Deep convolutional neural networks (DCNN for short) are vulnerable to
examples with small perturbations. Improving DCNN's robustness is of great
significance to the safety-critical applications, such as autonomous driving
and industry automation. Inspired by the principal way that human eyes
recognize objects, i.e., largely relying on the shape features, this paper
first employs the edge detectors as layer kernels and designs a binary edge
feature branch (BEFB for short) to learn the binary edge features, which can be
easily integrated into any popular backbone. The four edge detectors can learn
the horizontal, vertical, positive diagonal, and negative diagonal edge
features, respectively, and the branch is stacked by multiple Sobel layers
(using edge detectors as kernels) and one threshold layer. The binary edge
features learned by the branch, concatenated with the texture features learned
by the backbone, are fed into the fully connected layers for classification. We
integrate the proposed branch into VGG16 and ResNet34, respectively, and
conduct experiments on multiple datasets. Experimental results demonstrate the
BEFB is lightweight and has no side effects on training. And the accuracy of
the BEFB integrated models is better than the original ones on all datasets
when facing FGSM, PGD, and C\&W attacks. Besides, BEFB integrated models
equipped with the robustness enhancing techniques can achieve better
classification accuracy compared to the original models. The work in this paper
for the first time shows it is feasible to enhance the robustness of DCNNs
through combining both shape-like features and texture features.

通过将边缘检测器作为层核心并设计一个二进制边缘特征分支（BEFB）来学习二进制边缘特征，结合纹理特征并将其与主干网络结合，以增强深度卷积神经网络（DCNN）的鲁棒性。实验证明，采用 BEFB 的集成模型在面对多种攻击时具有更高的分类准确性，并且具备了轻量化和不对训练造成负面影响的特点。该研究首次表明通过结合形状类特征和纹理特征可以增强 DCNN 的鲁棒性。

边缘检测器能够使深度卷积神经网络更加稳健

Edge Detectors Can Make Deep Convolutional Neural Networks More Robust

This paper presents an evaluation of edge detectors using receiver operating
characteristic (ROC) analysis on the BIPED dataset. Our study examines the
benefits and drawbacks of applying this technique in Matlab. We observed that
while ROC analysis is suitable for certain edge filters, but for filters such
as Laplacian, Laplacian of Gaussian, and Canny, it presents challenges when
accurately measuring their performance using ROC metrics. To address this
issue, we introduce customization techniques to enhance the performance of
these filters, enabling more accurate evaluation. Through our customization
efforts, we achieved improved results, ultimately facilitating a comprehensive
assessment of the edge detectors.

该论文使用 BIPED 数据集，使用接收器操作特征（ROC）分析方法来评估边缘检测器，探讨了在 Matlab 中应用该技术的优缺点。作者通过定制技术提高了 Laplacian，Laplacian of Gaussian 和 Canny 等过滤器的性能，从而更准确地评估边缘检测器的综合表现。