基于机械扫描与改进SURF的显微图像拼接算法  被引量:5

Microscopic Image Stitching Algorithm Based on Mechanical Scanning and Improved SURF

在线阅读下载全文

作  者:柴政 刘任任[1] 梁光明 CHAI Zheng;LIU Ren-ren;LIANG Guang-ming(School of Information Engineering,Xiangtan University,Xiangtan 411105,China;Hunan USER Technology Co.,Ltd.,Changsha 410075,China)

机构地区:[1]湘潭大学信息工程学院,湖南湘潭411105 [2]湖南友哲科技有限公司,湖南长沙410075

出  处:《计算机技术与发展》2019年第11期97-101,共5页Computer Technology and Development

基  金:湖南省自然科学基金(2016GK4014)

摘  要:针对传统图像拼接算法拼接速度慢,难以实现显微图像实时拼接的难点问题,通过分析显微镜扫描的规律和显微图像的特点,结合Canny边缘检测、线性搜索法、SURF及最小二乘法,提出了一种适用于显微图像的快速拼接算法。根据已知的重叠区域,对低倍镜图像和高倍镜图像分别使用Canny边缘检测和线性搜索法,获得特征点相对集中的区域,使用SURF对其进行特征点提取,并采用快速最近邻搜索算法匹配特征点。由于显微镜的震动较为规律,对于特征点稀少,容易出现误匹配的重叠区域,为了减小算法的复杂度,利用最小二乘法,根据前后拼接点的位置信息,进行预测拼接。显微镜的扫描过程可以视为只有平移,没有旋转,运用了对平移图像融合效果好、速度快的加权平滑融合算法消除拼接缝隙。实验结果表明,该算法拼接速度快,具有很强的鲁棒性和稳定性,可以很好地实现显微图像的实时拼接。Aiming at the difficulty of traditional image stitching algorithm which has a slow stitching speed and is difficult to realize real-time microscopic image stitching,we propose a fast stitching algorithm for microscopic images by analyzing the rules of microscope scanning and characteristics of microscopic images,combining Canny edge detection,linear search,SURF and least square method.According to known overlapping regions,Canny edge detection and linear search are used for low magnification and high magnification respectively to obtain the region with relative concentration of feature points.SURF is used to extract feature points,and fast nearest neighbor search algorithm is used to match feature points.Because the vibration of microscope is more regular,there are few feature points,and the overlapping areas of mismatching are easy to occur.To reduce the complexity of the algorithm,the least squares method is used to predict the stitching points according to the position information of the splicing points.The scanning process of a microscope can be regarded as translation without rotation.A weighted smoothing fusion algorithm is used to eliminate the splicing gap.The experiment shows that the proposed algorithm is fast,robust and stable,and can realize the real-time mosaic of microscopic images.

关 键 词:图像拼接 SURF 线性搜索法 最小二乘法 CANNY 

分 类 号:TP306.1[自动化与计算机技术—计算机系统结构]

 

参考文献:

正在载入数据...

 

二级参考文献:

正在载入数据...

 

耦合文献:

正在载入数据...

 

引证文献:

正在载入数据...

 

二级引证文献:

正在载入数据...

 

同被引文献:

正在载入数据...

 

相关期刊文献:

正在载入数据...

相关的主题
相关的作者对象
相关的机构对象