Faculty Publications
Permanent URI for this communityhttps://idr.nitk.ac.in/handle/123456789/18736
Publications by NITK Faculty
Browse
6 results
Search Results
Item Alternate Approaches to Scattering Networks in Image Classification(Institute of Electrical and Electronics Engineers Inc., 2023) Rao, S.; Varma, V.Scattering networks are a special class of convolutional neural networks (CNNs) which implement a windowed scattering transform in their initial layers while learning the rest of the parameters. For classification tasks requiring little data, scattering networks beat cutting-edge deep neural networks. When given a huge dataset, their performance is comparable to end-to-end trained networks, but they're better suited for real-time applications due to their lower latency. The use of a windowed scattering transform for tasks involving image classification on the CIFAR-10 dataset is examined in this paper. We replace the 2-D Gabor filterbank in the state-of-the-art scattering network with alternate filterbanks that provide better directional separation in the frequency domain. We also develop a trainable directional filterbank that uses data-driven directional filters in its construction. The directional filters are built using the weights of a 2D convolutional operator. We demonstrate the performance of the alternate approaches in supervised classification tasks and observe that the trainable implementation outperforms the traditional scattering networks. © 2023 IEEE.Item EXhype: A tool for mineral classification using hyperspectral data(Elsevier B.V., 2017) Adep, R.N.; Shetty, A.; Ramesh, H.Various supervised classification algorithms have been developed to classify earth surface features using hyperspectral data. Each algorithm is modelled based on different human expertises. However, the performance of conventional algorithms is not satisfactory to map especially the minerals in view of their typical spectral responses. This study introduces a new expert system named ‘EXhype (Expert system for hyperspectral data classification)’ to map minerals. The system incorporates human expertise at several stages of it's implementation: (i) to deal with intra-class variation; (ii) to identify absorption features; (iii) to discriminate spectra by considering absorption features, non-absorption features and by full spectra comparison; and (iv) finally takes a decision based on learning and by emphasizing most important features. It is developed using a knowledge base consisting of an Optimal Spectral Library, Segmented Upper Hull method, Spectral Angle Mapper (SAM) and Artificial Neural Network. The performance of the EXhype is compared with a traditional, most commonly used SAM algorithm using Airborne Visible/Infrared Imaging Spectrometer (AVIRIS) data acquired over Cuprite, Nevada, USA. A virtual verification method is used to collect samples information for accuracy assessment. Further, a modified accuracy assessment method is used to get a real users accuracies in cases where only limited or desired classes are considered for classification. With the modified accuracy assessment method, SAM and EXhype yields an overall accuracy of 60.35% and 90.75% and the kappa coefficient of 0.51 and 0.89 respectively. It was also found that the virtual verification method allows to use most desired stratified random sampling method and eliminates all the difficulties associated with it. The experimental results show that EXhype is not only producing better accuracy compared to traditional SAM but, can also rightly classify the minerals. It is proficient in avoiding misclassification between target classes when applied on minerals. © 2016 International Society for Photogrammetry and Remote Sensing, Inc. (ISPRS)Item GPUPeP: Parallel Enzymatic Numerical P System simulator with a Python-based interface(Elsevier Ireland Ltd, 2020) Raghavan, S.; Rai, S.S.; Rohit, M.P.; Chandrasekaran, K.Membrane computing is a computational paradigm inspired by the structure and behavior of a living cell. P Systems are the computing devices that are used to realize membrane computing models. Numerous theoretical studies on many variants of P Systems have shown them to be computationally universal. There is a wide range of applications of P Systems from modeling of biological processes to image processing. Among many variants of P Systems, one of the most important is Enzymatic Numerical P System (ENPS). ENPS is a class of P System in which membranes operate on numerical values. To realize the power of ENPS there are a few simulators developed. Each and every simulator has some advantages as well as some disadvantages. Here, a GPU based simulator using Python as a user interaction language is developed. This tool is a completely parallel variant, compatible with a Python based sequential simulator (PeP) which was the first Python based work for ENPS. The developed simulator uses CUDA to interact with GPU and gives the desired speed up, while processing the membranes. There are two important case studies which show the performance of the developed tool to be far better than the other serial simulators. © 2020 Elsevier B.V.Item Dehazing of Satellite Images using Adaptive Black Widow Optimization-based framework(Taylor and Francis Ltd., 2021) Suresh, S.; Rajan, M.; Pushparaj, J.; Cs, A.; Lal, S.; Chintala, C.S.Haze is a common atmospheric disturbance that adversely affects the quality of optical data, thus often restricting their usability. Since these effects are inherent in the process of spaceborne Earth sensing, it is important to develop effective methods to remove them. This work proposes a novel method for de-hazing satellite imagery and outdoor camera images. It is developed by modifying the transmission map used in Dark Channel Prior (DCP) method. A Weighted Variance Guided Filter (WVGF) is introduced for enhancing the image quality, which included a two-stage image decomposition and fusion process. The method also optimally combines the radiance and transmission components along with an additional stage modelling a fusion-based transparency function. A final guided filter-based image refinement scheme is incorporated to improve the processed image quality. The optimal tuning of the image-dependent parameters at various stages is achieved using the newly proposed Adaptive Black Widow Optimization (ABWO) algorithm, which makes the proposed de-hazing scheme fully automatic. Qualitative and quantitative performance analyses, and the results are compared with other state-of-the-art methods. The experimental results reveal that the proposed method performs better as compared with others, independent of the haze density, without losing the natural look of the scene. © 2021 Informa UK Limited, trading as Taylor & Francis Group.Item INFLUENCE OF LAND USE LAND COVER CHANGE ON RUNOFF CHARACTERISTICS OF NETRAVATHI RIVER CATCHMENT, KARNATAKA, INDIA(Zibeline International Publishing Sdn. Bhd., 2024) Dwarakish, G.S.; Pai, J.B.; Jubina, C.K.The effect of LU/LC on the streamflow characteristics of the Netravathi river basin, Karnataka, India, is studied using Soil and Water Assessment Tool (SWAT) model. Landsat images, soil map from FAO, ASTER DEM (30m grid) and streamflow data, forms the database for the present work. The most significant changes from 1981 to 2015, in the LU/LC includes agricultural land (31.86%), built-up area (67.9%), forest cover (-20.01%), coconut plantation (55.12%), other vegetation (-18.55%) and others (-11.82%). The verification of performance of model was carried out by the coefficient of determination values (R2 > 0.8) and N S E (NSE > 0.78) were obtained and hence proved that SWAT model performance in estimating streamflow.. The average streamflow is increased by 13.74% from 1981 to 2015, which is mainly due to dynamic changes in LU/LC. Hence, it can be concluded that changes in LU/LC have a direct impact on streamflow in the study area. © 2024, Zibeline International Publishing Sdn. Bhd.. All rights reserved.Item EffiCOVID-net: A highly efficient convolutional neural network for COVID-19 diagnosis using chest X-ray imaging(Academic Press Inc., 2025) Kumar, S.; Bhowmik, B.The global COVID-19 pandemic has drastically affected daily life, emphasizing the urgent need for early and accurate detection to provide adequate medical treatment, especially with limited antiviral options. Chest X-ray imaging has proven crucial for distinguishing COVID-19 from other respiratory conditions, providing an essential diagnostic tool. Deep learning (DL)-based models have proven highly effective in image diagnostics in recent years. Many of these models are computationally intensive and prone to overfitting, especially when trained on limited datasets. Additionally, conventional models often fail to capture multi-scale features, reducing diagnostic accuracy. This paper proposed a highly efficient convolutional neural network (CNN) called EffiCOVID-Net, incorporating diverse feature learning units. The proposed model consists of a bunch of EffiCOVID blocks that incorporate several layers of convolution containing (3×3) filters and recurrent connections to extract complex features while preserving spatial integrity. The performance of EffiCOVID-Net is rigorously evaluated using standard performance metrics on two publicly available COVID-19 chest X-ray datasets. Experimental results demonstrate that EffiCOVID-Net outperforms existing models, achieving 98.68% accuracy on the COVID-19 radiography dataset (D1), 98.55% on the curated chest X-ray dataset (D2), and 98.87% on the mixed dataset (DMix) in multi-class classification (COVID-19 vs. Normal vs. Pneumonia). For binary classification (COVID-19 vs. Normal), the model attains 99.06%, 99.78%, and 99.07% accuracy, respectively. Integrating Grad-CAM-based visualizations further enhances interpretability by highlighting critical regions influencing model predictions. EffiCOVID-Net's lightweight architecture ensures low computational overhead, making it suitable for deployment in resource-constrained clinical settings. A comparative analysis with existing methods highlights its superior accuracy, efficiency, and robustness performance. However, while the model enhances diagnostic workflows, it is best utilized as an assistive tool rather than a standalone diagnostic method. © 2025 Elsevier Inc.