Denoising an image is a significant problem in the processing of digital images. Any impulse noise damages the image and the aim of denoising is to remove noise and restore the high-quality image as much as possible. This paper aims to develop a method to discriminate between corrupted and uncorrupted pixels and develop a novel filter to denoise the image. It is also necessary to consider images with different level of noise of various applications to develop an optimal system to remove the noise for further processing. In this paper different filtering techniques such as Median-Filter (MF), Weighted -Median-Filter (WMF),Centre-Weighted Median filter (CWMF), and adaptive centre weighted median filter (ACWMF) are used for denoising, also a novel filter Asymmetric Trimmed Median Filter(ATMF) is designed which outperforms compared to the filters designed earlier. Experimentation is carried by considering the dataset size of 700 noisy images. The average PSNR value of proposed system is 28.12%(DB).
Identification of the ARX Model with Random Impulse Noise Based on Forgetting Factor Multi-error Information Entropy
