期刊名称:International Journal of Innovative Research in Computer and Communication Engineering
印刷版ISSN:2320-9798
电子版ISSN:2320-9801
出版年度:2016
卷号:4
期号:4
页码:7960
DOI:10.15680/IJIRCCE.2016.0404343
出版社:S&S Publications
摘要:Range scanning methods for underwater images are gaining interest in exploring underwater. Different tools have come up for representing seafloors. These scan images or videos captured by underwater cameras or robots usually results in an unstructured cloud of points. These may also have few defects like noise a nd outliers present in this kind of extreme environment. These problems can be solved by approximating these 3D points. Hence in this paper we presented an efficient 3D reconstruction system. Proposed method enhances the input images using CLAHE and makes use of points obtained using SURF descriptor to work along with 3D triangulation. Final reconstruction is done using Delaunay method. It is used in many applications. This algorithm is evaluated on underwater video for reconstruct the 3D view of it