Hazique Aetesam

Hyperspectral images (HSIs) aid in the investigation of objects beyond the visible ranges. However, the acquired data from aerial sensors suffer from mixed Gaussian-impulse (GI) noise. This hinders the application of high-level vision tasks like target tracking, classification and spectral unmixing. In this paper, we use dual data fidelity terms (in the form of combined 𝓁2 - 𝓁1 norms) as an energy functional to handle this mixed noise scenario. The resultant objective function is projected as a constrained minimization problem along with the estimation of local bounds on the expected absolute value (EAV) and variance of the combined GI noise. To account for the spatial variability, an adaptive hyper-parameter is appended to the total variation (TV) regularised objective function. To the best of our knowledge, this is the first attempt to perform local parameter selection under mixed data fidelity term. Finally, the original primal-dual hybrid gradient (PHDG) algorithm is modified to act as the minimizer of the locally adaptive optimization problem (named as ProxLocHyDen). Experimental results conducted over synthetic data and those acquired from real sensors suggest the potential utility and robustness of the proposed methodology.

Images obtained from hyperspectral sensors provide visual details about the target area beyond the visible parts of the electromagnetic spectrum. However, the acquired 3D datacube is corrupted by a combination of mixed Gaussian-impulse noise. In this letter, the problem is posed as the joint statistical characterization of this mixed noise scenario which is missing in the recent literature. We propose a Dual Fidelity based (ℓ ₂ − and ℓ ₁ −norms) Automated Parameter Adjustment scheme (named as Dual-FAPA) for recovering hyperspectral images (HSIs) corrupted by the mixed noise. Total variation (TV) is used as the regularization term due to its noise removal and detail preserving capability. To estimate the bounds on the aggregate expected absolute value (EAV) and variance under joint Gaussian and impulse noise, the objective function is remodelled as a constrained minimization problem whose solution aids in the automated selection of the parameters (α ₁ , α ₂ ) used to handle different levels of noise. Further, the mathematical tractability of the solution based on iterative monotonic solution and convergence analysis suggest the robustness of the proposed methodology. On experimentation with synthetic as well as real HSI data, we observe a significant improvement over state-of-the-art methods by multiple qualitative and quantitative assessments.

Images obtained from hyperspectral sensors provide information about the target area that extends beyond the visible portions of the electromagnetic spectrum. However, due to sensor limitations and imperfections during the image acquisition and transmission phases, noise is introduced into the acquired image, which can have a negative impact on downstream analyses such as classification, target tracking, and spectral unmixing. Noise in hyperspectral images (HSI) is modelled as a combination from several sources, including Gaussian/impulse noise, stripes, and deadlines. An HSI restoration method for such a mixed noise model is proposed. First , a joint optimisation framework is proposed for recovering hyperspectral data corrupted by mixed Gaussian‐impulse noise by estimating both the clean data as well as the sparse/impulse noise levels. Second , a hyper‐Laplacian prior is used along both the spatial and spectral dimensions to express sparsity in clean image gradients. Third , to model the sparse nature of impulse noise, an ℓ ₁ − norm over the impulse noise gradient is used. Because the proposed methodology employs two distinct priors, the authors refer to it as the hyperspectral dual prior (HySpDualP) denoiser. To the best of authors' knowledge, this joint optimisation framework is the first attempt in this direction. To handle the non‐smooth and non‐convex nature of the general ℓp − norm‐based regularisation term, a generalised shrinkage/thresholding (GST) solver is employed. Finally , an efficient split‐Bregman approach is used to solve the resulting optimisation problem. Experimental results on synthetic data and real HSI datacube obtained from hyperspectral sensors demonstrate that the authors’ proposed model outperforms state‐of‐the‐art methods, both visually and in terms of various image quality assessment metrics.

In this paper, we propose a variational approach for estimating the clean hyperspectral images (HSI) that are corrupted by the combined effect of Gaussian noise, impulse noise, stripes, and deadlines. Successful removal of noise from the corrupted observations is essential for subsequent downstream analyses like classification, spectral unmixing, and target tracking. The main contribution of this work is as follows: Firstly, an objective function is designed for the joint estimation of clean data and impulse corrupted pixels. A rationale is presented for using $\ell _{0}-$ norm to estimate the exact sparsity induced by impulse noise. Secondly, the problem is reformulated as a multi-convex problem which is solved using proximal projection and alternating minimization. Thirdly, to exploit the spatial-spectral similarity, a non-local and vectorized version of total variation (TV) regularization is proposed to estimate the clean data. Lastly , a study on the parameter sensitivity analysis empirically validates the convergence of the restoration results under different values of the regularization hyperparameters. The experiments conducted over synthetically corrupted and real HSI data obtained from hyperspectral sensors suggest the potential utility of the proposed methodology ( CompoHyDen ) at a scalable level.

Image denoising is an important preprocessing step in low-level vision problems involving biomedical images. Noise removal techniques can greatly benefit raw corrupted magnetic resonance images (MRI). It has been discovered that the MR data is corrupted by a mixture of Gaussian-impulse noise caused by detector flaws and transmission errors. This paper proposes a deep generative model (GenMRIDenoiser) for dealing with this mixed noise scenario. This work makes four contributions. To begin, Wasserstein generative adversarial network (WGAN) is used in model training to mitigate the problem of vanishing gradient, mode collapse, and convergence issues encountered while training a vanilla GAN. Second, a perceptually motivated loss function is used to guide the training process in order to preserve the low-level details in the form of high-frequency components in the image. Third, batch renormalization is used between the convolutional and activation layers to prevent performance degradation under the assumption of non-independent and identically distributed (non-iid) data. Fourth, global feature attention module (GFAM) is appended at the beginning and end of the parallel ensemble blocks to capture the long-range dependencies that are often lost due to the small receptive field of convolutional filters. The experimental results over synthetic data and MRI stack obtained from real MR scanners indicate the potential utility of the proposed technique across a wide range of degradation scenarios.