Enhanced Performance in Salt and Pepper Noise Denoising Method using Discrete Wavelet Transform

Many denoising algorithms are designed to recover a noise infected image. However, most of them cannot well recover a significant noise dirtied image with noise density higher than seventy percentages. A brand new approach projected to remove background noise with efficiency by detecting and modifying noisy pixels in a picture. If the middle pixel of a local window is classed to clamorous, this center pixel is replaced by a weighted norm on associate degree optimum direction, facultative impulse noise to be removed. Conversely, the middle pixel is unbroken unchanged once it's classified to noise-free, yielding the standard of repaired image being well maintained. The algorithmic rule of salt and pepper noise, DWT (Discrete Wavelet Transform) Denoising that exhibits excellent performance each in denoising and in restoration, may be simply and effectively parallelized to use the complete power of multi-core processor performance. The projected Discrete Wavelet Denoising implementation supported the quick Flow library achieves each close-to-ideal quickening and incredibly excellent wall-clock execution figures.