计算机工程 ›› 2010, Vol. 36 ›› Issue (2): 214-216.doi: 10.3969/j.issn.1000-3428.2010.02.076

• 多媒体技术及应用 • 上一篇    下一篇

视频传感器网络中基于动态注意力的图像融合

谭 励1,2,杨明华3,4,曹元大1,成保栋1   

  1. (1. 北京理工大学计算机学院智能信息技术北京市重点实验室,北京 100081;2.北京工商大学艺术与传媒学院,北京 100048;3. 中国人民解放军96627部队,北京 100085;4. 第二炮兵装备研究院博士后工作站,北京 100085)
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2010-01-20 发布日期:2010-01-20

Image Fusion Based on Dynamic Attention in Video Sensor Network

TAN Li1,2, YANG Ming-hua3,4,CAO Yuan-da1, CHENG Bao-dong1   

  1. (1. Beijing Key Laboratory of Intelligent Information Technology, School of Computer, Beijing Institute of Technology, Beijing 100081; 2. School of Art and Communication, Beijing Technology and Business University, Beijing 100048; 3. Unit 96627 of Peoples Liberation Army, Beijing 100085; 4. The Post-doctoral Workstations of Second Artillery Equipment Academy, Beijing 100085)
  • Received:1900-01-01 Revised:1900-01-01 Online:2010-01-20 Published:2010-01-20

摘要: 为了降低视频传感器网络中的网络负载,减少能量消耗并降低时延,提出一种基于动态注意力的图像分层融合方法。通过对网络的结构化部署和节点间的区域映射,对视频监测区域进行逻辑划分。利用动态注意力模型对局部重点区域进行精度优化,实现多质量图像融合。实验结果证明了该方法的有效性。

关键词: 视频传感器网络, 图像融合, 动态注意力, 区域分割, 区域映射

Abstract: In order to reduce the network load, energy consumption and delay in video sensor network, this paper proposes a stratification image fusion method based on dynamic attention. The region of video monitoring is divided logically according to the structural deployment of network and inter-node regional mapping. Refined optimization of local-important-region is achieved by using dynamic attention module and multi-quality image fusion is implemented. Experimental results show that the method is effective.

Key words: video sensor network, image fusion, dynamic attention, region segmentation, region mapping

中图分类号: