摘要
针对传统阈值法难以快速准确分割感光不均匀数字PCR(Polymerase Chain Reaction)荧光图像的问题,提出了一种两步分割算法。第一步,对预处理后的图像采用Otsu算法进行分割;第二步,提取第一步分割不完全的区域,基于图像复杂度确定待分割区域的灰度范围,并基于灰度统计信息计算最优分割阈值,分割结果与第一步分割结果合并形成最终分割图像。实验结果表明,该算法能够有效减少不均匀感光对数字PCR荧光图像分割的影响,分割结果准确,算法复杂度低且鲁棒性好,为有效腔室数量的统计提供了保障,进而提升了数字PCR绝对定量的准确度。
For the problem that the traditional threshold method is difficult to segment the non-uniform digital Polymerase Chain Reaction(PCR)fluorescence image quickly and accurately,a two-step segmentation algorithm is proposed.In the first step,Otsu algorithm is used to segment the preprocessed image.In the second step,the incomplete segmentation region in the first step is extracted,the gray range of the region to be segmented is determined based on the image complexity,and the optimal segmentation threshold is calculated based on the gray statistical information,and the segmentation result is combined with the segmentation result in the first step to form the final segmentation image.The experimental results show that the algorithm can effectively reduce the influence of uneven sensitivity on the segmentation of digital PCR fluorescence image.The segmentation results are accurate,the algorithm complexity is low and the robustness is good.It provides a guarantee for the statistics of the number of effective chambers,and further improves the accuracy of the absolute quantification of digital PCR.
作者
余跃
唐骏
肖旻
YU Yue;TANG Jun;XIAO Min(Department of Communication Engineering,Xiamen University of Technology,Xiamen 361024,China)
出处
《电讯技术》
北大核心
2022年第1期123-129,共7页
Telecommunication Engineering
基金
国家自然科学基金资助项目(61701422)
福建省自然科学基金面上项目(2018J01569)。
作者简介
余跃,男,1995年生于安徽安庆,硕士研究生,主要研究方向为图像处理;通信作者:唐骏,男,1977年生于湖南永州,2018年于厦门大学获博士学位,现为副教授、硕士生导师,主要研究方向为雷达信号处理、图像处理;肖旻,女,1982年生于福建龙岩,2011年于厦门大学获博士学位,现为副教授、硕士生导师,主要研究方向为信道编码、信源信道联合编码、网络编码。
