期刊名称：ELCVIA: electronic letters on computer vision and image analysis
印刷版ISSN：1577-5097
出版年度：2008
卷号：7
期号：3
语种：English
出版社：Centre de Visió per Computador
摘要：When observing a scene horizontally at a long distance in the near-infrared domain, degradations dueto atmospheric turbulence often occur. In our previous work, we presented two hybrid methods to restorevideos degraded by such local perturbations. These restoration algorithms take advantages of a space-timeWiener filter and a space-time regularization by the Laplacian operator. Wiener and Laplacian regularizationresults are mixed differently depending on the distance between the current pixel and the nearest edge point.It was shown that a gradation betweenWiener and Laplacian areas improves results quality, so that only thealgorithm using a gradation will be used in this article.In spite of a significant improvement in the obtained images quality, our restoration results greatly dependon the segmentation image used in the video processing. We then propose a method to select automaticallythe best segmentation image.
关键词：Video Surveillance; Image and Video Processing;Image Analysis and Processing; Applications
