期刊名称：International Journal of Image Processing (IJIP)
电子版ISSN：1985-2304
出版年度：2013
卷号：7
期号：1
页码：101-108
出版社：Computer Science Journals
摘要：The image pre-processing is very critical and important task in any digital image analysis system. The eventual success and failure of image analysis depends on the performance of preprocessing techniques applied on the image to be analyzed. In digital images, different noise types are noticed and to attenuate each type of noise, different pre-processing methods have been proposed in literature. The main focus of this paper is on pre-processing the microstructure images. Among many types of noise, impulse noise is the one which is generally noticed in microstructure images. This paper is to present a novel, efficient and suitable pre-processing method for negotiating the impulse noise that is generally present in microstructure images. Through this paper, a new filtering method, selective median switching filter (SMSF) has been proposed. The proposed method is compared with filtering methods those belong to median filter family for their efficiency in negotiating with impulse noise. The efficiency of proposed method is compared with other methods by computing the three image quality assessment methods, namely, mean square error (MSE), peak signal-to-noise ratio (PSNR) and correlation coefficient. The experimental results confirm that the proposed SMSF method is efficient in handling the impulse noise present in microstructure images of material. Also, the proposed SMSF method is quite efficient in preserving the edge information in images.
关键词：Pre-processing; SMSF; Median Filter; MSE; PSNR; Correlation Coefficient; Microstructure.
