计算机与现代化 ›› 2025, Vol. 0 ›› Issue (10): 14-19.doi: 10.3969/j.issn.1006-2475.2025.10.003

• 图像处理 • 上一篇    下一篇

基于变核卷积的HRNetV2模型对舌下络脉图像分割算法

  


  1. (1.安徽水利水电职业技术学院电子信息工程学院,安徽 合肥 231603; 2.合肥云诊信息科技有限公司人工智能实验室,安徽 合肥 230088; 3.安徽中医药大学第二附属医院,安徽 合肥 230012; 4.合肥工业大学计算机与信息学院,安徽 合肥 230009) 
  • 出版日期:2025-10-27 发布日期:2025-10-27
  • 作者简介:作者简介:蒋冬梅(1983—),女,安徽六安人,硕士研究生,研究方向:计算机应用,E-mail: 99744663@qq.com; 通信作者:彭成东(1984—),男,安徽六安人,工程师,博士,研究方向:图像处理,计算机视觉的理论与应用研究,E-mail: 2020020018@mail.hfut.edu.cn。
  • 基金资助:
    安徽省重点研究与开发计划项目(2022h11020018); 安徽省科技重大专项计划(202303a07020008); 新安医学与中医药现代化研究所揭榜挂帅项目(2024CXMMTCM002); 安徽省高校自然科学研究项目(2022AH052293)
        

Sublingual Vein Image Segmentation Based on HRNetV2 Model with Variable Kernel Convolution

  1. (1. Electronic Information Engineering College, Anhui Technical College of Water Resources and Hydroelectric Power, Hefei  231603, China; 2. Artificial Intelligence Laboratory, Hefei Yunzhen Information Technology Co., Ltd., Hefei 230088, China; 3. Second Affiliated Hospital, Anhui University of Chinese Medicine, Hefei 230012, China; 4. School of Computer Science and Information Engineering, Hefei University of Technology, Hefei 230009, China)
  • Online:2025-10-27 Published:2025-10-27

摘要: 摘要:针对现有舌下络脉分析采用卷积神经网络分类或分割的方法提取络脉图像,但存在对络脉细节提取精度不高的问题,本文提出一种改进的HRNetV2高分辨率语义分割网络算法对舌下络脉进行提取。该方法采用高分辨率保持的HRNetV2网络结构,将高分辨率到低分辨率子网络结构的输出并行连接成多尺度融合具有更高的空间精确度的特征图,改善舌下络脉细节信息丢失的问题。同时使用具有任意采样形状和任意数目参数的变核卷积AKConv代替普通卷积,提升卷积对脉形结构变化的适应能力,减少欠分割问题。在安徽中医药大学云诊科技舌象开放平台上抽取数据验证,该算法的像素精度PA、均像素精度mPA、均交并比mIoU分别为95.28%、92.33%和93.42%,优于Mask-RCNN模型、U-Net系列模型、HRNetV2模型。改进的HRNetV2方法对于舌下络脉图像分割精确度高,为进一步应用于脉色和脉形特征定量研究提供新方法。


关键词: 关键词:舌下络脉分割, 高分辨率分割, 变核卷积, 不规则卷积, 多尺度特征融合

Abstract:
Abstract: The existing analysis of sublingual vein often uses convolutional neural network (CNN) image classification methods or image segmentation methods to extract. But there is a problem of low accuracy in extracting details of meridians. Therefore, an improved HRNetV2 high-resolution semantic segmentation network algorithm is proposed to extract sublingual meridians. Adopting a high-resolution HRNetV2 network structure, the outputs of sub network structures from high to low resolution are connected in parallel to form multi-scale fused feature maps with higher spatial accuracy, improving the problem of loss of detailed information in sublingual veins. In addition, AKConv, convolutional kernel with arbitrary sampled shapes and arbitrary number of parameters instead of ordinary convolution can improve the convolution’s adaptability to change pulse structure and reduce the problem of under segmentation. The algorithm is validated through data extraction on the tongue image open platform of Anhui university of Traditional Chinese Medicine (TCM) cloud diagnosis technology, with pixel accuracy(PA), mean pixel accuracy(mPA), and mean intersection over union(mIoU) of 95.28%, 92.33%, and 93.42%, respectively, which is superior to the Mask-RCNN model, U-Net models, and HRNetV2 model. The improved HRNetV2 method has high accuracy in segmenting sublingual vein images, providing a new method for further quantitative research on pulse color and shape features.

Key words: Key words:sublingual vein segmentation, high-resolution segmentation, variable kernel convolution, irregular convolution, multi-scale feature fusion

中图分类号: