基于注意力机制的多尺度残差U-Net眼底血管分割

Multi-Scale Residual U-Net Fundus Blood Vessel Segmentation Based on Attention Mechanism

针对现有眼底血管分割方法难以辨别细小血管及交叉处血管分割断裂的问题,提出了一种基于注意力机制的多尺度U型网络。在编码阶段使用改进的残差块结构提取血管深度特征的同时有效解决过拟合问题,接着依次采用多尺度卷积模块和多尺度注意力模块进一步获取深度特征的多尺度特征信息。然后,使用MaxBlurPool进行池化,对数据进行降维并保证平移不变性。此外,在最后一个编码层引入混合注意力机制和并行空洞卷积,前者从通道和空间维度强调需要重点关注的信息,抑制背景区域的干扰;后者用来获取不同大小感受野的特征信息,且不会引入多余参数而导致计算负担。在解码部分,改进跳跃连接方式以抑制噪声的干扰并获得更加丰富的上下文信息。所提算法在公开的眼底数据集上取得了优于其他算法的分割效果。

Some existing retinal vessel segmentation methods have been unsuccessful in distinguishing weak blood vessels and have suffered from blood vessel segmentation disconnections at intersections.To solve this problem,a multiscale Ushaped network based on attention mechanism was proposed in this paper.In the encoding part,the proposed algorithm employed the improved residual block structure to extract the depth features of blood vessels while effectively solving the overfitting problem.In turn,the multiscale convolution module and multiscale attention module were used to obtain multiscale feature information of the depth features.Then,MaxBlurPool was used as the pooling method to reduce dimensions of data and ensure the translation invariance.In addition,hybrid attention module and parallel dilated convolution were presented in the last encoding layer,where the former emphasized the information that needs to be focused from the channel and space dimensions to suppress the interference of the background area and the latter was used to obtain the characteristic information of receptive fields with different sizes while not introducing redundant parameters to cause computational burden.In the decoding part,skip connection was improved to suppress noise and obtain more abundant context information.The proposed algorithm achieved better segmentation effect than other methods on public fundus datasets.