Abstract
Abstract
Morphological filter is a kind of nonlinear filter. Based on the erosion and dilation operations a complex morphological filter was proposed with multi-structure elements. In order to protect the image detail effectively, a noise detect method was introduced firstly. The noise pixels are removed using the proposed filter, while signal pixels hold their pixel value and are left unprocessed. The experimental results demonstrate that the proposed algorithm has good filtering performance, which can preserve edge details effectively during denoising. For both gray image and binary image with salt & pepper noise pollution, the proposed algorithm is superior to the conventional morphological methods in the aspect of objective performance evaluation and subjective visual effect. Meanwhile, the proposed method can also be extended to the color image processing.