摘要
提出了一种不同尺度的图像配准与自适应拼接算法。通过特征点匹配与图像间变换几何参数的自动辨识,实现了不同尺度和旋转图像间的匹配和拼合。利用尺度不变特征点提取算法(SIFT)提取出待拼接图像的特征点,根据互信息量最大原则实现特征点的匹配,然后通过得到的匹配对的几何信息自动辨识出两幅图像之间的变换参数关系,得到反映图像平移、旋转和尺度变换信息的投影变换矩阵,最终实现图像的无缝融合拼接。仿真结果表明:与现有算法相比,该算法较好地解决了不同尺度图像间的配准与自适应拼合问题。
An image matching and adaptive mosaic algorithm was proposed.The images with different scales and rotations were matched and stitched through the feature points matching and automatic identification of transform geometrical parameters between two images.The scale invariant feature transform（SIFT） algorithm was used to extract feature points of images,the matches was realized by use of rule of the maximal mutual information,then the transform parameters relation between the two images were identified automatically on the basis of geometry information of matching points,the projection transform matrix which reflected the displacement,rotation and scale change was also obtained,and finally the two images were stitched well and truly.The simulation results show that the proposed method can resolve the image matching and adaptive mosaic problem with varying scales effectively compared with the existing methods.
出处
《红外与激光工程》
EI
CSCD
北大核心
2011年第3期564-569,共6页
Infrared and Laser Engineering
基金
国防预研基金资助项目(9140A16050109DZ01
9140A16050310DZ01)
国家部委十一五科技项目(51316060205)
中央高校基本科研业务费专项资金资助项目(JY10000904017)
关键词
图像配准
图像拼接
尺度变化
参数辨识
image matching
image mosaic
scale change
parameter identification
