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 (10): 248-252. doi: 10.19678/j.issn.1000-3428.0056685

• 图形图像处理 • 上一篇    下一篇

基于高分辨率特征的舌象分割算法研究

马龙祥1,2, 杨浩1,2, 宋婷婷1, 翟鹏博1,2, 余亢1   

  1. 1. 中国科学院微电子研究所, 北京 100029;
    2. 中国科学院大学, 北京 100049
  • 收稿日期:2019-11-12 修回日期:2019-12-24 发布日期:2020-01-14
  • 作者简介:马龙祥(1995-),男,硕士研究生,主研方向为生物医学图像处理;杨浩,副教授、博士;宋婷婷,助理研究员、硕士;翟鹏博,硕士研究生;余亢,助理研究员、硕士。
  • 基金资助:
    国家科技重大专项(2018ZX01031201)。

Research on Tongue Image Segmentation Algorithm Based on High Resolution Feature

MA Longxiang1,2, YANG Hao1,2, SONG Tingting1, ZHAI Pengbo1,2, YU Kang1   

  1. 1. Institute of Microelectronics of Chinese Academy of Sciences, Beijing 100029, China;
    2. University of Chinese Academy of Sciences, Beijing 100049, China
  • Received:2019-11-12 Revised:2019-12-24 Published:2020-01-14

摘要: 舌象的精准分割对舌诊中舌体识别与分类具有重要意义,采用传统图像处理方法和深度学习方法分割舌象会丢失部分舌象边缘信息,从而降低舌体识别精确度。针对该问题,提出一种利用高分辨率网络的舌象分割算法。使用区域定位网络识别舌体并提取舌象原图特征生成建议框,对其进行分类和回归处理以定位舌象所在区域,同时构建高分辨率网络提取该区域高分辨率特征,最终完成舌象分割。实验结果表明,该算法可有效保留舌象边缘信息,其分割结果平均交并比达到98.2%,较SegNet、Mask-RCNN算法分割舌象更精准。

关键词: 舌诊, 舌象, 深度学习, 高分辨率特征, 实例分割

Abstract: The accurate segmentation of tongue images is of great significance for tongue recognition and classification in tongue diagnosis.The traditional image processing and deep learning methods will lose part of the edge information of tongue images,thus reducing the accuracy of tongue recognition.To solve the problem,this paper proposes a tongue image segmentation algorithm based on high-resolution network.The region location network is used to identify the tongue and extract the features of the original image of the tongue to generate suggestion boxes,which are classified and processed with regression to locate the tongue area.At the same time,a high-resolution network is constructed to extract the high-resolution features of the region,and finally complete the tongue image segmentation.Experimental results show that the proposed algorithm can effectively preserve the edge information of tongue images,and the mean Intersection over Union(mIoU) of segmentation results is 98.2%,which is more accurate than that of SegNet and Mask-RCNN algorithms.

Key words: tongue diagnosis, tongue image, deep learning, high resolution feature, instance segmentation

中图分类号: