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 ›› 2009, Vol. 35 ›› Issue (11): 178-180. doi: 10.3969/j.issn.1000-3428.2009.11.061

• 人工智能及识别技术 • 上一篇    下一篇

基于多传感器/目标识别的电子宠物视觉系统

王 慧1,介龙梅2   

  1. (1. 北京科技大学信息工程学院,北京 100083;2. 大庆师范学院计算机科学与信息系,大庆 163712)
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2009-06-05 发布日期:2009-06-05

Electronic Pet Vision System Based on Multi-sensor and Multi-target Recognition

WANG Hui1, JIE Long-mei2   

  1. (1. School of Information Engineering, Beijing University of Science & Technology, Beijing 100083; 2. Dept. of Computer Science & Information, Daqing Normal University, Daqing 163712)
  • Received:1900-01-01 Revised:1900-01-01 Online:2009-06-05 Published:2009-06-05

摘要: 将多传感器技术用于电子宠物“智能狗”的视觉系统,提出一种集娱乐与安全功能于一体的架构体系,阐述该系统的工作流程,介绍基于一阶差分链码图像的表示方法和句法模式识别理论,给出一种新的头部形状检测方法,仿真实验结果表明,该系统能够获得较好的性能,且具有一定市场应用价值。

关键词: 模式识别, 链码, 多传感器, 多目标识别

Abstract: The multi-sensor technique is applied into the vision system of electronic pet called electronic dog, and a system architecture is proposed, which integrates the recreation and the security capabilities. The flowchart of the system is explained. The method based on first-order difference chain code image and structural pattern recognition theory are introduced. A novel head shape detection method is given. Simulation experimental results show this system gets better performance and has the value of market application.

Key words: pattern recognition, chain code, multi-sensor, multi-target recognition

中图分类号: