注意力感知的多尺度语义视杯盘分割  

Multi-scale semantic OD/OC segmentation method based on attention perception

作　　者：燕杨[1] 曹娅迪 黄文博[1] YANG Yan;CAO Yadi;HUANG Wenbo(School of Computer Science and Technology,Changchun Normal University,Changchun 130032,China)

机构地区：[1]长春师范大学计算机科学与技术学院,吉林长春130032

出　　处：《光学精密工程》2023年第21期3203-3211,共9页Optics and Precision Engineering

基　　金：吉林省科技厅自然科学基金(联合基金项目)(No.YDZJ202101ZYTS147)。

摘　　要：为了解决编码器-解码器网络结构在目标提取中抑制无关语义、跨越语义鸿沟等问题,以获取更高精度,采用UNet作为提取特征的主干网络;为了减轻浅层特征与深层特征语义的差异,设计一种融合注意力感知的多尺度语义池化模块(Channel-Spatial-Pyramid,CSP),替代早期层中的跳跃链接。CSP模块从空间与通道两个层面强调更有意义的语义信息,通过4个不同池化核的并行分支提取不同尺度特征,聚合所有分支结果与后期层特征拼接。实验结果表明,CSP-Net在彩色眼底图像视盘分割中的Dice指数可达99.6%,视杯分割结果的Dice指数为92.1%,相比现有算法均有提高。所提出的CSP-Net对于眼底图像中的微小目标提取的有效性及抗干扰性较高,可为青光眼筛查与诊断临床提借鉴。To suppress irrelevant semantics and cross semantic gaps in object extraction using an encoderdecoder network structure,thereby achieving higher accuracy.U-Net is used as the backbone network for feature extraction.To reduce semantic differences between shallow and deep features,a multi-scale se⁃mantic pooling module(CSP,Channel-Spatial-Pyramid)integrates attention perception and replaces skip links in early layers.The CSP module emphasizes more meaningful semantic information from two levels corresponding to space and channel,extracts features at different scales through parallel branches of four different pooling cores,and aggregates all branch results to splice with the features of later layers.The ex⁃perimental results show that the Dice index of CSP-Net in color fundus image disc segmentation reaches 99.6%,whereas that of cup segmentation reaches 92.1%.Both results represent improvements over ex⁃isting algorithms.CSP-Net exhibits a high effectiveness and anti-interference ability for extracting small targets in fundus images,making it appropriate for clinical reference in glaucoma screening and diagnosis.

关 键 词：多尺度语义 注意力感知 目标提取 U-Net 

分 类 号：TP394.1[自动化与计算机技术—计算机应用技术] TH691.9[自动化与计算机技术—计算机科学与技术]