福建电脑2025，Vol.41Issue(7)：42-46,5.DOI:10.16707/j.cnki.fjpc.2025.07.009
用于RGB-T显著性目标检测的特征过滤网络
Feature Filtering Network for RGB-T Salient Object Detection
何海洋 1陈炳锡1
作者信息
- 1. 信阳艺术职业学院信息工程学院 河南 信阳 464000
- 折叠
摘要
Abstract
To efficiently retrieve salient information from features,this paper proposes a feature filtering network for RGB-T salient object detection.By using group attention mechanism,the salient features of multimodal fusion are filtered,gradually guiding the network model to obtain more accurate salient targets.The experimental results show that compared with 8 algorithms,the algorithm proposed in this paper performs well in 5 indicators,verifying the effectiveness of the method.关键词
显著性目标检测/特征净化/多模态融合/注意力机制Key words
Salient Object Detection/Feature Purification/Multimodality Fusion/Attention Mechanism分类
信息技术与安全科学引用本文复制引用
何海洋,陈炳锡..用于RGB-T显著性目标检测的特征过滤网络[J].福建电脑,2025,41(7):42-46,5.
