Faculty Publications
Permanent URI for this communityhttps://idr.nitk.ac.in/handle/123456789/18736
Publications by NITK Faculty
Browse
2 results
Search Results
Item Secure transmission of Hyperspectral Images(Institute of Electrical and Electronics Engineers Inc., 2020) Srujana, O.S.; Mhala, N.C.; Pais, A.R.Hyperspectral Images (HSIs) are images that are captured across the electromagnetic spectrum. Hyperspectral images are represented using a three-dimensional (x,y,λ) data cube where dimension x and y represent the spatial dimension of a scene, and λ represents the spectral dimension of a scene. These images contain abundant information that has to be transmitted securely among the users for further processing. Visual Secret Sharing (VSS) is a modern cryptographic method used to send the visual data securely among n users. VSS scheme generates multiple shares of the secret image, and to recover secret image, these all shares need to be stacked together.In this paper, we propose a scheme for the secure transmission of hyperspectral images using VSS. We also introduce a band selection technique as a pre-processing step to reduce the redundancy and size of the image cube. The proposed scheme uses the super-resolution concept to increase the contrast of the resultant image, obtained from the VSS. We have performed the visual quality assessment of the reconstructed image using quantitative measurement parameters and we have compared the results with the existing VSS based Randomised VSS (RVSS) scheme. The experimental results showed that our proposed scheme achieves better reconstruction quality than the RVSS scheme and reconstructs HSI with a similarity of almost 76-95%. © 2020 IEEE.Item An Illumination Invariant Approach to Salient Object Detection in Hyperspectral Images(Institute of Electrical and Electronics Engineers Inc., 2025) Lone, Z.A.; Pais, A.R.; Murali Krishna, M.M.; Mhala, N.C.Salient Object Detection (SOD) in Hyperspectral Images (HSI) has traditionally relied on low-level features, with limited exploration of high-level features. Low-level features lack semantic context, making them ineffective for detecting complex or subtle salient objects. Moreover, the inherent sensitivity of HSI to uneven illumination poses additional challenges, particularly when there is uneven illumination on the objects. To address these challenges, a novel SOD algorithm is proposed, that employs manifold ranking on high-level features learned through a deep segmentation network. Unlike existing methods that use raw HSI as input, the proposed approach utilizes spectral gradient as input, mitigating the effects of spectral variation. The learned features are ranked using a graph-based manifold ranking approach, which enhances the robustness of the detection process. Experimental evaluation on the HS-SOD dataset demonstrates that the proposed approach performs better, achieving superior results compared to state-of-the-art methods, with a precision of 0.63, an f1 score of 0.51, and a Mean Absolute Error (MAE) of 0.08. © 2025 IEEE.