基于动态阈值和全局信息的SIFT量子图像拼接

量子电子学报 ›› 2020, Vol. 37 ›› Issue (1): 22-28.

基于动态阈值和全局信息的SIFT量子图像拼接

唐泽恬1,2，丁召1,2，曾瑞敏1,2，钟岷哲1,2，朱登玮1,2，王昱皓1,2，王阳1,2，杨晨1,2,3

1贵州大学大数据与信息工程学院,贵州贵阳 550025； 2贵州大学半导体功率器件可靠性教育部工程研究中心,贵州贵阳 550025； 3贵阳朗玛信息技术股份有限公司，贵州贵阳 550022

收稿日期:2019-07-16 修回日期:2019-11-05 出版日期:2020-01-28 发布日期:2020-01-28
通讯作者: 杨晨（1982-），贵州贵阳人，博士，副教授，主要从事光电系统设计和机器视觉方面的研究。 E-mail:eliot.c.yang@163.com
作者简介:唐泽恬（1995-），贵州六盘水人，研究生，主要从事图像处理方面的研究。E-mail：tang_zetian@foxmail.com
基金资助:
Supported by National Natural Science Foundation of China(国家自然科学基金, 61604046),Guizhou Science and Technology Plan Project（贵州省科技计划项目，黔科合平台人才[2017]5788号、[2018]5781号),Semiconductor Power Device Reliability Ministry of Education Engineering Research Center Open Fund（半导体功率器件可靠性教育部工程研究中心开放基金，黔科合平台人才20176103号)。

Quantum image stitching of SIFT based on dynamic threshold and global information

TANG Zetian1,2, DING Zhao1,2, ZENG Ruimin1,2, ZHONG Minzhe1,2, ZHU Dengwei1,2, WANG Yuhao1,2, WANG Yang1,2, YANG Chen1,2,3

1 College of Big Data and Information Engineering, Guizhou University, Guiyang 550025, China; 2 Power Semiconductor Device Reliability Engineering Center of the Ministry of Education, Guizhou University, Guiyang 550025, China; 3 Longmaster Information and Technology Co., Guiyang 550022, China

Received:2019-07-16 Revised:2019-11-05 Published:2020-01-28 Online:2020-01-28

摘要/Abstract

摘要： 针对量子图像拼接使用传统SIFT算法易造成特征点数量不合理和误匹配率高的问题，提出基于动态阈值和全局信息的SIFT量子图像拼接算法。在特征点数量不合理方面，利用特征点数量与量子点或量子环数量呈现正比例关系这一特性，通过量子点或环的密度对SIFT算法的对比度阈值进行设置，以得到合适的特征点数量。在误匹配率高方面，通过构建全局信息描述子，并与SIFT局部描述子相结合，以降低误匹配率。实验表明：改进的算法有效地完成了量子环、量子线和量子点图像的拼接；将量子图像的特征点有效的控制在一个合理的范围内，并将误匹配率从17.34%~33.02%降低至10.84%~20%，使量子图像的拼接具有更好的可靠性。

关键词:

尺度不变特征变换, 量子图像, 全局信息, 动态阈值

Abstract: When applied to quantum image mosaic, the traditional SIFT algorithm will result in problems such as unreasonable feature point distribution and high mismatch rate. An improved SIFT quantum image stitching algorithm based on dynamic threshold and global information is proposed. In the aspect of unreasonable feature point number, as the number is proportional to the number of quantum dots or quantum loops, the density of quantum dots or rings is calculated to set the contrast threshold of the SIFT algorithm for the appropriate feature point number. In the aspect of reducing high false mismatch rate, a global information descriptor is constructed and combined with the local SIFT descriptor. Experiments show that the improved algorithm effectively completes the stitching of quantum loops, quantum lines and quantum dot images, effectively controls the feature points of quantum images within a reasonable range, and reduces the mismatch rate from 17.34% ~ 33.02% to 10.84% ~ 20%, which makes the quantum image mosaic have better reliability.

Key words:

scale-invariant feature transform, quantum image, global information, dynamic threshold

中图分类号:

TP391.41

唐泽恬丁召曾瑞敏钟岷哲朱登玮王昱皓王阳杨晨. 基于动态阈值和全局信息的SIFT量子图像拼接[J]. 量子电子学报, 2020, 37(1): 22-28.

TANG Zetian, DING Zhao, ZENG Ruimin, ZHONG Minzhe, ZHU Dengwei, WANG Yuhao, WANG Yang, YANG Chen, . Quantum image stitching of SIFT based on dynamic threshold and global information[J]. Chinese Journal of Quantum Electronics, 2020, 37(1): 22-28.

参考文献

[1] Zhou Xun, Luo Zijiang, Guo Xiang, et al. Surface segregation of GaAs films by the evolution of reflection high-energy electron diffraction patterns [J]. Chinese Physics B, 2012, 21(4): 046103. [2] Zhou Xun, Yang Zairong, Luo Zijiang, et al. Temperature calibration and surface phase transition of molecular beam epitaxy growing GaAs crystal substrate under real-time monitoring by reflection high-energy electron diffraction [J]. Acta Physica Sinica (物理学报), 2011, 60(1): 481-485 (in Chinese). [3] Ghosh D, Kaabouch N. A survey on image mosaicing techniques [J]. Journal of Visual Communication and Image Representation, 2016, 34(C): 1-11. [4] Zhang Kai, Li Xuzhi, Zhang Jiuxing. A robust point-matching algorithm for remote sensing image registration [J]. IEEE Geoscience and Remote Sensing Letters, 2014, 11(2): 469-473. [5] Wu Yue, Ma Wenping, Gong Maoguo, et al. A novel point-matching algorithm based on fast sample consensus for image registration [J]. IEEE Geoscience and Remote Sensing Letters, 2015, 12(1): 43-47. [6] Chen Yue, Zhao Yan, Wang Shigang. Fast image stitching method based on SIFT with adaptive local image feature[J]. Chinese Optics（中国光学）, 2016, 9(4): 415-422 (in Chinese). [7] Lowe D G. Distinctive image features from scale-invariant keypoints [J]. International Journal of Computer Vision, 2004, 60(2): 91-110. [8] Li Yufeng, Li Guangze, Gu Shaohu, et al. Image mosaic algorithm based on area blocking and SIFT[J]. Optics and Precision Engineering (光学精密工程), 2016, 24(5):1197-1205 (in Chinese). [9] He Bin, Tao Dan, Peng Bo, High real-time F-SIFT image mosaic algorithm [J]. Infrared and Laser Engineering (红外与激光工程), 2013,42(S2): 440-444. (in Chinese). [10] Ji Hua, Wu Yuanhao, Sun Honghai, et al. SIFT feature matching algorithm with global information [J]. Optics and Precision Engineering (光学精密工程), 2009,17(2): 439-444 (in Chinese). [11] Wang Rui, Zhu Zhengdan, SIFT matching with color invariant characteristics and global context [J]. Optics and Precision Engineering (光学精密工程), 2015, 23(1): 295-301 (in Chinese). [12] Tang Zetian, Yang Chen, Tang Jiawei, et al. Research on quantum dot STM morphological image recognition based on machine vision[J]. Journal of Atomic and Molecular Physics (原子与分子物理学报), 2019, 36(5): 824-830 (in Chinese).