Moran's / for impulse noise detection and removal in color images

摘要

A method for impulse noise detection and removal in color images based on Moran's I statistic using Moran's I vector median filter (MIVMF) which comprises detection and removal components. The detection module determines if a pixel is noise or is noise-free. If it is a noise pixel, the vector median filter (VMF) will be used to remove the noise. This detection capability meets the so-called “switching” mechanism, which only selects noisy pixels for suppression. The inventive filter expedites processing time with the reduced number of vector calculations due to this new detection function, achieved with Moran's I index and the indication of 1-D Laplacian kernels. Compared to other vector-type median filters, the inventive MIVMF was faster in the filtering process and efficient in removing random impulse noise with different noise levels in color images. The MIVMF demonstrated a promising de-noising result based on the criteria of Peak Signal-to-Noise Ratio (PSNR) and Structural Similarity Index Metric (SSIM). With the visualization of processed images, the MIVMF avoids image blurring, preserves the image edge details, and achieves superior noise reduction.