首页> 美国政府科技报告 >Scale-Space-Based Visual-Motion-Cue for Autonomous Navigation
【24h】

Scale-Space-Based Visual-Motion-Cue for Autonomous Navigation

机译:基于尺度空间的自主导航视觉运动提示

获取原文

摘要

This paper presents a new visual motion cue, the authors call the Hybrid VisualThreat Cue (HVCT) that provides some measure for a change in relative range as well as absolute clearances, between a 3D surface and a fixated observer in motion. The visual field associated with the HVTC can be used to demarcate the regions around a moving observer into safe and danger zones of varying degree, which may be suitable for autonomous navigation tasks, in particular collision avoidance and maintenance of clearance. The HVTC is independent of the 3D environment and needs almost no a-priori information about it. It is rotation independent, and is measured in (time(sup -1)) units. When there is a relative motion between a fixation point on a 3D surface and an observer, the perceived texture details the image vary. The rate at which the details vary provides an indication of the observer's relative motion with respect to the 3D surface. Scale space representation which is a multiscale approach provides a concrete way to analyze the variations of image details The authors derive a relation between the relative temporal variations of the image inner scale and the HVTC. A practical method to extract the HVTC from a sequence of images of a 3D textured surface obtained by a fixated, fixed-focus monocular camera in motion is also presented. A global dissimilarity measure is extracted directly from the raw data of the gray level of textured images from which the HVTC is obtained. This approach of extracting the HVTC is independent of the type of 3D surface texture and needs no optical flow information, 3D reconstruction, segmentation, feature tracking. It needs almost no camera calibration. This algorithm to extract the HVTC was applied to a set of twelve different texture patterns (of 3D scenes) from the Brodatz's album, where the authors observed a similar behavior for most of the textures.

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号