光学 精密工程, 2016, 24 (11): 2830, 网络出版: 2016-12-26   

采用显著性分析与改进边缘方向直方图特征的红外与可见光图像配准

Registration of infrared image and visible image based on saliency and EOH feature analysis
作者单位
1 武汉大学 测绘遥感信息工程国家重点实验室, 湖北 武汉 430079
2 北京市混合现实与新型显示工程技术研究中心 北京理工大学 光电学院, 北京 100081
3 国家质量监督检验检疫总局 信息中心, 北京 100088
摘要
为了实现红外图像与可见光图像的信息融合, 弥补单一模态图像的不足, 提出了一种基于显著性分析与改进的边缘方向直方图EOH(Edge Orientation Histogram)特征的红外与可见光图像配准算法。该算法首先利用显著性分析技术找到可见光图像中的重要信息, 得到显著性图; 将其与可见光图像融合, 实现可见光图像中重要信息的划分。然后, 利用自适应FAST(Features from Accelerated Segment Test)算法, 探测可见光与红外图像上的特征点; 利用改进的EOH, 描述特征点。最后, 根据描述计算特征点的相似性, 在可见光与红外图像上找出对应的特征点, 实现红外与可见光图像的匹配。在3种不同情况下对红外与可见光图像数据进行了配准实验。结果表明: 在红外图像与可见光图像采集条件相似情况下, 特征点正确匹配率为96.55%, 而在图像采集条件差异较大的情况下, 特征点正确匹配率可达74.21%。该算法可实现红外与可见光图像的精确快速匹配, 即使红外图像与可见光图像采集的角度与位置均存在较大差异的情况下, 仍可以满足红外与可见光图像匹配对精度和稳定性的要求。
Abstract
To realize the information fusion of infrared and visible images and make up the deficiency of the single modality image, a new algorithm based on saliency and Edge Orientation Histogram(EOH) features was proposed. Firstly, the saliency analysis was used to find the important information of the visible image and to obtain the saliency map. By fusing it with the visible image, the important information in the visible image was divided. Then, adaptive Features from Accelerated Segment Test(FAST) algorithm was employed in detecting feature points on the visible image and infrared image, and the improved EOH was used to describe the detected feature points. Finally, corresponding feature points were found by calculating the similarity of feature points in the visible and infrared images and the infrared and visible images were matched. An image matching experiments at three conditions were carried out, and the results indicate that when the collection conditions between the infrared and visible images are similar, the feature matching accuracy reaches 96.55%. When the difference of collection conditions between the infrared and visible images is large, the feature matching accuracy still can reach 74.21%. The algorithm realizes fast and accurate matching of infrared and visible images, and meets the requirements of image matching for accuracy and stability, especially under a collection condition that the infrared and visible images are bigger different.

徐军, 付天宇, 杨健, 丰苏. 采用显著性分析与改进边缘方向直方图特征的红外与可见光图像配准[J]. 光学 精密工程, 2016, 24(11): 2830. XU Jun, FU Tian-yu, YANG Jian, FENG Su. Registration of infrared image and visible image based on saliency and EOH feature analysis[J]. Optics and Precision Engineering, 2016, 24(11): 2830.

本文已被 2 篇论文引用
被引统计数据来源于中国光学期刊网
引用该论文: TXT   |   EndNote

相关论文

加载中...

关于本站 Cookie 的使用提示

中国光学期刊网使用基于 cookie 的技术来更好地为您提供各项服务,点击此处了解我们的隐私策略。 如您需继续使用本网站,请您授权我们使用本地 cookie 来保存部分信息。
全站搜索
您最值得信赖的光电行业旗舰网络服务平台!