Local Spectral Component Decomposition for Multi-Channel Image Denoising

Abstract

We propose a method for local spectral component decomposition based on the line feature of local distribution. Our aim is to reduce noise on multi-channel images by exploiting the linear correlation in the spectral domain of a local region. We first calculate a linear feature over the spectral components of an M -channel image, which we call the spectral line, and then, using the line, we decompose the image into three components: a single M -channel image and two gray-scale images. By virtue of the decomposition, the noise is concentrated on the two images, and thus our algorithm needs to denoise only the two gray-scale images, regardless of the number of the channels. As a result, image deterioration due to the imbalance of the spectral component correlation can be avoided. The experiment shows that our method improves image quality with less deterioration while preserving vivid contrast. Our method is especially effective for hyperspectral images. The experimental results demonstrate that our proposed method can compete with the other state-of-the-art denoising methods.

Hyperspectral image restoration based on spatio-spectral structure tensor regularization

“…As for BM4D, we use the MATLAB code published by the authors. Test images were taken from the images in [8], [25], [26]. In order to objectively evaluate denoising performance, two indexes: peak signal to noise ratio (PSNR) and SSIM [27] are used.…”

“…The set B y,ϵ is defined as B y,ϵ = {x ∈ R M N | ∥x − y∥ 2 ≤ ϵ} using the prescribed error tolerance ϵ, and ι By,ϵ (x) is defined in the same way as (8).…”

“…To extract signals buried in the noise, many denoising methods for the hyperspectral images have been proposed [3]- [8]. BM4D [3] is a highly accurate nonlocal method for images with three or more channels, which is an extended version of the state-of-the-art 2D image denoising method [9].…”

Joint Local Abundance Sparse Unmixing for Hyperspectral Images

“…Hyperspectral data Y ∈ R l×n have linearity in their spectral [38] and spatial [30] domains. Qu et al [30] provided prior knowledge that the high spatial correlation of the hyperspectral data, implies linearly dependent abundance vectors in the abundance matrix X ∈ R m×n .…”

Remote Sensing Satellite Image Processing Techniques for Image Classification: A Comprehensive Survey

“…Yang et al, [40] proposed Moving Weighted Harmonic Analysis (MWHA) method to reconstruct high quality vegetation NDVI time series data (removing residual effects and noise levels) of SPOT 5. MWHA method provides moving support domain to assign the weights for all the points which makes easier in determining the frequency number.…”