首页 > 论文 > 激光与光电子学进展 > 56卷 > 15期(pp:151002--1)

基于区域蛙跳搜索与轮廓匹配的显微图像拼接

Microscopic Image Stitching Based on Regional Frog Leaping Search Algorithm and Image Contour Matching

  • 摘要
  • 论文信息
  • 参考文献
  • 被引情况
  • PDF全文
分享:

摘要

为解决传统显微图像拼接中产生的几何畸变和错位,及特征稀少造成的正确匹配率低、时效性差等问题,提出基于区域蛙跳搜索和图像轮廓匹配的拼接算法。提取连续采集且有重叠区域的图像轮廓曲线;引入轮廓线索感知相似度和均方误差距离,计算图像轮廓曲线间的相似度或曲线离散距离,并将其作为匹配的衡量指标;在决策域内采用区域蛙跳算法更新鸣叫分贝和蛙跳策略,搜索图像轮廓最优匹配,实现图像快速精确的拼接。结果表明,所提算法不仅具有较高的拼接精度和较强的稳健性,还减小了其简化匹配策略的计算量,具有较强的时效性。

Abstract

To solve geometric distortion and misalignment generated in the traditional microscopic image stitching process and avoid low matching rate and high computational burden caused by rare features, a stitching algorithm based on the regional frog leaping search algorithm and image contour matching is proposed. First, the image contour curves of continuous acquisition and overlapping regions were extracted. Then, the clue-aware trajectory similarity or mean square error distance was introduced to calculate the similarity and discrete distance between image contour curves. Finally, the regional frog leaping algorithm was used in the decision domain, and by updating the tweet decibel and frog leaping strategy, optimal contour matching for the image was searched. Consequently, the images were stitched quickly and accurately. Experimental results verify that the proposed algorithm demonstrates high stitching precision and strong robustness. In addition, the simplified matching strategy reduces the amount of calculation and exhibits strong timeliness.

Newport宣传-MKS新实验室计划
补充资料

DOI:10.3788/LOP56.151002

所属栏目:图像处理

基金项目:国家自然科学基金(21365008,61105004);

收稿日期:2019-01-17

修改稿日期:2019-03-04

网络出版日期:2019-08-01

作者单位    点击查看

颜振翔:桂林电子科技大学电子工程与自动化学院, 广西 桂林 541004
王寒迎:桂林电子科技大学电子工程与自动化学院, 广西 桂林 541004
石齐双:桂林电子科技大学计算机与信息安全学院, 广西 桂林 541004
莫艳红:桂林电子科技大学电子工程与自动化学院, 广西 桂林 541004
杨辉华:桂林电子科技大学计算机与信息安全学院, 广西 桂林 541004北京邮电大学自动化学院, 北京 100876

联系人作者:颜振翔(670927905@qq.com)

备注:国家自然科学基金(21365008,61105004);

【1】Qiu G Q, Feng H Q, Jiang T Y et al. Improved image mosaic algorithm based on Harris corner. Computer Science. 39(11), 264-266, 297(2012).
仇国庆, 冯汉青, 蒋天跃 等. 一种改进的Harris角点图像拼接算法. 计算机科学. 39(11), 264-266, 297(2012).

【2】Xu J J. Fast image registration method based on Harris and SIFT algorithm. Chinese Optics. 8(4), 574-581(2015).
许佳佳. 结合Harris与SIFT算子的图像快速配准算法. 中国光学. 8(4), 574-581(2015).

【3】Xu X, Sun S Y, Sha Y J et al. A method of infrared image mosaic based on improved RANSAC. Laser & Optoelectronics Progress. 51(11), (2014).
徐鑫, 孙韶媛, 沙钰杰 等. 一种基于改进RANSAC的红外图像拼接方法. 激光与光电子学进展. 51(11), (2014).

【4】Huang Q D, Qiu Y H and Tian X P. Image automatic mosaic method based on feature points and optimization theory. Acta Photonica Sinica. 38(8), 2139-2143(2009).
黄琼丹, 邱跃洪, 田小平. 基于特征点及优化理论的图像自动拼接方法. 光子学报. 38(8), 2139-2143(2009).

【5】Feng J B, Su Z X and Liu X P. An similar-curve based automatic mosaic algroithm of panoramic image. Chinese Journal of Computers. 26(11), 1604-1608(2003).
封靖波, 苏志勋, 刘秀平. 一种基于相似曲线的全景图自动拼接算法. 计算机学报. 26(11), 1604-1608(2003).

【6】Hung C C, Peng W C and Lee W C. Clustering and aggregating clues of trajectories for mining trajectory patterns and routes. The VLDB Journal. 24(2), 169-192(2015).

【7】Ouyang Z and Zhou Y Q. Self-adaptive step glowworm swarm optimization algorithm. Journal of Computer Applications. 31(7), 1804-1807(2011).
欧阳喆, 周永权. 自适应步长萤火虫优化算法. 计算机应用. 31(7), 1804-1807(2011).

【8】Li Z, Shen X M and Miao T J. Image mosaic based on contract threshold adaptive SIFT algorithm. Infrared Technology. 39(10), 946-950(2017).
李尊, 申小萌, 苗同军. 对比度阈值自适应的SIFT图像拼接算法. 红外技术. 39(10), 946-950(2017).

【9】Xia Y, Liu Z and Wang J R. Research on SURF registration algorithm in image mosaic. Journal of Changchun University of Science and Technology (Natural Science Edition). 40(2), 98-101(2017).
夏岩, 刘智, 王俊然. 图像拼接中SURF配准算法的研究. 长春理工大学学报(自然科学版). 40(2), 98-101(2017).

引用该论文

Zhenxiang Yan, Hanying Wang, Qishuang Shi, Yanhong Mo, Huihua Yang. Microscopic Image Stitching Based on Regional Frog Leaping Search Algorithm and Image Contour Matching[J]. Laser & Optoelectronics Progress, 2019, 56(15): 151002

颜振翔, 王寒迎, 石齐双, 莫艳红, 杨辉华. 基于区域蛙跳搜索与轮廓匹配的显微图像拼接[J]. 激光与光电子学进展, 2019, 56(15): 151002

您的浏览器不支持PDF插件,请使用最新的(Chrome/Fire Fox等)浏览器.或者您还可以点击此处下载该论文PDF