A Hybrid Method for the Removal of RVIN Using Self Organizing Migration with Adaptive Dual Threshold Median Filter

Day-by-day impulse detection seems critical because removing the random-valued impulse noise with high density from the noisy image is a typical task. The specified noise does not regularly obstruct the pixels of the image. The proposed work, which combines self-organizing migration & adaptive dual-threshold techniques, gives optimal output for the noisy image in less time. The proposed methodology has been applied on various noisy images with different window sizes and it has proven better results while reconstructing the original image with high quality from the noisy image. To evaluate the performance of the proposed work MSE and PSNR has considered as evaluation parameters. PSNR value is evaluated by varying the noise from 10 to 70%. As per the comparative analysis, it is proved that proposed method gives better results and able to preserve the object information in the given noisy image.