大视场空间目标光电探测起伏背景抑制算法

A background clutter suppression algorithm for space target detection in wide field-of-view opto-electronic observation

针对大视场空间目标光电探测系统中目标星像大小不一(星像直径在几个像素至几十个像素)、背景受月光、云层等影响呈现大面积连续起伏或强烈起伏等问题,提出一种基于形态学重构的背景抑制算法.该算法综合利用目标和背景的信息计算标记图像,根据目标信息计算时采用多个方向上线状结构元素,在提取高亮度背景的同时,减少了大尺寸结构元素带来的计算量;根据背景信息计算与目标大小无关,避免了形态学背景抑制算法中结构元素选取的难题.以某一大视场空间目标光电观测系统获取的不同背景起伏的实测数据进行测试,结果表明,本算法具有良好的背景自适应能力,在较好抑制背景的同时,可提高目标信噪比,背景抑制效果优于对比方法.

A background clutter suppression algorithm based on morphological reconstruction, which calculated the marker image based on information of target and background, is proposed for the suppressing of background clutter of wide field-of-view opto-electronic observation. The images of these targets varied in size （from several to dozens pixels in diameter）, and the backgrounds undulated consecutively or strongly due to moonlight, clouds, etc. The marker image according to target information was calculated by a plurality of direction of linear structuring elements, which extracted high gray background better and reduced the huge amount of calculation. The mark image according to background information was independent of the target＇s imaging size, which avoided the problem of structuring element choices in the morphological background suppression algorithm. The proposed algorithm was tested and compared through real data obtained by a wide field-of-view opto-electronic observation system. The results show that the proposed algorithm has excellent background adaptability and achieves a higher signal-to-noise ratio, and the background suppression effect is more efficient than those done by other testedbackground suppression algorithms.