基于特征的水下图像拼接技术研究  

Underwater Image Mosaicking Technologies Based on Features

作　　者：孙杰[1] 薛纯[2] 余义德[1] 

机构地区：[1]91550部队 [2]中国海洋大学信息科学与工程学院电子工程系

出　　处：《舰船电子工程》2017年第6期118-121,共4页Ship Electronic Engineering

摘　　要：论文针对水下图像的特点,运用四种图像颜色校正的算法,针对多幅图片进行批量预处理后,根据它们能够平均得到的特征点的个数,找到最适用于水下图像拼接的算法。同时,在进行SIFT特征点描述时,不仅考虑了邻域的梯度信息,而且加入纹理信息,能够在一定程度上降低误匹配率,改善图像融合的效果。通过分析错位距离以及拼接处的色度差异,量化了评价的好坏,对最终的拼接效果有了更为客观的评价。According to the characteristics of the underwater image,this paper applies four image color correction algorithms for Batch preprocessing multiple images. Based on the average number of feature points that obtained in the pretreatment,a more suitable algorithm for underwater image mosaicking is found. At the same time,in the description of SIFT feature points,not only the gradient information of the neighborhood,but also the texture information are taken into account,which can reduce the mismatching rate to some extent and improve the effect of image fusion. The difference between the dislocation distance and the chromaticity of the splice is quantitatively analyzed,and the quality of the evaluation is quantified. The final mosaicking effect is evaluated more objectively.

关 键 词：水下图像 图像拼接 颜色校正 SIFT特征点描述 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]