散斑噪声污染的激光水下图像滤波算法

Filtering algorithm on underwater laser image stained by speckle noise

水是一种特殊的介质 ,在激光水下图像中总存在因水的后向散射引起的散斑噪声。充分利用形态学的不同结构元多方向形态滤波的多分辨率特性 ,以及形态学运算固有的能将大量的复杂图像处理运算转换为基本的逻辑与移位运算的组合来完成的特点 ,提出了一种多方向的形态滤波算法。不仅有效地抑制了因海水后向散射引起的散斑噪声 ,同时保留了激光水下目标图像的几何结构细节特征。实验表明这是一种有效的滤除水下激光图像散斑噪声的快速滤波算法。

Water is a special medium. In the underwater laser image there are always speckle noise which is caused by backscattering of water. A multidirectional morphological filtering algorithm is proposed by use of the characteristics of morphology's multisolution analysis and fastly-computing. It can not only efficiently reduce the speckle noise but also preserve the detailed features of the image. Experimental results indicate this algorithm is a fast efficient algorithm to filter speckle noise in the underwater laser image.