A Review: New Histogram for Bilateral Filtering and Nearest Neighbor Searching

Abstract: Noise not only degrades quality of image but also results in loss of important information in images. There are different types of noise are available such as Gaussian, impulse, mixed noise etc. Filtering plays vital role in image processing and computer vision to remove noise. In this work, we have reviewed and analyzed different nearest neighbor field (NNF) methods using bilateral filtering to preserve edges. Computing NNF is nothing but, for each patch in one image, find out the most similar patch in other image. Bilateral filtering overcomes limitation of using box spatial filter kernel by using locality sensitive histogram (LSH). The computational complexity of bilateral filter is linear in number of pixels. Also new bilateral weighted histogram (BWH) is proposed for edge preserving patch-match. In this paper, we have studied and reviewed different patch-matching methods.

Keywords: locality sensitive histogram, bilateral weighted histogram, bilateral filter, nearest neighbor field.