Detection and Removal of Impulse Noise from Colour Image using Lagrange Interpolation and Centre Weighted Vector Median Filter

Sally Khaidem, Keisham Pritamdas, Loitongbam Surajkumar Singh and Khumanthem Manglem Singh

In this paper, Impulse noise is detected using Lagrange interpolation method. Lagrange Interpolation method helps in predicting the pixel values in a given dataset. The predicted pixel value is compared with the original pixel value. When the comparison exceeds a specified threshold value, the concerned pixel is said to be noisy else it is a good pixel. The detected noisy pixel is removed with the help of Centre Weighted Vector Median Filter (CWVMF). The simulation of the algorithm and calculation of performance parameters are done in Matlab_2019b simulation tools. The performance parameters considered are Peak Signal to Noise Ratio (PSNR), Mean Absolute Error (MAE), Mean Square Error (MSE) and Structure Similarity Index (SSIM).

Volume 12 | Issue 3

Pages: 50-56

DOI: 10.5373/JARDCS/V12I3/20201166