节点文献
基于规则的物体识别定位方法
A Rule-Based Approachto the Recognition and Localization of Objects
【摘要】 提出了用于物体识别定位的传感器规划方法,并利用该方法对物体进行了识别定位仿真实验。该方法利用规则在识别定位过程中动态地规划传感器的方位,较好地显示了主动视觉的作用。实验表明,该方法简明有效且具有较高的可靠性。
【Abstract】 Dynamic sensor placement is crucial to the recognition and localization of objects using computer visionprinciples.In this paper,we have proposed an approach to the sensor placement for the recognition andlocalization of objects,and developed a simulation system by use of this approach.The approach utilizesrules to dynamically plan sensor positions in the process of recognition and localization,and therefore it canwell demonstrate the effectiveness of active vision principles.Experiments have shown that the approach isquite efficient and reliable.
【关键词】 计算机视觉;
物体识别定位;
动态传感器规划;
物体模型;
特征库;
几何不变性;
【Key words】 Computer vision; Recognition and localization of objects; Dynamic sensor placement; Object models; Feature base; Geometric invariant constraints;
【Key words】 Computer vision; Recognition and localization of objects; Dynamic sensor placement; Object models; Feature base; Geometric invariant constraints;
【基金】 863计划资助项目
- 【文献出处】 高技术通讯 ,HIGH TECHNOLOGY LETTERS , 编辑部邮箱 ,1995年04期
- 【分类号】TP242.05
- 【被引频次】1
- 【下载频次】148