[1]姚莉,杜俊康,李长顺.针对伪影改善的图像拼接方法[J].郑州大学学报(工学版),2021,42(01):35-41.[doi:10.13705/j.issn.1671-6833.2021.01.006]
 Yao Li,Du Junkang,Li Changshun,et al.Image Stitching Method for Artifact Improvement[J].Journal of Zhengzhou University (Engineering Science),2021,42(01):35-41.[doi:10.13705/j.issn.1671-6833.2021.01.006]
点击复制

针对伪影改善的图像拼接方法()
分享到:

《郑州大学学报(工学版)》[ISSN:1671-6833/CN:41-1339/T]

卷:
42卷
期数:
2021年01期
页码:
35-41
栏目:
出版日期:
2021-03-14

文章信息/Info

Title:
Image Stitching Method for Artifact Improvement
作者:
姚莉杜俊康李长顺
东南大学计算机科学与工程学院;

Author(s):
Yao Li; Du Junkang; Li Changshun;
School of Computer Science and Engineering, Southeast University;

关键词:
Keywords:
DOI:
10.13705/j.issn.1671-6833.2021.01.006
文献标志码:
A
摘要:
针对图像拼接中的伪影问题,提出了一整套的拼接解决方案。图像拼接方法分为两大阶段:图像配准阶段和图像融合阶段。在图像配准阶段,提出了一种基于网格优化的图像配准方法。通过使用点特征和线特征相结合的特征提取算法,提高特征的数量和质量。同时提出了基于先验的RANSAC特征点对筛选方法,剔除误匹配点对,并提高了模型计算的速度。为了进一步提升对齐效果和优化伪影问题,提出了多项图像对齐约束。最后采用了柱面投影模型,将图像对投影到同一平面。在图像融合阶段,使用了基于图像对重叠区域细节增强的缝合线检测算法,优化多焦段场景下的拼接效果。利用渐入渐出融合算法对同一平面的投影图像进行融合,得到一幅宽视野的高分辨率图像拼接结果。通过将现有的拼接方法和本文方法进行实验数据集验证对比,结果表明本文的方法能够更好地应对伪影问题,在拼接质量上有着更优的表现。
Abstract:
This paper focuses on solving the problem of artifact in image mosaic technology, and proposes a whole set of mosaic scheme to optimize the problem. I mage stitching method is divided into two stages: image registration stage and image fusion stage. In the stage of image registration, we propose an image registration method ba<x>sed on grid optimization. By using the feature extraction algorithm combining point feature and line feature, the quantity and quality of features can be improved. At the same time, we propose a prior screening method of RANSAC feature point pairs, which can eliminate mismatching point pairs and improve the speed of model calculation. In order to further improve the alignment effect and optimize the artifacts, a number of image alignment constraints are proposed. Finally, the cylinder projection model is used to project the image pairs to the same plane. In the stage of image fusion, a seam-line detection algorithm ba<x>sed on the enhancement of the details of the overlapping areas of the image is used to optimize the stitching effect in the multi-focus scene. Finally, the gradual image fusion algorithm is used to fuse the projection image of the same plane, and a wide field of vision high-resolution image mosaic result is obtained. The existing splicing methods are compared with the method in this paper. The results show that the methods in this paper can better deal with the problem of artifacts, and have better performance in image stitching quality
更新日期/Last Update: 2021-03-15