基于Horn-Schunck光流运动矢量法的印章图像匹配
Seal Image Matching Based on Horn-Schunck Optical Flow Motion Vector Method
摘要: HS光流法是基于机器视觉的运动目标方法中的主流方法之一，1981年由Horn和Schunck等人提出光流约束方程，对光流计算奠定了深厚的基础。本文通过引入HS光流计算方法，生成待识别印章与存档印章的光流速度矢量场，再计算矢量误差的局域方差从而计算综合判定待识别印章的真伪。实验结果证明，HS光流法一定程度上可以鉴别真伪印章。
Abstract: The HS optical flow method is one of the mainstream methods of moving target methods based on machine vision. In 1981, the optical flow constraint equation was proposed by Horn and Schunck et al., which laid a deep foundation for optical flow calculation. In this paper, by introducing the HS optical flow calculation method, the optical flow velocity vector field of the seal to be identified and the archived seal is generated, and the local variance of the vector error is calculated to com-prehensively determine the authenticity of the seal to be identified. The experimental results prove that the HS optical flow method can discriminate authentic seals to a certain extent.
文章引用：张娜, 付小宁, 苏煜琴. 基于Horn-Schunck光流运动矢量法的印章图像匹配[J]. 计算机科学与应用, 2020, 10(3): 521-528. https://doi.org/10.12677/CSA.2020.103054

参考文献

[1] Horn, B.K.P. and Schunck, B.G. (1981) Determining Optical Flow. [Google Scholar] [CrossRef] 
[2] 朱璐瑶. 改进的Horn-Schunck光流法在目标追踪中的应用[J]. 中国体视学与图像分析, 2015, 20(3): 218-226.
[3] 李秀丽, 李旭健, 高林. 一种改进的视频运动目标检测算法[J]. 软件导刊, 2016, 15(8): 26-29.
[4] Okamura, H. and Fujii, Y. (1988) Lorentzian Force Type Fibre-Optic AC Magnetic Sensor Realising Frequency-Independent Sensitivity. Electronics Letters, 24, 1090-1091. [Google Scholar] [CrossRef] 
[5] Wang, F.F., Han, J.L. and Guo, L.J. (2020) Optical Coherence Tomography and Fractional Flow Reserve Guided Treatment of Woven Coronary Artery Anomaly Presenting as Acute Myocardial Infarction: A Case Report. Medicine, 99, e19163. [Google Scholar] [CrossRef] 
[6] Meng, X., Gao, X.Y. and Du, Y.L. (2020) Influence of O2/Ar Flow Ratio on the Structure, Optical and Electrical Properties of DC Reactively Magnetron-Sputtered Face-Centered Cubic NiO Films at Room Temperature. Physica B: Physics of Condensed Matter, 579, Article ID: 411877. [Google Scholar] [CrossRef] 
[7] Ramaiah, J., Ajithaprasad, S., Gannavarpu, R. and Ambrosini, D. (2020) Fast and Robust Method for Flow Analysis Using GPU Assisted Diffractive Optical Element Based Background Oriented Schlieren (BOS). Optics and Lasers in Engineering, 126, Article ID: 105908. [Google Scholar] [CrossRef] 
[8] Yoshitaka, F. and Takeharu, S. (2020) The Effect of Optical Flow Motion Direction on Vection Strength. i-Perception, 11. [Google Scholar] [CrossRef] [PubMed]
[9] Yoshinori, K., Tomoyo, S., Masahiro, H., Eisuke, U., Rikuta, H., Yoshihisa, K., Tadashi, M., Tetsumin, L., Taishi, Y. and Tsunekazu, K. (2020) Optical Coherence Tomogra-phy-Defined Plaque Vulnerability in Relation to Functional Stenosis Severity Stratified by Fractional Flow Reserve and Quantitative Flow Ratio. Catheterization and Cardiovascular Interventions: Official Journal of the Society for Cardiac Angiography & Interventions. [Google Scholar] [CrossRef] [PubMed]
[10] Priya, D., Pascal, A., Charuta, F., Berenice, F.I., Stephan, J., Khokha, M.K., Kahle, K.T. and Engin, D. (2020) Author Correction: Visualizing Flow in an Intact CSF Network Using Optical Coherence Tomography: Implications for Human Congenital Hydrocephalus. Scientific Reports, 10, Article No. 2791. [Google Scholar] [CrossRef] [PubMed]