泽尼克多项式
计算机科学
斑点图案
人工智能
计算机视觉
散斑噪声
特征(语言学)
噪音(视频)
相位一致性
模式识别(心理学)
相(物质)
合成孔径雷达
图像(数学)
光学
语言学
哲学
物理
波前
化学
有机化学
作者
Bo Wang,Qinghong Sheng,Rui Ren,Li Jun,Weilan Xu,Xiaoting Yu
标识
DOI:10.1080/2150704x.2023.2215941
摘要
ABSTRACTAs SAR adopts an oblique distance imaging mode, due to obvious speckle multiplicative noise, the registration accuracy of SAR and optical images is low. This study proposes using Zernike polynomials to fit the front of the complex SAR echo data to obtain its phase feature, simultaneously extracting the SAR Zernike matrix feature from the complex echo phase and optical image, and using it to construct the registration descriptor to realize the registration of SAR and optical images. Experimental results show that the proposed method avoids the problem of obvious coherent speckle noise, the target characteristics are more prominent, the registration accuracy is higher, and the ability to recognize SAR targets and scenes is improved.KEYWORDS: Multisource image registrationregistration descriptorphase featurespeckle noiseZernike matrix Disclosure statementNo potential conflict of interest was reported by the authors.Additional informationFundingThis work was supported in part by the National Key Laboratory Foundation(Contract No. 6142411203416). We would also like to convey our heartfelt gratitude to the reviewers and members of the editorial team for their comments and contributions.
科研通智能强力驱动
Strongly Powered by AbleSci AI