CO2激光相干成像系统具有分辨率高、抗干扰能力强、能三维成像等特点,但其所成图像易受到散斑噪声的污染造成分辨率下降,因此研究适用于此成像系统图像散斑噪声抑制的算法具有重要的实用价值.文中对常用的噪声抑制算法:中值滤波、同态均值滤波和Lee滤波进行了介绍,并将其应用于CO2脉冲激光相干成像系统图像的散斑噪声抑制处理,比较了三种算法的处理结果.结果表明,采用小窗口、不迭代Lee滤波进行处理时,既明显抑制了散斑噪声,又有效保持了图像的边缘信息,所以用它进行激光成像系统图像散斑噪声抑制处理是合适的.

The coherent CO₂ laser imaging system has high resolution, strong anti-interference and 3D imaging, but its imagery is usually affected by speckle noise, imaging resolution descend. It's important to study on suppressing speckle algorithms for this system. In the paper, three common algorithms——median filtering, homomorphic mean filtering, and Lee filtering were introduced and used in suppressing speckle in coherent pulse CO₂ laser imaging system. Then these algorithms were compared. The results show that non-iterative Lee filtering with small windows is better. The speckle noises are suppressed clearly, and image edges are maintained.