zheran, born in 1996, M. S. candidate. Her research interests include medical image processing, image classification.
    TIAN Zepei, born in 1994, M. S. candidate. Her research interests include medical image processing, object detection.
    ZHANG Yajuan, born in 1984, M. S. Her research interests include data mining, image processing.
  • Supported by:
    the National Natural Science Foundation of China(61702157);the Key Program of Natural Science Foundation of Hebei Province(F2016202144)

摘要:

为了解决肺结节图像检索中特征提取难度大、检索精度低下的问题,提出了一种深度网络模型——LMSCRnet用于提取图像特征。首先采用多种不同尺寸滤波器卷积的特征融合方法以解决肺结节大小不一引起的局部特征难以获取的问题,然后引入SE-ResNeXt块来得到更高级的语义特征同时减少网络退化,最后得到肺结节图像的高级语义特征表示。为满足现实中大数据量检索任务的需求,将距离计算及排序过程部署到Spark分布式平台上。实验结果表明,基于LMSCRnet的特征提取方法能够更好地提取图像高级语义信息,在肺结节预处理数据集LIDC上能够达到84.48%的准确率,检索精度高于其他检索方法,而且使用Spark分布式平台完成相似度匹配及排序过程使得检索方法能够满足大数据量检索任务需求。

关键词: 肺结节图像, 图像检索, 特征融合, 并行优化, Spark, 深度学习

Abstract:

In order to solve the difficulty of feature extraction and low accuracy of retrieval in pulmonary nodule image retrieval, a deep network model named LMSCRnet was proposed to extract image features. Firstly, the feature fusion method of convolution of filters with different scales was adopted to solve the problem of difficulty in obtaining local features caused by different sizes of pulmonary nodules. Then, the SE-ReSNeXt block was introduced to obtain the semantic features with higher level and reduce network degradation. Finally, the high-level semantic feature representation of pulmonary nodule image was obtained. In order to meet the needs of massive data retrieval tasks in real life, the distance calculation and sorting process were deployed on the Spark distributed platform. The experimental results show that the feature extraction method based on LMSCRnet can better extract the image high-level semantic information, and can achieve 84.48% accuracy on the preprocessed dataset of lung nodules named LIDC, and has the retrieval precision higher than other retrieval methods. At the same time, using Spark distributed platform to complete similarity matching and sorting process enables the retrieval method to meet the requirements of massive data retrieval tasks.

Key words: pulmonary nodule image, image retrieval, feature fusion, parallel optimization, Spark, deep learning

中图分类号: