인공 구조물 환경에서 자율 수중 로봇의 위치 인식을 위한 크기 변화에 강인한 영상 기반 물체 인식

Abstract

This paper introduces vision-based object detection techniques for underwater localization. The accurate localization is very essential to make autonomous underwater robots perform underwater tasks completely. However the sensors or environments for the accurate localization are not available. So, in the previous work[7], we proposed the weighted sum-based correlation coefficient for template matching technique and multi-template based section method for object detection using a camera in structured environment. However, the performance was not satisfactory in case that the size of object is varying. In this paper, the object detection method has been enhanced to be robust in scale change by employing color-based segmentation. For the segmentation techniques, we have assumed deep-sea environment that the light source installed on the AUV is dominant and the hue and saturation of the artificial landmark is higher than those of environment. Using the assumptions, the regions of which hue and saturation are higher than background have been extracted, and some candidate regions have been selected considering the size and aspect ratio of those. The selected candidate regions have been scaled as template size and classified relying on the correlation coefficients calculated by weighted correlation coefficient-based template matching technique. The performance of the proposed algorithm has been demonstrated through the waments for the accurate localization are not available. So, in the previous work[7], we proposed the weighted sum-based correlation coefficient for template matching technique and multi-template based section method for object detection using a camera in structured environment. However, the performance was not satisfactory in case that the size of object is varying. In this paper, the object detection method has been enhanced to be robust in scale change by employing color-based segmentation. For the segmentation techniques, we have assumed deep-sea environment that the light source installed on the AUV is dominant and the hue and saturation of the artificial landmark is higher than those of environment. Using the assumptions, the regions of which hue and saturation are higher than background have been extracted, and some candidate regions have been selected considering the size and aspect ratio of those. The selected candidate regions have been scaled as template size and classified relying on the correlation coefficients calculated by weighted correlation coefficient-based template matching technique. The performance of the proposed algorithm has been demonstrated through the wa