摘要
为了将紫外图像和可见光图像进行融合让人们及时清晰地观测到设备故障,该文提出了一种基于相机模型和相机间透视变换的图像配准方法和相应的标定装置,能够实现高精度、小计算量的高效紫外图像和可见光图像的对齐。紫外和可见光的信息融合不仅能够探测目标发射的紫外光谱,还可以对目标进行清晰成像、精确定位,生成的多光谱图像改善单一光谱成像信息的局限性。利用多源图像的信息互补,提高图像检测的准确性、及时性,也更适合计算机的后续分析与处理。
For the fusion of the ultraviolet image and the visible light image can facilitate people to find the equipment damage quickly and clearly.In this paper,an image registration and fusion method based on the camera model and the homograph relationship between cameras was proposed,which can perform the registration of real-time ultraviolet images and visible images with high precision and low computation complexity.Multi-spectral information fusion can not only detect the ultraviolet spectrum of target emission,but also clearly image and accurately locate the problem target,generate multi-spectral images, improve the performance of single spectral image information.With the complement information from different optical spectrum,the accuracy and timeliness of image detection were improved,and the fusion was also helpful for the subsequent data analysis and processing of computers.
作者
陈锦龙
陈俊全
张旭
吴珂
王兴国
CHEN Jinlong;CHEN Junquan;ZHANG Xu;WU Ke;WANG Xingguo(Guizhou Power Grid Co.,Ltd.,Guiyang 550002,China;Nanjing GM Innovation Technology Co.,Ltd.,Nanjing 210014,China)
出处
《测绘科学》
CSCD
北大核心
2019年第8期70-74,共5页
Science of Surveying and Mapping
基金
南方电网有限责任公司科技项目(066500KK52170034)
贵州省科学技术基金项目(黔科合基础[2016]1036)
关键词
多光谱
图像配准
图像处理
紫外
放电检测
multispectral
image registration
image processing
ultraviolet
discharge detection
