傅建安, 万文, 熊震宇. 双能图像的亚像素匹配方法[J]. 南昌航空大学学报(自然科学版), 2015, 29(4): 40-44,84. DOI: 10.3969/j.issn.1001-4926.2015.04.008
引用本文: 傅建安, 万文, 熊震宇. 双能图像的亚像素匹配方法[J]. 南昌航空大学学报(自然科学版), 2015, 29(4): 40-44,84. DOI: 10.3969/j.issn.1001-4926.2015.04.008
FU Jian-an, WAN Wen, XIONG Zhen-yu. The Subpixel Match Method of Dual Energy Image[J]. Journal of nanchang hangkong university(Natural science edition), 2015, 29(4): 40-44,84. DOI: 10.3969/j.issn.1001-4926.2015.04.008
Citation: FU Jian-an, WAN Wen, XIONG Zhen-yu. The Subpixel Match Method of Dual Energy Image[J]. Journal of nanchang hangkong university(Natural science edition), 2015, 29(4): 40-44,84. DOI: 10.3969/j.issn.1001-4926.2015.04.008

双能图像的亚像素匹配方法

The Subpixel Match Method of Dual Energy Image

  • 摘要: 在很多应用领域,要求图像配准的精度达到亚像素级。针对X射线双能图像的特点,以双能图像为研究对象,构建了基于标准化互相关算子的相似度函数,纠正双能图像在竖直方向上存在的若干像素的偏差;建立二次插值迭代模型精确计算亚像素配准位置;最后,基于双线性插值算法实现对双能图像的亚像素平移。试验结果表明,这种方法能有效消除双能图像竖直方向上存在的若干像素偏差,精度能达到0.1级别的亚像素配准要求。

     

    Abstract: In many application fields, the image registration precision must reach subpixel level. Based on the characteristics of X-ray dual-energy image's characteristic, regrading dual energy image as the research object, the standardization of cross-correlation operator similarity function is constructed, dual-energy images the number of pixel deviation in the vertical direction is corrected. Quadratic interpolation iterative model is build to compute the subpixel registration location precisely. Finally, the bilinear interpolation algorithm is used to implement the image pixel translation. Experimental results show that this method can effectively eliminate the dual energy image of several pixels deviation in the vertical direction, precision can reach 0.1 level.

     

/

返回文章
返回