BriefGPT.xyz
Apr, 2025
利用生物启发滤波器增强卷积神经网络对遮挡的鲁棒性
Enhancing CNNs robustness to occlusions with bioinspired filters for border completion
HTML
PDF
Catarina P. Coutinho, Aneeqa Merhab, Janko Petkovic, Ferdinando Zanchetta, Rita Fioresi
TL;DR
本研究针对卷积神经网络在处理遮挡图像时的鲁棒性不足问题，提出了一种基于视觉皮层边界完成机制的定制滤波器的方法。研究发现，经过修改的LeNet 5在处理被遮挡的MNIST图像时，准确性有显著提升，显示出这种新方法的有效性。
Abstract
We exploit the mathematical modeling of the visual cortex mechanism for
Border Completion
to define custom filters for CNNs. We see a consistent improvement in performance, particularly in accuracy, when our modified
Le
→