摘要
利用遗传算法的最佳熵阈值法进行图像分割 ,缩短了寻找阈值的时间 .分割后提取可疑区域 ,再运用人工智能中的基于缺省规则推理对可疑区域进行分别处理 .本系统运用于非法入侵对象识别和火灾识别 ,经实验验证实用可靠 ,满足实时性要求 .对采集的彩色图像具有与灰度图像同样良好的效果 ,能将报警信号和可疑区域图像通过网络传送到报警中心 .
General image segmentation techniques and rule-based inference were used to recognize fire and intruder. After segmentation and getting the suspected light-area in the image, this paper used rule-based inference for intelligent recognition of the are. The results show that the system is practical, reliable and in real-time. The system has a good result on processing color images as well as gray images. It also can send the alarm signal and the suspected light-area image to the alarm center through networks.
出处
《上海交通大学学报》
EI
CAS
CSCD
北大核心
2002年第4期498-498,共1页
Journal of Shanghai Jiaotong University
基金
上海市教育基金会"曙光计划"资助项目
