无线电工程2025，Vol.55Issue(6)：1161-1176,16.DOI:10.3969/j.issn.1003-3106.2025.06.004
基于双树复小波变换和U-Net的视网膜血管分割
Retinal Blood Vessel Segmentation Based on Dual-Tree Complex Wavelet Transform and U-Net Network
摘要
Abstract
Biomedical image segmentation has become one of the key tasks in medical diagnosis.However,due to the complex morphology of tissues and organs and the diversity of their structures,the practical application of medical image segmentation techniques faces significant technical challenges.In traditional Convolutional Neural Network(CNN),the maximum pooling operation often results in irreversible loss of information.Although the incorporation of the wavelet transform mitigates this issue to some extent,the wavelet transform itself also has its own limitations.To solve this problem,a retinal blood vessel segmentation model named DTCWU-Net based on Dual-Tree Complex Wavelet Transform(DTCWT)and U-Net is proposed.The model replaces the traditional pooling layer with DTCWT and traditional upsampling layer with the Inverse DTCWT(IDTCWT).This significantly enhances the feature extraction ability,particularly in preserving image details.DTCWU-Net also introduces Low and High Feature Fusion Attention(LHFFA)and Multi-Scale Gate Attention(MSGA)modules to further improve the segmentation performance.The experimental results show that DTCWU-Net achieved an Accuracy(ACC)of 0.968 6 and an Area Under the ROC(Receiver Operating Characteristic)Curve(AUC)of 0.986 7 on the DRIVE dataset,an ACC of 0.975 0 and an AUC of 0.990 3 on the CHASE_DB1 dataset,an ACC of 0.975 7 and an AUC of 0.990 1 on the STARE dataset,The proposed model demonstrates superior performance in key metrics when compared to other mainstream methods,including F1,Sensitivity(SE),ACC and AUC.Through the collaborative optimization of multiple modules,DTCWU-Net significantly improves and demonstrates the accuracy of retinal vessel segmentation and the ability to recover details.关键词
双树复小波变换/小波变换/U-Net/医学图像分割/注意力机制/深度学习Key words
DTCWT/wavelet transform/U-Net/medical image segmentation/attention mechanism/deep learning分类
计算机与自动化引用本文复制引用
陶寅涵,朱家明,吴军..基于双树复小波变换和U-Net的视网膜血管分割[J].无线电工程,2025,55(6):1161-1176,16.基金项目
国家自然科学基金(62473362) National Natural Science Foundation of China(62473362) （62473362）
