Modified adaptive image filtering and enhancement method for speckle pattern based on visual characteristics

Abstract

A modified adaptive image filtering and enhancement method for speckle pattern based on human visual characteristics is described in this paper. Because there are various noises in speckle pattern, especially signal-dependent noise such as speckle noise, it is very difficult to directly and accurately extract useful information from it. In the light of this case, a local variable-direction non-linear adaptive image filtering method that can simultaneously remove various noises in speckle pattern is presented here, in which, the characteristics of linear local minimum mean square error filter and some non-linear filters are combined, and the local fringe direction and neighborhood statistics are considered. Experimental results show that the proposed filtering and enhancement method are effective for speckle pattern and it is easy to implement, the loss of useful information is very small as well.