多媒体传感器网络中基于颜色空间的图像融合方案  

An Image Fusion Method Based on Color Space for Multimedia Sensor Networks

作　　者：王玉斐[1] 王汝传[1,2] 曾鸣[1] 黄海平[1] 孙力娟[1] 肖甫[1] 

机构地区：[1]南京邮电大学计算机学院,江苏南京210003 [2]南京大学计算机软件新技术国家重点实验室,江苏南京210093

出　　处：《电子学报》2009年第8期1659-1663,共5页Acta Electronica Sinica

基　　金：国家自然科学基金(No.60573141;No.60773041);国家863高技术研究发展计划(No.2006AA01Z201;No.2006AA01Z219;No.2006AA01Z439);江苏省高技术研究计划(No.BG2006001);2006江苏省软件专项;南京市高科技项目(2007软资106);江苏省博士后基金(No.0801019C)

摘　　要：基于彩色图像的YUV颜色空间和相邻节点间的视角相关性,将同一场景的监控任务分配给3个相关度较大的传感器节点,每个节点仅需处理亮度分量或色度分量.使用深度信息模型,以及基于自适应四叉树分割和分块空间变换的方法,对解码后的亮度和色度分量进行融合,实现监控场景的彩色图像重构.仿真实验结果表明,该方法有效可行,在视频传感器节点存储量、传输量和场景监控质量之间能取得良好的折中.The monitor task of the same scene was assigned to three highly-correlated video sensor nodes based on YUV color space and the view correlation of adjacent nodes. Each sensor node only needed to take charge of compressing and transmitting luminance or chrominance part. With the proposed depth information model, luminance and chrominance parts were fused using fusion method based on adaptive quadtree partitioning and space transform. And then the color image of the scene was reconstructed. The experimental and simulation results show that our method is effective and feasible. A well tradeoff is achieved between store,transmission cost and scene monitoring quality.

关 键 词：多媒体传感器网络 颜色空间 图像融合 四叉树分割 

分 类 号：TP37[自动化与计算机技术—计算机系统结构]