Edge Detectors Can Make Deep Convolutional Neural Networks More Robust

arXiv:2402.16479v1 Announce Type: new
Abstract: 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, …

abstract applications arxiv automation autonomous autonomous driving convolutional neural networks cs.cv driving edge examples human industry networks neural networks objects robust robustness safety safety-critical significance small type vulnerable