Open AccessScene motion detection in imagery with anisoplanatic optical turbulence using a tilt-variance-based Gaussian mixture model
Author(s) -
Richard Van Hook,
Russell C. Hardie
Publication year - 2021
Publication title -
applied optics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.668
H-Index - 197
eISSN - 2155-3165
pISSN - 1559-128X
DOI - 10.1364/ao.424181
Subject(s) - computer science , artificial intelligence , computer vision , image warping , tilt (camera) , turbulence , pixel , benchmark (surveying) , optics , gaussian , physics , mathematics , geology , geometry , geodesy , quantum mechanics , thermodynamics
In long-range imaging applications, anisoplanatic atmospheric optical turbulence imparts spatially- and temporally varying blur and geometric distortions in acquired imagery. The ability to distinguish true scene motion from turbulence warping is important for many image-processing and analysis tasks. The authors present a scene-motion detection algorithm specifically designed to operate in the presence of anisoplanatic optical turbulence. The method models intensity fluctuations in each pixel with a Gaussian mixture model (GMM). The GMM uses knowledge of the turbulence tilt-variance statistics. We provide both quantitative and qualitative performance analyses and compare the proposed method to several state-of-the art algorithms. The image data are generated with an anisoplanatic numerical wave-propagation simulator that allows us to have motion truth. The subject technique outperforms the benchmark methods in our study.