改进结构相似度的红外两波段图像目标配准

Target registration for two wave band infrared images based on improved structural similarity

针对红外两波段图像目标配准问题,研究了目标整体结构特征,在高斯尺度空间对结构相似度、多尺度结构相似度和边缘结构相似度进行改进,提出基于改进的多尺度边缘结构相似度算法的红外两波段图像目标配准算法。利用高斯尺度空间中不同尺度的图像及其边缘图像计算多尺度结构相似度,结合了红外图像中保留的目标完整的边缘结构信息,提高了目标配准效果。实验结果表明基于改进的多尺度边缘结构相似度算法的红外两波段图像目标配准算法的目标配准正确率和目标配准精度都优于其他算法,目标相似度值高于0.98。

In order to register targets between two wave band infrared images, the overall structure feature is studied in this paper. A novel target registration algorithm for two wave band infrared images based on improved multi-scale edge structural similarity is proposed by means of improving structural similarity algorithm, multi-scale structural similarity algorithm and edge structural similarity algorithm in Gauss scale space. Improved rnuhi-scale edge structural similarity is calculated through using images of different scales and their edge images in Gauss scale space. Then the performance of target registration can be improved due to that target＇ s complete edge structure infomlation in the infrared images is utilized fully. The experimental results show that the accuracy of target registration based on the proposed method is better than other algorithms, and the target similarity between two wave band infrared images is higher than 0.98.