当前位置: X-MOL 学术arXiv.cs.CV › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Dense Steerable Filter CNNs for Exploiting Rotational Symmetry in Histology Images
arXiv - CS - Computer Vision and Pattern Recognition Pub Date : 2020-04-06 , DOI: arxiv-2004.03037
Simon Graham, David Epstein and Nasir Rajpoot

Histology images are inherently symmetric under rotation, where each orientation is equally as likely to appear. However, this rotational symmetry is not widely utilised as prior knowledge in modern Convolutional Neural Networks (CNNs), resulting in data hungry models that learn independent features at each orientation. Allowing CNNs to be rotation-equivariant removes the necessity to learn this set of transformations from the data and instead frees up model capacity, allowing more discriminative features to be learned. This reduction in the number of required parameters also reduces the risk of overfitting. In this paper, we propose Dense Steerable Filter CNNs (DSF-CNNs) that use group convolutions with multiple rotated copies of each filter in a densely connected framework. Each filter is defined as a linear combination of steerable basis filters, enabling exact rotation and decreasing the number of trainable parameters compared to standard filters. We also provide the first in-depth comparison of different rotation-equivariant CNNs for histology image analysis and demonstrate the advantage of encoding rotational symmetry into modern architectures. We show that DSF-CNNs achieve state-of-the-art performance, with significantly fewer parameters, when applied to three different tasks in the area of computational pathology: breast tumour classification, colon gland segmentation and multi-tissue nuclear segmentation.

中文翻译:

用于利用组织学图像中的旋转对称性的密集可控滤波器 CNN

组织学图像在旋转下本质上是对称的,其中每个方向出现的可能性相同。然而,这种旋转对称性并未被广泛用作现代卷积神经网络 (CNN) 中的先验知识,导致数据饥渴模型在每个方向学习独立特征。允许 CNN 旋转等变消除了从数据中学习这组变换的必要性,而是释放了模型容量,允许学习更多的判别特征。所需参数数量的减少也降低了过度拟合的风险。在本文中,我们提出了密集可转向滤波器 CNN(DSF-CNN),它在密集连接的框架中使用组卷积和每个滤波器的多个旋转副本。每个过滤器被定义为可操纵基础过滤器的线性组合,与标准过滤器相比,可以实现精确旋转并减少可训练参数的数量。我们还首次对用于组织学图像分析的不同旋转等变 CNN 进行了深入比较,并展示了将旋转对称编码到现代架构中的优势。我们表明,当应用于计算病理学领域的三个不同任务时,DSF-CNN 实现了最先进的性能,参数显着减少:乳腺肿瘤分类、结肠腺分割和多组织核分割。我们还首次对用于组织学图像分析的不同旋转等变 CNN 进行了深入比较,并展示了将旋转对称编码到现代架构中的优势。我们表明,当应用于计算病理学领域的三个不同任务时,DSF-CNN 实现了最先进的性能,参数显着减少:乳腺肿瘤分类、结肠腺分割和多组织核分割。我们还首次对用于组织学图像分析的不同旋转等变 CNN 进行了深入比较,并展示了将旋转对称编码到现代架构中的优势。我们表明,当应用于计算病理学领域的三个不同任务时,DSF-CNN 实现了最先进的性能,参数显着减少:乳腺肿瘤分类、结肠腺分割和多组织核分割。
更新日期:2020-07-21
down
wechat
bug