摘要
形态滤波和顺序滤波是图像处理中常用的两种方法,用这两种方法分别对含噪图像进行预处理,之后用阈值分割的办法提取出了目标,其中阈值是根据预处理前图像的灰度均值和方差来确定的。文中所处理的图像来自晴空背景下(日出前后)采集的恒星像光斑,经过实验发现用这种方法能较好提取出目标。同时实验证明图像中当目标所占面积比背景噪声点大时,用顺序滤波和阈值分割方法,不但能很好的提取出目标,而且运算速度也较快。最后给出用这种方法对其它图像的处理结果,结果表明此方法也很有效。
Morphology filtering and order filtering are two methods used in image processing. First the image is pretreated by using the two methods, and then the target is distracted by using threshold division. The threshold is decided by the mean of the pretreated image gray and the variance. The image to be processed is star image under sky background. The result indicates that the target can be distracted from its noisy background in this way. At the same time, it also proves that while the target is larger than noise particles, using the two types of method can not only distract the target well, but also process fast. Moreover, the processing result of other image proves that the method is effecitve.
出处
《量子电子学报》
CAS
CSCD
北大核心
2008年第1期29-32,共4页
Chinese Journal of Quantum Electronics
基金
国家863计划项目资助课题
关键词
图像处理
恒星目标检测
灰度形态学
顺序滤波
闽值分割
image processing
star target detecting
morphology tilting
order filtering
gray threshold dividing
