基于知识的红外图像背景抑制

在背景抑制时，先对可能的地物或地面的像素点进行标识，然后用像素灰度减本行（相邻行）灰度均值的方法却除大云图，大片地物和消除大气温度分布不均的影响，用模式识别方法识别图像下部存在的大地物或地面，将地面或与地面连通的地物去除，提出的基于知识的背景抑制方法具有运算量小，便于实时处理并能去除大部分的背景等特点，实验表明，该方法能有效地抑制复杂红外图像背景。

Knowledge Based Infrared Image Background Suppression

When suppressing the background, pixels of presumable surface or ground features are first marked. Extensive bulks of cloud clusters, surface features and differences in the distribution of atmospheric temperature are then removed by subtracting the mean greyness of the row (neighboring row). The ground and surface features are suppressed when recognized, through pattern recognition. The paper presents a knowlede based algorithm for background suppression. It uses lesser amount of operands and is capable of suppressing the most part of the background. Experimental results show that it can be used to suppress the background very effectively.