Conference Papers
Permanent URI for this collectionhttps://idr.nitk.ac.in/handle/123456789/28506
Browse
5 results
Search Results
Item M-CAD: Towards Multi-Categorical Auto Diagnosis of Varied Diseases using Deep Learning(Institute of Electrical and Electronics Engineers Inc., 2021) Praveen, K.; Patil, N.; Srikanth, C.S.; Nayaka, J.The economic burden and the number of lives lost due to diagnostic errors are higher than ever due to the onset of pandemics and new viruses, Specially in medium and low-economic status nations (including India) are affected heavily in terms of capital and human resources. Due to limited expertise in diagnostic technologies in remote parts of India and many low-economic nations of Africa, autonomous diagnostics can save millions of lives and lower the costs. To accomplish this goal we propose a method that uses modern developments in Deep Learning in semantic segmentation and classification to predict multiple diseases from multiple medical images. To conduct the study we test the model with Dermoscopy images and CT-Scans to predict 8 classes relating to Melanoma cancer, Covid-19 virus and different types of Carcinoma. The setup is tested on largest publicly available ISIC Dermoscopy dataset, 1061 CT-scan images combined for the classification and Segmentation(only for Melanoma). Classification model(M-CAD) is progressively tested by increasing the number of classes and data that it trains on. This pilot study is conducted on a small subset of the complete data, segmentation of Melanoma images obtained an accuracy of 96.6% compared to human expert agreement which is 90.9%. we were able to produce average accuracy of 81.5% and AUC of 0.94 for 6 classes using CT-Scans whereas accuracy and AUC for all the 8 classes is 80.2% and 0.97 respectively. These results were quite promising for a model that classifies different images with no apparent relation at all. © 2021 IEEE.Item LiCamPos : An Indoor Positioning System using Light to Camera Communication(Institute of Electrical and Electronics Engineers Inc., 2021) Salvi, S.; Geetha, V.; Praveen, K.; Kiran, C.; Nayaka, V.J.Due to the limitations of the Global Positioning System to track or locate ultra-small resolutions, Indoor Positioning System(IPS) came into existence. IPS is used to provide useful indoor location-based information. Mostly Radio Frequency (RF) based system is used to achieve IPS. However, alternatives to RF are getting popular due to the debatable health risks of RF. In this paper, a light-based camera communication technique for IPS is proposed and evaluated. The proposed method uses a light aperture to encode location beacon information and is decoded using image processing techniques on the user's mobile phone. An On-Off Keying (OOK) modulation technique is employed to encode and broadcast beacon information, and software-based custom modified mobile camera parameters are used for position estimation. A phenomenon called the "Rolling Shutter"effect is used to extract useful information from the beacon. During experimentation, it was observed that the number of observable rolling shutter stripes is inversely proportional to the distance between the beacon and the receiver, and it is directly proportional to the frequency of OOK. It was also observed that the stripe's width is independent of the distance between beacon and receiver. Finally, an android application is built to show location-relevant messages/ads based on the identified beacon. © 2021 IEEE.Item Sliding wear behaviour of Ni-5 %Al coating deposited by detonation spray on IN718(Elsevier Ltd, 2022) Purushotham, N.; Rajasekaran, B.; Parthasarathi, N.L.; Praveen, K.; Govindarajan, G.Ni-5 %Al metallic coating was deposited on Nickel-based superalloy (IN718) specimens using the detonation spray coating (DSC) method. Detonation spraying yielded coating with extreme chemical bond strength, hardness, and less porosity. The microstructure, microhardness, and room temperature pin-on-disc sliding wear behavior of the Ni-5 %Al coating and the as-received IN718 superalloy were evaluated. Sliding wear tests were done at room temperature (25 °C), under different loading conditions (6 N and 10 N), using an alumina (Al2O3) ball-on-disk tribometer and friction coefficients were measured. The study of worn surfaces conducted by SEM indicated that both Ni-5 %Al coating and the substrate suffered significant abrasive wear with occasional adhesion and spalling of the coating. The 3D topography of the wear track was examined by a 3D non-contact profilometer, which enabled the quantification of the wear. The friction coefficient values of the tests and the wear in terms of mass loss were in good correlation. © 2022Item SCaLAR NITK at Touché: Comparative Analysis of Machine Learning Models for Human Value Identification(CEUR-WS, 2024) Praveen, K.; Darshan, R.K.; Reddy, C.T.; Anand Kumar, M.This study delves into task of detecting human values in textual data by making use of Natural Language Processing (NLP) techniques. With the increasing use of social media and other platforms, there is an abundance in data that is generated. Finding human values in these text data will help us to understand and analyze human behavior in a better way, because these values are the core principle that influence human behavior. Analyzing these human values will help not only in research but also for practical applications such as sentiment evaluation, market analysis and personalized recommendation systems. The study tries to evaluate the performance of different existing models along with proposing novel techniques. Models used in this study range from simple machine learning model like SVM, KNN and Random Forest algorithms for classification using embeddings obtained from BERT till transformer models like BERT and RoBERTa for text classification and Large Language Models like Mistral-7b. The task that has be performed is a multilabel, multitask classification. QLoRA quantization method is used for reducing the size of weights of the model which makes it computationally less expensive for training and Supervised Fine Tuning (SFT) trainer is used for fine tuning LLMs for this specific task. It was found that LLMs performed better compared to all other models. © 2024 Copyright for this paper by its authors.Item An Effective Approach for Deepfake Video Detection using Binarized Neural Network(Institute of Electrical and Electronics Engineers Inc., 2025) Praveen, K.; Pandey, A.; Rudra, B.The rise of DeepFake technologies, especially in audio and video, poses significant threats to information integrity, security, and privacy. Artificially driven Artificial Intelligence (AI) methods and their advancement make it difficult to trace synthetic media through deepfakes that closely approximate real speech, facial expressions, and body movements. Consequently, traditional methods of detecting these are losing the race because they cannot compete with the newly invented methods that are more advanced in comparison. This paper proposes a lightweight and scalable approach to deepfake video detection using Binarized Neural Networks (BNNs). We integrate BNNs with Convolutional Neural Networks (CNNs) and Multi-task Cascaded Convolutional Networks (MTCNN) to boost feature extraction and analysis while making sure that this is done at a computational efficiency, especially to be deployed in resource-constrained systems such as mobile and embedded devices. The binarization of network weights and activations naturally deals with the trade-off regarding detection accuracy and computational cost. Our approach introduces a practical solution for real-time deepfake detection, thus advancing toward more secure and trusted digital environments. Our proposed model has achieved an accuracy of 80%. © 2025 IEEE.