像素级图像融合技术在视频判读中的应用

The Pixel Level Image Fusion Technology Applied by Video Judgement

在视频判读中,往往因目标丢失而不能准确地提取目标的脱靶量。而随着多传感器系统的广泛应用,采集到的数字图像具有多样性。利用多传感器图像的多样性,把图像融合技术应用于视频判读中,使在具有多传感器的观测设备上,图像融合算法可以针对来自不同传感器的多源图像信息,通过对多幅图像信息的提取与综合,从而获得对同一场景或目标的更为准确、全面、可靠的图像描述,以便更为精确地定位目标,以及在因环境因素某一传感器目标丢失时,不影响视频判读。

In video-judging, we can＇t get the undershoot exactly because the target lost frequently. By the use of multi-sensor, we can get various images. In this paper, The variety of multi-sensor images has been used by video-judging. So in the observing equipment with multi-sensor, image fusion method aims at multi-source image information of different sen- sors, used by the collection and integration of the image information, to get more accurate, comprehensive, reliable description, so as to local the target exactly and judge normally when the target of one sensor lost.