Adaptive probability filter for removing salt and pepper noises

• DOI: 10.1049/iet-ipr.2017.0910
• Publication Date: 2018-01-04T02:40:19Z
• Authors: Jiayi Chen; Yinwei Zhan; Huiying Cao; Xingda Wu
• Keywords: Salt-and-pepper noise

To overcome the drawbacks of existing filters for salt and pepper noises, an adaptive probability filter is proposed. For image, it detects noises based on characteristic minimum maximum intensity values images, as well distribution noise. If noise‐free intensities in neighbourhood repeat with a certain probability, highest repeated frequency used to remove noise statistical significance; otherwise, median pixels Experiments show that proposed method capable detecting more accurately perform much better than distinguished terms peak‐signal‐to‐noise ratio, image enhancement factor, visual representation at all densities.