现代电子技术2026，Vol.49Issue(2)：44-48,5.DOI:10.16652/j.issn.1004-373x.2026.02.007
基于权重共享的轻量化注意力卷积编码器
Lightweight attention convolutional encoder based on weight sharing
摘要
Abstract
Due to the complexity and diversity of the environment,there is a large amount of interference information in the images obtained by vehicle mounted sensors,which poses significant difficulties for the recognition of traffic signs.A lightweight attention convolutional encoder based on weight sharing pre-training is proposed.Important features of the encoded features are focused by combining attention mechanisms with asymmetric convolution structures.The parallel structure in contrastive learning is used to pre-training the encoder,so as to improve its sample learning ability in complex environments.The output feature encoding is predicted to complete the recognition of traffic signs.The experimental results show that the accuracy rates of the proposed algorithm under blurry and occluded conditions can reach 94.65%and 91.23%respectively,which is practical.关键词
通道注意力机制/卷积编码器/非对称卷积结构/权重共享/对比学习/图像识别Key words
channel attention mechanism/convolutional encoder/asymmetric convolution structure/weight sharing/contrastive learning/image recognition分类
信息技术与安全科学引用本文复制引用
肖芷翊,奚峥皓..基于权重共享的轻量化注意力卷积编码器[J].现代电子技术,2026,49(2):44-48,5.基金项目
国家自然科学基金项目(12104289) （12104289）
