제목

Vision Based Navigation on Top-view Virtual Plane

저자명

Qing Lin, Youngjoon Han, Hernsoo Hanhn

초 록

  This paper presents a vision-based obstacle detection method in an outdoor sidewalk environment. One of the main problems of using single camera in outdoor navigation task is the discrimination of various types of obstacles and clutter pavement. To solve this problem, this paper uses the property that pavement edges are sub-sampled while obstacle edges are oversampled when mapping the original image to a top-view virtual plane. Morphology filters are used to enhance obstacle edges as edge-blobs with larger size, whereas sparse edges from pavement are filtered out. Based on the identified obstacles, safe walking areas are estimated by tracking a polar edge-blob histogram. The algorithm is tested in different sidewalk scenes with complex pavements, and its efficiency has been confirmed.

원문 수록처 제 7회 로봇종합학술대회 (KRoC 2012), Page 566 ~ 569
자료유형 학술대회자료