双目视觉下红外遥感舰船图像三维重构设计研究  

Research on three-dimensional reconstruction design of infrared remote sensing ship image under binocular vision

作　　者：赵金 ZHAO Jin(Zhengzhou University of Science and Technology,Zhengzhou 450000,China)

机构地区：[1]郑州科技学院,河南郑州450000

出　　处：《舰船科学技术》2023年第22期202-205,共4页Ship Science and Technology

基　　金：郑州市社会科学调研课题(ZSLX20210678)。

摘　　要：为提升舰船三维重构结果的细节丰富度,在双目视觉支持下提出红外遥感舰船图像三维重构设计方法。通过像素坐标系与世界坐标系的转换,融合可见光图像与红外遥感图像的点云信息。从可见光图像与红外遥感图像融合结果中提取舰船目标的质心、质心区域灰度以及舰船区域面积的特征点。利用半全局匹配算法,匹配提取特征点。依据匹配结果采用三角化曲面算法实现舰船图像三维重构。实验结果表明,该方法重构获取的舰船图像,细节丰富,未出现空洞或细节丢失情况。To enhance the detail richness of ship 3D reconstruction results,a design method for infrared remote sensing ship image 3D reconstruction is proposed with the support of binocular vision.By converting the pixel coordinate system to the world coordinate system,the point cloud information of visible light images and infrared remote sensing images is fused.Extract the centroid,centroid region grayscale,and ship area feature points of ship targets from the fusion results of visible light images and infrared remote sensing images.Utilize semi global matching algorithm to extract feature points for matching.Based on the matching results,the triangulation surface algorithm is used to achieve three-dimensional reconstruction of ship images.The experimental results show that the reconstructed ship image obtained by this method has rich details and no voids or loss of details.

关 键 词：双目视觉 红外遥感 舰船图像 三维重构 特征基元 三角化曲面 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]