摘要
基于条件随机场,提出一种由近及远的在线、自适应大范围场景地形感知框架.首先,把当前场景图片划分为超像素,将近视场超像素的特征向量和地形类别作为学习样本整合到地形数据库中;然后,利用条件随机场和地形数据库对远视场超像素的特征信息和空间关系进行建模;最后,利用在线学得的模型参数对远视场超像素所属地形类别进行推理.分类实验结果表明,该方法相对已有的其他方法在分类的精度、鲁棒性以及对动态环境的自适应能力三方面均有极大提高.
Based on conditional random fields(CRFs),an online,adaptive,and near-to-far long-range terrain perception approach is proposed.First,the current image is segmented into superpixels,and feature vectors and terrain categories of near-field superpixels are incorporated into terrain database as learning samples.Second,superpixel features and spatial relationships between far-field superpixels are modeled by using conditional random fields and terrain database.Finally,terrain categories of far-field superpixels are inferred based on the on-line learned model parameters.Experimental results show that the proposed approach outperforms other existing methods in terms of accuracy,robustness and adaptability to dynamic environments.
作者
王明军
周俊
屠珺
刘成良
WANG Mingjun;ZHOU Jun;Tu Jun;LIU Chengiang(Shanghai Jiaotong University,Shanghai 200240,China;Nanjing Agricultural University,Nanjing 210031,China)
出处
《机器人》
EI
CSCD
北大核心
2010年第3期326-333,共8页
Robot
基金
国家863计划资助项目(2006AA10A304
2006AA10Z259
2008AA100905)
关键词
移动机器人
视觉导航
条件随机场
mobile robot vision navigation
conditional random field
