盲环境移动定位(Ⅰ):序列编码图形路标识别算法  

Mobile Positioning in Blind Environments(Ⅰ):Pattern Recognition Algorithms for Coded Sequence Pattern Signposts

作　　者：王植[1] 吴立新[1,2] 孔祥勤[1] 

机构地区：[1]东北大学资源与土木工程学院,辽宁沈阳110819 [2]北京师范大学减灾与应急管理研究院,北京100875

出　　处：《东北大学学报（自然科学版）》2011年第5期732-735,共4页Journal of Northeastern University(Natural Science)

基　　金：国家杰出青年科学基金资助项目(50525414);国家自然科学基金资助项目(40901220);霍英东青年教师基金资助项目(122025);中央高校基本科研业务费专项资金资助项目(N100401009);河南理工大学矿山空间信息技术国家测绘局重点实验室开放基金资助项目(KLM200913)

摘　　要：针对无卫星导航信号的盲环境中移动目标定位需要,提出一种适合于盲环境移动定位的序列编码图形路标技术方案,包含更多可用信息的序列编码图形路标方案.在测试环境中布设序列编码图形标志牌,建立了实现盲环境中移动目标快速定位定姿的路标模型实时采集包含编码图形路标的图像,利用SIFT(scale invariance feature transform)算法定位编码图形路标,使用Otsu算法进行图像分割和路标边界跟踪,利用快速Hough变换确定编码中心,全自动解译出路标信息,并根据编码库查询得到编码标志牌中心点的三维坐标.实验结果表明,方法可行、算法有效,为盲环境中移动目标的精确定位定姿技术开发奠定了路标基础.In order to meet the needs of navigation in blind environments lacking satellite navigation signals,a pattern recognition algorithm is proposed for positioning moving targets in a blind environment based on coded sequence pattern signposts and a monocular vision system.A new coded sequence pattern signpost carrying more valuable information was designed and fixed along the blind environment,with the signpost identified by its cental 3D coordinate.A calibrated monocular vision system was mounted on the target object to capture images with coded sequence pattern signposts.To interpret the coded sequence patterns,the SIFT（scale invariance feature transform） algorithm was used for identifying signposts,the Otsu method for image segmentation and boundary tracking,and the fast Hough algorithm for positioning signpost centers,thereby enabling inquiry into the coordinates of each signpost center according to its code pattern data base.Experimental results validated the proposed method and showed that the algorithm was effective,which established a signpost foundation of technology development of precision positioning for moving targets in blind environments.

关 键 词：机器视觉 盲环境 移动定位 序列编码图形 路标 

分 类 号：TG391.4[金属学及工艺—金属压力加工]