海岸海岛重要基地图像型火灾探测系统研究  

Research of Fire Image Detection System based on the Important bases in Coast and Island

作　　者：毕振波[1] 杨花[2] 高禹[1] 朱顺乐[1] 

机构地区：[1]浙江海洋学院数理与信息学院,浙江舟山316004 [2]浙江海洋学院船舶与海洋工程学院,浙江舟山316004

出　　处：《浙江海洋学院学报（自然科学版）》2013年第2期123-128,134,共7页Journal of Zhejiang Ocean University(Natural Science Edition)

基　　金：浙江省自然科学基金项目(LY12F02021)

摘　　要：针对当前海岸海岛重要基地防火方法不能适应实际需求,提出了基于云计算的图像型火灾探测系统。作为一种新的服务计算模式,其具有弹性计算、简化应用和降低成本的特点,然后对云模式图像型的火灾探测系统行了深入的研究,提出了系统架构,探讨了火灾探测的功能服务设计,提出了云模式火灾探测中关键阶段融合多种数据处理技术提高探测准确率的方法。最后借助开源Hadoop平台,对系统方案进行了有效验证。As the current fire prevention methods for important base in coast and island can＇t meet the actual demand, Fire Image Detection System based on cloud computing is proposed. As a new service model, the system has elastic computing, simplified applications and cost reduction characteristics. Therefore, the indepth study of fire image detection system based on cloud computing is done, the system architecture is proposed, functional services design of the system is discussed, the key technologies of integrating a variety of data processing methods in fire image detection based on cloud computing are also put forward, including pretreatment, image segmentation and fire image feature extraction and identification. Finally, with the help of open source Hadoop platform, an effective verification of system solutions was done.

关 键 词：海岸海岛 云计算 图像处理 图像型火灾探测 

分 类 号：X932[环境科学与工程—安全科学] TP311.5[自动化与计算机技术—计算机软件与理论]