摘要
电力设备的红外图像和可见光图像从不同角度反映了设备的状态,对它们进行配准可以提高在线监测的准确性。将2维经验模态分解(BEMD)方法引入到电力设备的红外与可见光图像配准中。通过理论分析得出了基于BEMD的图像配准原理,把图像的配准问题转化成了BEMD分解后子图像的配准问题,并确定了原始图像、趋势图像及2维仿射变换在配准中的关系;基于此原理提出了一种红外与可见光图像配准方法。实验结果表明该方法可取得良好的配准效果,同时也验证了该配准原理的正确性和方法的先进性。
Infrared image and visible image of power equipment can reflect its state from different points of view. The accuracy of on-line monitoring can be improved after registration of infrared/visible image. Bidimensional empirical mode decomposition （BEMD） is introduced to registration of infrared/visible image of power equipment. The principle of image registration based on BEMD is given by theoretic analysis. The registration of image is changed into registration of their sub-images after BEMD decomposed. And the mathematical relation among original images, remnant images and two-dimensional affine transformation in image registration is determined. On the basis of the registration principle, an infrared/visible image registration method is proposed. The experimental results demonstrate that this method can obtain better registration effect and that the registration principle is correct and the proposed method is advanced.
出处
《电力系统保护与控制》
EI
CSCD
北大核心
2011年第23期25-29,共5页
Power System Protection and Control
基金
中央高校基本科研业务费专项资金(09QG10)~~
关键词
2维经验模态分解
图像配准
红外图像
可见光图像
电力设备
小波分解
bidimensional empirical mode decomposition
image registration
infrared image
visible image
power equipment
wavelet decomposition
