基于蚁群优化云计算的小区车辆监控系统研究  

Research of Neighborhood Vehicle monitoring System Based on Ant Colony Optimization Cloud Computing

作　　者：刘丹[1] 程跃华[2] 马世霞[1] 

机构地区：[1]河南机电高等专科学校计算机科学与技术系,河南新乡453000 [2]焦作大学信息工程学院,河南焦作454003

出　　处：《计算机测量与控制》2013年第5期1240-1242,共3页Computer Measurement &Control

基　　金：河南省教育厅自然科学研究计划项目(2011B520010)

摘　　要：提出一种基于蚁群算法和云计算的小区车辆监控系统设计方法。系统采用模块化的设计方案,在车辆图像检测模块中,运用了蚁群优化算法对车辆特征进行分割,能够有效地排除多车重叠状态下的特征干扰问题。在数据调度模块设计中,采用云调度平台,对海量特征图像实现有序调度,消除了系统中数据冲突的现象。给出了系统详细的硬件和软件模块设计方案。测试表明,系统能够完成监控图像内的有效定位,系统反应时间从3秒提高到不超过0.9秒,延迟符合应用要求,定位的准确性从原来的3.7米缩短到0.5米以内,大幅提高了系统的定位精度。Proposed based on ant colony algorithm and cloud computing neighborhood vehicle monitoring system design method. System adopts the modular design, in the vehicle image detection module, the use of the ant colony optimization algorithm for vehicle feature segmentation, can effectively eliminate many cars overlap state characteristics of interference problem. In data scheduling module design, the cloud dispatching platform, the mass characteristic image to realize order scheduling, eliminate the system data conflict phenomenon. System were given detailed hardware and software module design scheme. System test shows that the system can complete monitoring image within the effective positioning, the system response time from 3 seconds to improve to not more than 0. 9 seconds, delayed coincidence application requirements, the accuracy of positioning from the original 3. 7 meters shorten to 0.5 meters of less than, greatly improve the positioning accuracy of system.

关 键 词：蚁群算法 云计算 车辆监控 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]