摘要
微型航天目标探测器主要是对空间目标探测识别,然而在微型航天探测器在空间对目标物成像时目标物、背景和探测器之间存在相对运动,目前还没有好的运动背景中运动目标识别的算法。因此,针对微型航天探测器在星空背景下对目标物识别的要求,提出了把图像配准与双差分法相结合目标识别的算法。该算法对连续采集的三帧图像中的前两帧和后两帧分别进行配准,差分,二值化,然后将两副二值化图像逻辑乘,识别出目标。计算机仿真的结果表明该算法计算量比较小,对目标的识别效果好。因而,基于图像配准的双差分法是一个相对较好适用于微型航天探测器对空间目标探测识别的算法。
The mainly task of microspacecraft target detector is to identify the target in the space. But there are relatively movememt among target, background and detector, andThere isn＇ t a good algorithm to identify target from the moving background. In order to meet the needs of detecting the target under the moving space background by microspacecraft target detector. This paper proposed an algorithm that combine image registration with double difference method. The algorithm separately use image registration, image difference to the former two images and the later two images which were collected three seriate images , then multiply the two difference images by pixel after binarizing to detect the target. The result of simulation indicate that the calculation is small and recognized result is good. So the double difference based on image registration is relatively fit for microspacecraft target detector to identify the spacial target.
出处
《计算机仿真》
CSCD
2007年第5期34-36,86,共4页
Computer Simulation
关键词
微型航天探测器
双差分
图像配准
Microspacecraft target detector
Double deference
Image registration
