S. Shang, Zhan-He Ou, Yang Zhou, Yuhao Yang, Pin Li
{"title":"A novel method of heterologous image registration based on SURF and feature inertial following","authors":"S. Shang, Zhan-He Ou, Yang Zhou, Yuhao Yang, Pin Li","doi":"10.1117/12.2661631","DOIUrl":null,"url":null,"abstract":"In this paper, a novel method of heterologous image registration method based on feature inertial following is proposed, which can perform high-precision and rapid registration of SAR image and optical image. The first image pair in the sequence is registered based on improved SURF feature to achieve high-precision registration. Based on this registration result, the other image pairs in the sequence are registered by the method of feature inertia following to achieve rapid registration. The proposed method makes full use of the correlation and gradient properties between image sequence frames. It maintains the registration accuracy of the preceding images, improves the registration speed greatly.","PeriodicalId":16181,"journal":{"name":"Journal of Infrared, Millimeter, and Terahertz Waves","volume":null,"pages":null},"PeriodicalIF":1.8000,"publicationDate":"2023-04-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of Infrared, Millimeter, and Terahertz Waves","FirstCategoryId":"5","ListUrlMain":"https://doi.org/10.1117/12.2661631","RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"ENGINEERING, ELECTRICAL & ELECTRONIC","Score":null,"Total":0}
引用次数: 0
Abstract
In this paper, a novel method of heterologous image registration method based on feature inertial following is proposed, which can perform high-precision and rapid registration of SAR image and optical image. The first image pair in the sequence is registered based on improved SURF feature to achieve high-precision registration. Based on this registration result, the other image pairs in the sequence are registered by the method of feature inertia following to achieve rapid registration. The proposed method makes full use of the correlation and gradient properties between image sequence frames. It maintains the registration accuracy of the preceding images, improves the registration speed greatly.
期刊介绍:
The Journal of Infrared, Millimeter, and Terahertz Waves offers a peer-reviewed platform for the rapid dissemination of original, high-quality research in the frequency window from 30 GHz to 30 THz. The topics covered include: sources, detectors, and other devices; systems, spectroscopy, sensing, interaction between electromagnetic waves and matter, applications, metrology, and communications.
Purely numerical work, especially with commercial software packages, will be published only in very exceptional cases. The same applies to manuscripts describing only algorithms (e.g. pattern recognition algorithms).
Manuscripts submitted to the Journal should discuss a significant advancement to the field of infrared, millimeter, and terahertz waves.