摘要
在分析了几种非均匀校正算法特点的基础上,提出了一种易于硬件实现的新算法,并给出了具体的硬件实现方法。与许多基于场景的非均匀校正算法不同,它不依赖目标的运动来抑制固定的图像噪声,而是将图像中的固定图像噪声和随时间温度的缓慢变化带来的信号漂移,看作是一种低频信号,利用时域高通滤波,将低频滤除;由于滤除的低频信号远低于帧频,因此不会影响图像细节。应用结果表明,该算法能有效消除红外探测器的非均匀噪声并有效抑制温度变化带来的漂移。
On the basis of analysis for several kinds of nonuniformity correction algorithms, a new scene-based nonunifortuity correction algorithm easy to implement in hardware is put forward. And the implementation method in hardware is presented in detail. Different from other scene-based algorithms, it does not suppress the noise by the movement of the object, but considers the fixed-pattern noise and the signal drift followed by temperature as low frequency signal, and it can be filtered by the temperal high-pass filter correction （THPFC）. The image details are not influenced, because the frequency of the low frequency signal filtered is more less than the frame frequency. The results of application prove that this algorithm can eliminate the nonuniformity noise and suppress the signal drift varies with temperature effectively.
出处
《激光与红外》
CAS
CSCD
北大核心
2006年第7期568-570,共3页
Laser & Infrared
