摘要
研究了复杂背景下红外小目标图像的去噪问题,鉴于小波阈值法去噪的缺点,结合小波变换的去相关性和能量紧支性,提出一种新的去噪方法。考虑到实际中的复杂背景和大量干扰,弱小目标通常占有很少像素,首先对红外小目标图像进行二级小波变换,然后根据新的算法对变换所得小波细节系数进行邻域运算,最后通过小波逆变换得到处理后的图像。实验中采用Db3小波基函数,分别对两帧低信噪比原始图像进行仿真。仿真结果表明,该算法能很好地保存小目标的形状特征,抑制背景,达到较好的去噪效果。
The infrared small target de-noising in complicated background is studied. According to de-correlation and energy compaction of wavelet transforms, a new de-noising method is proposed. The method is put forward with the consideration of the practicality of the method, especially while the image involves complex background and a lot of noise. Firstly, the image is decomposed twice using wavelet transforms. Secondly, each wavelet detail coefficient is recalculated according to the new method in order to avoid the shortcomings of the hard or soft wavelet shrinkage. Lastly, the de-noised image is obtained by reconstruction from the processed coefficients. Simulation results show the de-noising method can preserve the shape character of small target, attenuate the background, and work efficiently in de-noising. The simulation used Db3 wavelet and two original images with low SNR. The size of one experimental image is 181 × 250 pixels and the other is 175 × 247 pixels.
出处
《强激光与粒子束》
EI
CAS
CSCD
北大核心
2007年第3期399-402,共4页
High Power Laser and Particle Beams
关键词
红外小目标
去噪
小波变换
细节系数
邻域运算
Infrared small target
De-noising
Wavelet transforms
Detail coefficients
Neighborhood operating