作者投稿和查稿 主编审稿 专家审稿 编委审稿 远程编辑

计算机工程 ›› 2020, Vol. 46 ›› Issue (1): 255-261,270. doi: 10.19678/j.issn.1000-3428.0055495

• 图形图像处理 • 上一篇    下一篇

基于分水岭修正与U-Net的肝脏图像分割算法

亢洁1a, 丁菊敏1a, 万永2, 雷涛1b   

  1. 1. 陕西科技大学 a. 电气与控制工程学院;b. 电子信息与人工智能学院, 西安 710021;
    2. 西安交通大学第一附属医院 老年外科, 西安 710061
  • 收稿日期:2019-07-16 修回日期:2019-08-29 出版日期:2020-01-15 发布日期:2019-09-02
  • 作者简介:亢洁(1973-),女,副教授、博士,主研方向为数字图像处理、模式识别;丁菊敏,硕士研究生;万永,主治医师、博士研究生;雷涛,教授、博士。
  • 基金资助:
    国家自然科学基金(61871259,61811530325)。

Liver Image Segmentation Algorithm Based on Watershed Correction and U-Net

KANG Jie1a, DING Jumin1a, WAN Yong2, LEI Tao1b   

  1. 1a. School of Electrical and Control Engineering;1b. School of Electronic Information and Artificial Intelligence, Shaanxi University of Science and Technology, Xi'an 710021, China;
    2. Department of Geratic Surgery, The First Affiliated Hospital of Xi'an Jiaotong University, Xi'an 710061, China
  • Received:2019-07-16 Revised:2019-08-29 Online:2020-01-15 Published:2019-09-02

摘要: 在利用卷积神经网络分割肝脏边界较模糊的影像数据时容易丢失位置信息,导致分割精度较低。针对该问题,提出一种基于分水岭修正与U-Net模型相结合的肝脏图像自动分割算法。利用U-Net分层学习图像特征的优势,将浅层特征与深层语义特征相融合,避免丢失目标位置等细节信息,得到肝脏初始分割结果。在此基础上,通过分水岭算法形成的区域块对肝脏初始分割结果的边界进行修正,以获得边界平滑精确的分割结果。实验结果表明,与传统的图割算法和全卷积神经网络算法相比,该算法能够实现更为精准的肝脏图像分割。

关键词: 肝脏图像分割, 卷积神经网络, U-Net模型, 分水岭算法, 边界修正

Abstract: When Convolutional Neural Network(CNN) is used for liver image segmentation with blurred boundaries,the segmentation precision is reduced due to the frequent loss of location information.To address the problem,this paper proposes an automated liver image segmentation algorithm that combines the watershed correction and the U-Net model.The algorithm takes advantages of U-Net in layered learning of image features,so as to achieve fusion of shallow features and deep features without loss of detailed information,such as the location of the target.After the initial result of liver image segmentation is obtained,the boundaries of the initial result is corrected by using blocks formed by the watershed algorithm,so as to obtain a segmentation result with smooth and precise boundaries.Experimental results show that the proposed algorithm can implement more precise liver image segmentation compared with the existing graph-cut algorithm and the Fully Convolutional Network(FCN) algorithm.

Key words: liver image segmentation, Convolutional Neural Network(CNN), U-Net model, watershed algorithm, boundary correction

中图分类号: