Faculty Publications
Permanent URI for this communityhttps://idr.nitk.ac.in/handle/123456789/18736
Publications by NITK Faculty
Browse
5 results
Search Results
Item Curvature driven diffusion coupled with shock for image enhancement/reconstruction(Inderscience Publishers, 2011) Padikkal, P.; George, S.Curvature driven diffusion is widely used for image denoising and inpainting. Among the curvature driven diffusion techniques Gauss Curvature Driven Diffusion (GCDD) became a prominent image denoising method due to its capability to retain some important structures with non zero curvatures, like curved edges, corners etc. Unlike many other non-linear diffusion techniques, the curvature driven diffusion hardly has any inverse diffusion characteristics. In this work we propose to introduce a shock term along with the GCDD term to enhance the edges while smoothing-out the noise. This technique will preserve some important structures and enhance them while denoising the image. The experiments clearly demonstrates the efficiency of the method. Copyright © 2011 Inderscience Enterprises Ltd.Item Shock coupled fourth-order diffusion for image enhancement(Elsevier Ltd, 2012) Padikkal, P.; George, S.In this paper a shock coupled fourth-order diffusion filter is proposed for image enhancement. This filter converges at a faster rate while preserving and enhancing edges, ramps and textures present in the images. The proposed filter diffuses with varying magnitudes in the directions normal to the level-curve and along it. The magnitude of the directional diffusion is controlled by a diffusion function, meant to provide a good response in the direction along the level-curves, than across them. The proposed filter can still preserve the planar approximation of the image, thereby avoiding the discrepancy caused due to the staircase effect, as in the second-order counterparts. The anisotropic property of the filter is thoroughly studied, analyzed and demonstrated with perspective and quantitative results. The performance of the proposed filter is compared with the state-of-the-art methods for image enhancement. The quantitative and perspective measures provided endorse the capability of the method to enhance various kinds of images. © 2012 Elsevier Ltd. All rights reserved.Item A time-dependent switching anisotropic diffusion model for denoising and deblurring images(2012) Padikkal, P.; George, S.A conditionally anisotropic diffusion based deblurring and denoising filter is introduced in this paper. This is a time-dependent curvature based model and the steady state can be attained at a faster rate, using the explicit time-marching scheme. The filter switches between isotropic and anisotropic diffusion depending on the local image features. The switching of the filter is controlled by a binary function, which returns either zero or one, based on the underlying local image gradient features. The parameters in the proposed filter can be fine-tuned to get the desired output image. The filter is applied to various kinds of input test images and the response is analyzed. The filter is found to be effective in the reconstruction of partially textured, textured, constant-intensity and color images, as is evident from the results provided. © 2011 Copyright Taylor and Francis Group, LLC.Item A self-attention driven retinex-based deep image prior model for satellite image restoration(Elsevier Ltd, 2024) Shastry, A.; Padikkal, J.; George, S.; Bini, A.A.A self attention driven Deep Image Prior (DIP) framework has been proposed in this work for restoring satellite images corrupted by speckled interference and contrast deficiency. The retinex-based framework incorporated here-in leverages the benefits of DIP approach for image restoration, thus requiring only a single input image, eliminating the need for ground truth or training data. An attention framework is incorporated into the architecture of DIP networks to effectively capture fine textures, enhancing the restoration capability of the model. Two generative networks are employed to obtain the luminance and reflectance maps, with the model's loss functions specifically designed to tackle speckle interference and contrast distortions present in the input. These generated maps eventually reconstruct the enhanced version of the image. Satellite images from different sensors are used to demonstrate and compare the performance of the model. Various state-of-the-art models are evaluated and compared with the proposed strategy using different image quality metrics and statistical tests. The experimental results, incorporating both visual and statistical inferences, demonstrate the superiority and efficiency of the model. Additionally, an ablation analysis is performed to determine optimal regularization parameters, and the significance of integrating attention modules at different architecture layers is also demonstrated. © 2023 Elsevier LtdItem A weighted nuclear norm (WNN)-based retinex DIP framework for restoring aerial and satellite images corrupted by gamma distributed speckle noise(Springer, 2024) Shastry, A.; Padikkal, J.; George, S.; Bini, A.A.Restoration and enhancement are crucial preprocessing steps in the satellite domain. Mainly in active remote sensing such as Synthetic Aperture Radar (SAR), the images are more prone to speckle distortions and their reduction is not so trivial. Traditional deep learning models require large training datasets, limiting their applicability. This paper introduces a novel approach that combines the Deep Image Prior (DIP) model with a weighted nuclear norm (WNN) within a variational retinex framework to address these challenges. DIP leverages prior knowledge about noise distribution and works effectively with a single noisy image, eliminating the need for a large number of training images or ground truth. The WNN assigns non-negative weights to singular values, capturing the significance of each value and preserving crucial information during restoration. This approach offers a promising solution for satellite image restoration without relying on huge training data. The proposed method is evaluated through extensive experiments using various image quality metrics, including PSNR, SSIM, ENL, CNR, Entropy, and GCF. The comparative studies provide compelling evidence that the proposed method surpasses existing techniques in effectively restoring and enhancing speckled input images. Furthermore, statistical analysis performed using the Friedman test demonstrates the superior denoising performance of the model. Additionally, an ablation study is conducted to empirically determine the optimal regularization parameters, ensuring the optimal performance of the model. However, the theoretical selection of parameters for achieving optimal results remains an area that requires further exploration. © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2023.