CAJ | 학술논문

随着对地观测技术的发展, 光学遥感图像分辨率不断提高, 使得利用舰船局部特征进行舰船的精细识别成为可能. 基于经典的SIFT特征提取算法对高分辨率光学遥感图像中的舰船目标进行了特征点提取、特征点描述和特征匹配研究, 通过同一类型舰船目标切片的匹配实验, 结果表明SIFT算法可以有效用于遥感图像舰船切片的特征点提取和匹配, 证明了该算法在高分辨率遥感图像中舰船目标精细识别的应用潜力.
수착대지관측기술적발전, 광학요감도상분변솔불단제고, 사득이용함선국부특정진행함선적정세식별성위가능. 기우경전적SIFT특정제취산법대고분변솔광학요감도상중적함선목표진행료특정점제취、 특정점묘술화특정필배연구, 통과동일류형함선목표절편적필배실험, 결과표명SIFT산법가이유효용우요감도상함선절편적특정점제취화필배, 증명료해산법재고분변솔요감도상중함선목표정세식별적응용잠력.
Along with the development of earth observation technology,the spatial resolution of optical remote sensing image is becoming increasingly high,which leads to the possibility of identifying ships using the local features of ship slices. The feature extrac-tion,description and matching of ship slices of high-resolution optical remote sensing images are studied using the classical SIFT meth-od.The ship matching experiment of the same type shows that the SIFT method is valid for feature extraction and matching of ship slices,and it demonstrates the application potential of this method in ship identification of high-resolution remote sensing images.