Title :
Fast and auto-adaptative morphological segmentation for implementation in a 3-D vision sensor
Author :
Quiguer, Thieny ; Miche, P. ; Debrie, Roland
Author_Institution :
INSA de ROUEN, Mont Saint Aignan, France
Abstract :
An intelligent 3-D vision sensor is under development in the authors´ laboratory for applications in robot guidance and autonomous vehicle control. The aim in this context is to compute a depth map from two images (512×512×8 bits) in less than 0.1 seconds. The state of the field indicates the necessity for the development of new, fast, auto-adaptive operators to segment images meeting two constraints: short computational times, and automatic primitive extraction. For this reason the authors have introduced a new operator to segment image-lines: a morphological gradient with moving kernel dimension. This operator is described
Keywords :
CCD image sensors; computer vision; image segmentation; mathematical morphology; robots; 3-D vision sensor; CCD image sensors; auto-adaptative morphological segmentation; automatic primitive extraction; autonomous vehicle control; depth map; image line segmentation; intelligent sensor; morphological gradient; moving kernel dimension; robot guidance; Automatic control; Image segmentation; Intelligent robots; Intelligent sensors; Intelligent vehicles; Mobile robots; Navigation; Remotely operated vehicles; Robot sensing systems; Robot vision systems;
Conference_Titel :
Industrial Electronics, 1992., Proceedings of the IEEE International Symposium on
Conference_Location :
Xian
Print_ISBN :
0-7803-0042-4
DOI :
10.1109/ISIE.1992.279550
