ABSTRACT
In spatial filtering, the processed pixel value for the existing pixel is dependent on both itself and neighboring pixels. Smoothing filters are mainly used to reduce noise of an image and for blurring. A smoothing linear filter is basically the mean of the neighborhood pixels of the filter mask. Order-statistics smoothing filters are basically nonlinear spatial filter. The response of this filter is constructed by ordering or ranking the pixels enclosed in the image area covered by the filter. Frequency filtering is also more suitable if there is no direct kernel that can be created in the spatial domain, in which case they may also be more effective. A low-pass filter is generally used to smooth an image. The standard forms of low-pass filters are Ideal, Butterworth, and Gaussian low-pass filters. A high-pass filter is generally used to sharpen an image and to highlight the edges and fine details associated with the image.
