作者投稿和查稿 主编审稿 专家审稿 编委审稿 远程编辑

计算机工程 ›› 2020, Vol. 46 ›› Issue (8): 264-270. doi: 10.19678/j.issn.1000-3428.0054946

• 图形图像处理 • 上一篇    下一篇

引入通道注意力机制的SSD目标检测算法

张海涛, 张梦   

  1. 辽宁工程技术大学 软件学院, 辽宁 葫芦岛 125105
  • 收稿日期:2019-05-17 修回日期:2019-08-11 发布日期:2019-08-21
  • 作者简介:张海涛(1974-),男,教授、博士,主研方向为图像处理、计算机视觉;张梦,硕士研究生。
  • 基金资助:
    辽宁省自然科学基金(20170540426)。

SSD Target Detection Algorithm with Channel Attention Mechanism

ZHANG Haitao, ZHANG Meng   

  1. College of Software, Liaoning Technical University, Huludao, Liaoning 125105, China
  • Received:2019-05-17 Revised:2019-08-11 Published:2019-08-21

摘要: 为提升原始SSD算法的小目标检测精度及鲁棒性,提出一种基于通道注意力机制的SSD目标检测算法。在原始SSD算法的基础上对高层特征图进行全局池化操作,结合通道注意力机制增强高层特征图的语义信息,并利用膨胀卷积结构对低层特征图进行下采样扩大其感受野以增加细节与位置信息,再通过级联的方式将低层特征图与高层特征图相融合,从而实现小目标及遮挡目标的有效识别。实验结果表明,与原始SSD算法相比,该算法在PASCAL VOC数据集上的平均精度均值提升了2.2%,具有更高的小目标检测精度和更好的鲁棒性。

关键词: SSD算法, 全局池化, 通道注意力机制, 膨胀卷积, PASCAL VOC数据集

Abstract: In order to improve the accuracy and robustness of the original Single Shot Multibox Detector(SSD) algorithm for small target detection,this paper proposes a SSD target detection algorithm based on channel attention mechanism.The algorithm implements global pooling on high-level feature maps on the basis of the original SSD algorithm.Then the semantic information of high-level feature maps is enhanced by using the channel attention mechanism,and the dilated convolution structure is introduced for subsampling of low-level feature maps to enlarge their receptive field,so as to improve details and location information.Finally,different levels of feature maps are fused by cascading to implement effective recognition of small objects and occluded objects.Experimental results on the PASCAL VOC dataset show that compared with the original SSD algorithm,the proposed algorithm improves the mean Average Precision(mAP) by 2.2% with a higher accuracy and robustness for small object detection.

Key words: SSD algorithm, global pooling, channel attention mechanism, dilated convolution, PASCAL VOC dataset

中图分类号: