Faculty Publications
Permanent URI for this communityhttps://idr.nitk.ac.in/handle/123456789/18736
Publications by NITK Faculty
Browse
7 results
Search Results
Item Implementation of Reversible Logic Gates with Quantum Gates(Institute of Electrical and Electronics Engineers Inc., 2021) Mummadi, M.; Rudra, B.Quantum is an emerging technology in future computers. Reversibility is the main advantage of quantum computers. In conventional computers, the computation is irreversible i.e. the input bits are lost once the logic block generates the output and input bits cannot be restored but it can be done in reversible computation because in reversible computation the inputs and outputs have a one-to-one correspondence. Therefore, a reversible gate input could even be uniquely determined from their output which leads to less power consumption. Hence the complexity of the digital circuits can be reduced by using reversible computing. In quantum computer to perform reversible operations, we need to implement the reversible gates using quantum gates. In this paper, we discussed various reversible logic gates like Feynman, Toffoli, R, Peres and TR gates using basic quantum gates like CNOT, Pauli, Swap gates and their implementation using IBM quantum experience. © 2021 IEEE.Item Quantum-Enhanced Deep Q Learning with Parametrized Quantum Circuit(Institute of Electrical and Electronics Engineers Inc., 2024) Manjunath, T.D.; Bhowmik, B.Quantum Computing (QC) is fastly growing to replace current computing systems for many high-performance applications. Due to the rapid increase of computational power, machine learning models based on artificial neural networks (ANN) have become highly effective. Even though classical machine learning models have been performing well, quantum computing with machine learning, i.e., Quantum Machine Learning (QML), will enhance the performance in multiple ways. Subsequently, Deep Q learning is a prominent approach to reinforcement learning used in solving complex applications for a desired performance. But this performance can be improved using a quantum computer. This paper proposes a quantum-enhanced Deep Q Learning algorithm to see its advantages over its classical counterparts. We implement our approach using parametrized quantum circuit (PQC). The evaluation of the proposed method achieves the metrics rewards and episodes up to 500 points and 1200, respectively. The performance shows that it collects roughly four times more rewards in a given time and takes significantly fewer episodes to converge compared to the classical approach. © 2024 IEEE.Item Deepfake Audio Detection Using Quantum Learning Models(Institute of Electrical and Electronics Engineers Inc., 2024) Pandey, A.; Rudra, B.Artificial intelligence makes it easy for humans to create high-quality images, speech, audio dubbing, and more. However, this technology is often misused to create fake content, such as phony speech, which is then made public to tarnish someone's image. This technology is known as deepfake, which uses deep learning, a field of artificial intelligence, to generate fake content. Advancements in deepfake technology pose the challenge of detecting fake content. Although many classical models exist to detect fake content, they often do not consider suitable audio features, and training these classical models is resource-intensive. Therefore, in this paper, we use a recently created real-time AI-generated fake speech dataset and propose a method to detect fake content using quantum learning models. This emerging technology leverages the properties of quantum mechanics to increase processing speed. We have trained the quantum learning models using the Lightning Qubit simulator. © 2024 IEEE.Item CCD Sensor Based Cameras for Sustainable Streaming IoT Applications With Compressed Sensing(Institute of Electrical and Electronics Engineers Inc., 2023) Gambheer, R.; Bhat, M.S.This paper presents a comprehensive study of compressed sensing (CS) techniques applied to Charge Coupled Device (CCD) and Complementary Metal-Oxide Semiconductor (CMOS) sensor-based cameras. CS is a powerful technique for reducing the number of measurements required to capture high-quality images while maintaining a high signal-to-noise ratio (SNR). In this study, we propose a novel CS method for CCD and CMOS sensor-based cameras that combines a new sampling scheme with a sparsity-inducing transform and a reconstruction algorithm to achieve high-quality images with fewer measurements. This paper focuses on an efficient CCD image capturing system suitable for embedded IoT applications. Hardware implementation has been done for proof of concept with an onboard Field Programmable Gate Array (FPGA) performing the compression. This hardware module is used over a wireless network to transmit and receive images under different test conditions with both CMOS and CCD sensors. For each use case, Peak Signal to Noise Ratio (PSNR), average power, and memory usage are computed under different ambient lighting conditions from dark to very bright. The results show that, a 640× 480 CCD sensor with compressed sensing with a sparsity of 0.5, provides 13% power saving and 15% memory saving compared to uncompressed sensing in no-light condition, resulting in 25.76 dB PSNR. Whereas, in no light condition, CMOS sensor does not capture any image at all. These results shows that the CCD image capturing system with compressed sensing can be conveniently used for embedded IoT applications. The data recovery from wireless sensor network is done at a central office where computing time and processing power resources are not constrained. The weight of the CCD camera is approximately 100 grams with modular build approach. © 2013 IEEE.Item Quantum-inspired hybrid algorithm for image classification and segmentation: Q-Means++ max-cut method(John Wiley and Sons Inc, 2024) Roy, S.K.; Rudra, B.Finding brain tumors is a crucial step in medical diagnosis that can have a big impact on how patients turn out. Conventional detection techniques can be laborious and demand a lot of computing power. Brain tumor detection could be made more effective and precise, thanks to the quickly developing field of quantum computing. In this article, we propose a quantum machine learning (QML)-based method for brain tumor extraction and detection based on quantum computing. To implement our strategy, we use a Hybrid Quantum-Classical Convolutional Neural Network (HQC-CNN) that has been trained using a collection of brain MRI images. Additionally, we employ Batchwise Q-Means++ Clustering for segmenting the images and a Max-cut approach with Adiabatic Quantum Computation (AQC) to extract the tumor region from the segmented MRI image. Our results highlight the strength of Quanvolutional Layer in Neural Network and reduced time complexity exponentially or quadratically in clustering and max-cut algorithms respectively and see the potential of quantum computing for improving the accuracy and speed of medical diagnosis and have implications for the future of healthcare technology. © 2024 Wiley Periodicals LLC.Item Quantum-inspired Arecanut X-ray image classification using transfer learning(John Wiley and Sons Inc, 2024) Naik, P.; Rudra, B.Arecanut X-ray images accurately represent their internal structure. A comparative analysis of transfer learning-based classification, employing both a traditional convolutional neural network (CNN) and an advanced quantum convolutional neural network (QCNN) approach is conducted. The investigation explores various transfer learning models with different sizes to identify the most suitable one for achieving enhanced accuracy. The Shufflenet model with a scale factor of 2.0 attains the highest classification accuracy of 97.72% using the QCNN approach, with a model size of 28.40 MB. Out of the 12 transfer learning models tested, 9 exhibit improved classification accuracy when using QCNN models compared to the traditional CNN-based transfer learning approach. Consequently, the exploration of CNN and QCNN-based classification reveals that QCNN outperforms traditional CNN models in accuracy within the transfer learning framework. Further experiments with qubits suggest that utilising 4 qubits is optimal for classification operations in this context. © 2024 The Author(s). IET Quantum Communication published by John Wiley & Sons Ltd on behalf of The Institution of Engineering and Technology.Item Framework for Quantum-Based Deepfake Video Detection (Without Audio)(John Wiley and Sons Inc, 2025) Pandey, A.; Rudra, B.; Kumar Krishnan, R.Artificial intelligence (AI) has made human tasks easier compared to earlier days. It has revolutionized various domains, from paper drafting to video editing. However, some individuals exploit AI to create deceptive content, such as fake videos, audios, and images, to mislead others. To address this, researchers and large corporations have proposed solutions for detecting fake content using classical deep learning models. However, these models often suffer from a large number of trainable parameters, which leads to large model sizes and, consequently, computational intensive. To overcome these limitations, we propose various hybrid classical–quantum models that use a classical pre-trained model as a front-end feature extractor, followed by a quantum-based LSTM network, that is, QLSTM. These pre-trained models are based on the ResNet architecture, such as ResNet34, 50, and 101. We have compared the performance of the proposed models with their classical counterparts. These proposed models combine the strengths of classical and quantum systems for the detection of deepfake video (without audio). Our results indicate that the proposed models significantly reduce the number of trainable parameters, as well as quantum long short-term memory (QLSTM) parameters, which leads to a smaller model size than the classical models. Despite the reduced parameter, the performance of the proposed models is either superior to or comparable with that of their classical equivalent. The proposed hybrid quantum models, that is, ResNet34-QLSTM, ResNet50-QLSTM, and ResNet101-QLSTM, achieve a reduction of approximately 1.50%, 4.59%, and 5.24% in total trainable parameters compared to their equivalent classical models, respectively. Additionally, QLSTM linked with the proposed models reduces its trainable parameters by 99.02%, 99.16%, and 99.55%, respectively, compared to equivalent classical LSTM. This significant reduction highlights the efficiency of the quantum-based network in terms of resource usage. The trained model sizes of the proposed models are 81.35, 88.06, and 162.79, and their equivalent classical models are 82.59, 92.28, and 171.76 in MB, respectively. © © 2025 Atul Pandey et al. International Journal of Intelligent Systems published by John Wiley & Sons Ltd.