[1]覃 丹,崔 岩,刘兴林.基于经纬映射与透视投影的球形图像校正[J].计算机技术与发展,2020,30(11):94-99.[doi:10. 3969 / j. issn. 1673-629X. 2020. 11. 018]
　QIN Dan,CUI Yan,LIU Xing-lin.Spherical Image Correction Based on Latitude and Longitude Mapping and Perspective Projection[J].,2020,30(11):94-99.[doi:10. 3969 / j. issn. 1673-629X. 2020. 11. 018]
点击复制

基于经纬映射与透视投影的球形图像校正()
分享到：

《计算机技术与发展》[ISSN:1006-6977/CN:61-1281/TN]

卷:
30
期数:
2020年11期
页码:
94-99
栏目:
智能、算法、系统工程
出版日期:
2020-11-10

文章信息/Info

Title:
Spherical Image Correction Based on Latitude and Longitude Mapping and Perspective Projection
文章编号:
1673-629X(2020)11-0094-06
作者:
覃 丹1; 崔 岩1; 2; 刘兴林1
1. 五邑大学 智能制造学部,广东 江门 529000; 2. 珠海四维时代网络科技有限公司,广东 珠海 519000
Author(s):
QIN Dan1; CUI Yan1; 2; LIU Xing-lin1
1. Faculty of Intelligent Manufacturing,Jiangmen University of Wuyi,Jiangmen 529000,China; 2. 4DAGE Technology Co. ,Ltd. ,Zhuhai 519000,China
关键词:
球形图像; 鱼眼图像; 经纬映射; 球面透视投影; 信息熵
Keywords:
spherical image; fisheye image; longitude and latitude mapping; spherical perspective projection; information entropy
分类号:
TP309
DOI:
10. 3969 / j. issn. 1673-629X. 2020. 11. 018
摘要:
图像校正问题是图像处理领域内一个普遍存在的问题,由于成像系统的像差、畸变和带宽、摄像机姿态和成像设备扫描的非线性、运动模糊、辐射失真和噪声引入造成的图像失真都会给图像处理带来一定阻碍,所以需要通过图像校正将畸形图像恢复为原始图像的外观。 由特殊镜头所拍摄得到的广角 360o?视场的球形全景图虽然能将更多的场景容纳到视场中,但是其图像也产生了一定程度的畸变。 为了解决广角 360o?视场的球形图像畸变的问题,提出一种基于经纬映射与透视投影的球形图像校正方法。 通过改进的经纬映射法和改进的球面透视投影法对球形图像进行校正处理,将球形图像转化为复杂度较低的鱼眼图像,然后再进一步实验达到最终的校正效果,以更适应人的视觉效果。 与其他两种校正图像的算法比较表明,该方法降低了图像的扭曲程度,建立映射关系的时间更少。
Abstract:
Image correction problem is a common problem in the field of image processing. The image distortion caused by introduction of the imaging system aberration,distortion and bandwidth, camera and imaging equipment scanning nonlinearity,motion blur,radiation distortion and noise will bring certain obstacles to the image processing,so it is necessary to restore the deformed image to the appearance of the original image through image correction. Although the spherical panorama with a wide angle?360o field of view captured by the special lens can accommodate more scenes into the field of view,the image also produces a certain degree of distortion. In order to solve the problem of spherical image distortion in wide-angle?360o field of view,a spherical image correction method based on longitude and latitude mapping and perspective projection is proposed. The spherical image is corrected by the improved longitude and latitude mapping method and the improved spherical perspective projection method,and the spherical image is transformed into a fish-eye image with low complexity. Compared with the other two image correction algorithms,the proposed algorithm reduces the distortion degree of the image and takes less time to establish the mapping relation.

相似文献/References:

[1]斯闻豪,杨庚. 一种基于经线模型的鱼眼图像校正算法[J].计算机技术与发展,2014,24(10):38.
　SI Wen-hao,YANG Geng. A Fisheye Image Correction Method Based on Longitude Model[J].,2014,24(11):38.
[2]国艺美,冯时雨,毕 波*.一种未校正鱼眼图像的特征匹配的组合算法[J].计算机技术与发展,2022,32(S2):64.[doi:10. 3969 / j. issn. 1673-629X. 2022. S2. 011]
　GUO Yi-mei,FENG Shi-yu,BI Bo *.A Combination Algorithm for Feature Matching of Uncorrected Fisheye Images[J].,2022,32(11):64.[doi:10. 3969 / j. issn. 1673-629X. 2022. S2. 011]
[3]韩 叙,席 萌.面向 3D 动画球幕电影的鱼眼图像制作方法[J].计算机技术与发展,2023,33(11):209.[doi:10. 3969 / j. issn. 1673-629X. 2023. 11. 031]
　HAN Xu,XI Meng.Method for Making Fisheye Images for 3D Animated Dome Shows[J].,2023,33(11):209.[doi:10. 3969 / j. issn. 1673-629X. 2023. 11. 031]

更新日期/Last Update: 2020-11-10