许金鑫,李庆武,马云鹏,钱荣.基于斜率一致性的电气设备红外与可见光图像配准方法[J].光电子激光,2017,28(7):794~802 |
基于斜率一致性的电气设备红外与可见光图像配准方法 |
Registration method between infrared and visible images of electrical equipment based on slope consistency |
投稿时间:2016-08-19 |
DOI: |
中文关键词: 红外与可见光图像 配准 斜率一致性 粗边缘 SURF算法 |
英文关键词:infrared and visible image registration slope consistency rough ed ges SURF algorithm |
基金项目:江苏省重点研发计划-产业前瞻与共性关键技术(SBE2016000963)资助项目 (1.河海大学 物联网工程学院,江苏 常州 213022; 2.常州市传感网与环境感知重点实 验室,江苏 常州 213022; 3.江苏久创电气科技有限公司,江苏 常州 213022) |
|
摘要点击次数: 1181 |
全文下载次数: 0 |
中文摘要: |
针对电气设备同一场景间红外与可见光图像间难以 匹配的问题,提出了一种基于斜率一致性的图像配准方法。首先通 过基于多方向结构元素、不同权值的数学形态学边缘检测算法分别提取红外与可见光图像的 边缘,得到粗边缘图像; 然后通过SURF算法检测两幅边缘图像的特征点,根据正确的匹配点对之间斜率一致性的先 验知识,进行特征点匹配; 最后通过最小二乘法求得仿射变换模型参数实现两幅图像的配准。实验结果表明,本文方法 有效提高了匹配点对的正 确率,特征点的定位也更加精确,能够对电气设备红外和可见光图像实现高精度的配准 。 |
英文摘要: |
In terms of the difficulties to match between infrared and visible images of electrical equipment in the same scene,a registration method based on slope consistency is proposed.Firstly,the edges of infrared and visible images are extracted by the edge detection algorit hm mathematical morphology based on the multi-directional structure elements and different weights,and the rough e dges are obtained.Then,the feature points of two edge images obtained by the above method are detected respectively by speeded up robust features (SUR F) algorithm,and the feature points are matched by the prior knowledge of the s lope consistency between the correct matching points.Finally,affine transformation parameters are obtained by a least-squar es solution,and the registration between the infrared and visible image is achieved by using these p arameters.Experimental results show that the proposed method can effectively improve the accuracy of ma tching points and make the location of the feature points be more precise,which can achieve high pr ecision in the registration of infrared and visible images of electrical equipme nt. |
查看全文 下载PDF阅读器 |
关闭 |
|
|
|
|
|