摘要
为了实现多幅连续图像的快速、准确、完整地拼接,使用MSER算法快速、精确地检测图像的特征点,通过预先确定重叠范围,减少无效特征点的提取。利用非下采样剪切波变换(Non-Subsampled Shearlet Transform,NSST)进行图像融合,保证了融合效果。实验证明,该算法是一种快速有效、具有良好鲁棒性及融合效果的大视场高分辨率图像拼接方法。
In order to realize the fast,accurate and complete mosaic of multiple continuous images,this paper uses MSER algorithm to detect the feature points of the image quickly and accurately,and reduces the extraction of invalid feature points by determining the overlapping range in advance;and uses Non-Subsampled Shearlet Transform(NSST)for image fusion to ensure the fusion effect.The experimental results show that the algorithm is a fast and effective image mosaic method with good robustness and fusion effect.
作者
庞丽东
潘维东
林博文
PANG Lidong;PAN Weidong;LIN Bowen(School of Architectural Engineering,Jinggangshan University,Ji’an 343009;School of Mechanical and Electrical Engineering,Jinggangshan University,Ji’an 343009)
出处
《现代制造技术与装备》
2021年第5期4-5,共2页
Modern Manufacturing Technology and Equipment
基金
2018年度江西省文化艺术科学规划项目“大幅面文物数字化保护关键技术研究”(YG2018168)。
