Multi-Scale Deep Transfer Learning and Attention-Guided Feature Fusion for Robust Cervical Cell Categorization J Pradeep, M Harikrishnan, Sanjay P, Mohamed Siddiq R, Shivsharan N Proceedings of 2nd International Conference on Multi Agent Systems for Collaborative Intelligence Icmsci 2026, 2026 Conventional cervical cancer diagnosis, including Pap smear and histopathological examination, is frequently subjective, time-consuming and variable. To tackle the above limitations, in this paper, this work presents a robust AI-based framework for automated cervical cell classification, termed Multi-Scale Deep Transfer Learning and Attention-Guided Feature Fusion. The model combines several pre-trained CNN models (EfficientNetB0, InceptionV3, ConvNeXt-Tiny, Swin Transformer, DenseNet121, and MobileNetV3-Large) to generate multi-scale features. An attention guided feature fusion module with SE and CBAM mechanisms is utilized to improve the feature representation by focusing on the diagnostic-related regions. The addition of Focal Loss with Label Smoothing enhances class balance management and mitigates overconfidence. In addition, the stacked RBM network in the proposed hybrid deep learning method can be visually explained by Grad-CAM heatmaps, providing incremental value not only in boosting the classification accuracy and its generalization for imbalanced datasets but also visual explanatory capability. The proposed framework achieves a classification accuracy of 97.4%, with precision of 97.1%, recall of 97.6%, and an F 1 score of 97.3% on the SIPaKMeD dataset. Thus, the proposed scheme offers a scalable, data-driven, and noninvasive approach to the accurate early screening of cervical cancer.
Low-Power Non-Volatile Flip-Flop Design Using NAND-Spin Logic for Energy Efficient Systems J. Pradeep, S. Radhika 2025 International Conference on Emerging Technologies in Engineering Applications Icetea 2025 Proceedings, 2025 Non-Volatile Flip-Flops (NVFFs) are intended for addressing the growing leakage power consumption problem linked with CMOS technology advancements. As complexity increases, power dissipation has become a key design restriction. The demand for quick, low-power devices and digital circuits has increased rapidly during the past several years. The use and exploitation of the FF have gained popularity in the research area of low-power and voltage VLSI circuits at the gate level, as well as low-power applications systems. Power Gating (PG) techniques in NVFFs have the ability to solve the rising leakage power consumption problem caused by CMOS (Complex Metal Oxide semiconductor) technology advancement. The MTJs is excellent in developing the NVFF due to its less power consumption, fast efficiency, and greater CMOS integration. A new magnetic NVFF based on NAND-SPIN, a recently created storage technology. By using the SPIN Orbit Torque (SOT) and Spin Transfer Torque (STT) the data erasing function and the programming functions are done respectively. The feature enables us to create a magnetized NVFF that is both erasehidden and improves drivability. Moreover, it can be done by the principle of shorted gate FINFET logic, which intends to improve the noise factor, and achieve high write speeds and endurance. Furthermore, the study will address scalability difficulties at the nanoscale. According to simulation results, the suggested NVFF outperforms the traditional slave- latchdriven, hold-up time, and power.
Frequency-Adaptive Low-Light Image Enhancement Using Laplacian Pyramid and Blending Scheme R. Kurinjimalar, J. Pradeep, M. Harikrishnan 2025 International Conference on Emerging Technologies in Engineering Applications Icetea 2025 Proceedings, 2025 Image processing is essential in many fields, including security, medical image analysis, remote sensing, face detection, and pattern recognition. However, the enhancement of low-light images remains a significant challenge due to issues like detail loss, noise amplification, and limited dynamic range. Existing systems, including global contrast enhancement and histogram equalization, often fall short in preserving image details or introduce visual artifacts under poor lighting conditions. The Laplacian low-light image enhancement technique uses gamma correction to purposefully underexpose the image, which overcomes the difficulty of low-light conditions. Several exposed images are produced when frequency domain performance is improved. These images are then combined using a multi-scale Laplacian blending strategy to improve the output. By manipulating gamma values, a set of images with varying exposures is created, enabling the algorithm to capture a broader range of information in low-light conditions. Extensive experimental evaluations, encompassing qualitative and quantitative analysis, validate the method's effectiveness to preserve details while minimizing artifacts. The results show significant improvements in decreasing the mean-squared value (MSE) and improving the pixel-to-noise ratio (PSNR), especially in scenarios where traditional methods struggle. Optimizing the transformation in the frequency domain emphasizes the potential of multi-scale. Laplacian blending as a valuable tool in addressing challenges posed by low-light conditions, marking substantial advancements in image enhancement.
Radar Cross SectionReduction Using Metasurfaces Methods, Metrics, and Future Scope Madurakavi Karthikeyan, Sukumar Vidhya, Natarajan Rajesh, J Pradeep, M Harikrishnan Terahertz Technology in Microwave and Photonics for Effective Communications, 2025 This chapter deliberates metasurface based Radar Cross Section reduction (RCSR) techniques. Initially, the importance of metamaterial is emphasized in various domains and also its prominence in RCSR is discussed in detail. The further discussion starts with the evolution of metamaterials and its intrusion in various RCSR techniques. A thorough detailing is presented on the traditional methods for reducing RCS. RCSR focuses on two main categories: scattering and absorption. There are two major RCSR techniques: shaping and radar absorbing material (RAM), and an in-depth analysis is provided on the abovementioned methods. A thorough discussion is provided over active and passive methods for reducing RCS based on the metamaterials. The methods for reducing RCS are considered based on the parameters such as bandwidth, profile, polarization sensitivity, angular stability, conformality, cost-effectiveness, and design complexity. In this chapter, a summary is presented on the methods to reduce RCS, and a comparative tabulation is included for each method, which connects its size, operational bandwidth, fractional bandwidth, and the optimization method. The chapter also includes the research and investigation needed that can be performed in future.
Lora Based Remote Monitoring Emergency System for Underground Miners Suseendhar. P, T. S. Krishna Priya, J. Pradeep, Koushik Kabilan. A, Balakumaran. R, Keerthivasan. S 2025 International Conference on Emerging Technologies in Engineering Applications Icetea 2025 Proceedings, 2025 Coal mining is a high-risk occupation, where exposure to hazardous gases such as methane (CH₄) and low blood oxygen levels (SpO₂) can lead to severe health risks and accidents. This project proposes a Smart Mining Safety System that integrates CH₄ and SpO₂ sensors into a wearable device for real-time monitoring of miners’ health and environmental conditions. The system continuously tracks CH₄ concentrations and SpO₂ levels, comparing them to predefined safety thresholds. Also, the miners’ helmet is provided with an SOS button, which when pressed triggers an alert and the miner's location is transmitted to a control unit using an RSSI-based location tracking over LoRa technology. Workers can make quick decisions by monitoring remotely through the control unit that displays real-time data with its user-friendly interface as well as alarm features for alerts. The control unit can also send SOS signals to the miners for immediate evacuation in emergency situations. Health and safety of individuals remain protected through automatic system responses that also enable rescue professionals to act quickly. This proposed system implements fresh technological solutions to lower dangers in the coal mines by ensuring miners' safety. Through the combination of environmental monitoring technologies and location tracking service along with emergency response features this system brings important safety improvements to underground mining operations. Reliable data transfer through low-power LoRa communication systems improves performance of this underground system.
Absorbers as biosensors: Leveraging absorption phenomena for enhanced biosensing Madurakavi Karthikeyan, J Pradeep, M Harikrishnan, R Sitharthan, M Rajesh, Avinash Chandra, Rajkishor Kumar Advances in all Optical Communication, 2024 The purpose of this chapter's exploration of absorber-based biosensors is to highlight how they might enhance chemical and biological sensing's sensitivity and selectivity. Through the measurement of variations in light absorption associated with the refractive index of the surrounding media, these sensors provide real-time, label-free biomolecule, gas, and liquid detection. Notwithstanding their extreme sensitivity and adaptability, drawbacks include integration problems, noise interference, and sensitivity limits. In an effort to stimulate more study and creativity in this rapidly evolving sector, the chapter addresses the mechanics of absorption, developments in material development, and future directions.
Underwater Image Enhancement Using Gaussian Pyramid, Laplacian Pyramid and Contrast Limited Adaptive Histogram Equalization R. Kurinjimalar, J. Pradeep, M. Harikrishnan 2024 IEEE 3rd World Conference on Applied Intelligence and Computing Aic 2024, 2024 Underwater imaging is essential for capturing submerged environments and it faces significant challenges due to light attenuation, water turbidity and color distortion. This paper presents a novel technique to enlighten underwater image quality through a combination of multiscale image processing techniques like Gaussian and Laplacian pyramids, combined with Contrast Limited Adaptive Histogram Equalization (CLAHE) to enhance underwater image. In this paper, five different datasets namely, UIEB, EUVP, RUIE, USR and U45 datasets are used to representing various underwater environments and conditions. Four filtering techniques are employed to sharpen images and the resulting enhancements are showcased. From the result, it is found that the parameters including PSNR, UISM, UICM, PCQI and UIQM reach their maximum value for Laplacian of Gaussian method. Conversely, the gamma correction technique exhibits better performance, particularly in terms of UCIQE and CCQ metrics. It is clear from the suggested method that is applied to the UIEB Real World Underwater Datasets various parameter values are evaluated. The PCQI value is 0.0119, the UIQM value is 0.9998, and the UCIQE value is 30.455, according to the results. It shown in comparison to the other approaches that have been proposed earlier, the UCIQE value is improvised. The proposed method offers a superior solution for enhancing underwater images.
Recent developments in flexible and printed reconfigurable antennas for medical and Internet of things applications Madurakavi Karthikeyan, J Pradeep, M Harikrishnan, R Sitharthan, Naveen Mishra, M Rajesh, Thamizharasan Sivanesan Advances in Flexible and Printed Electronics Materials Fabrication and Applications, 2023 There are numerous possibilities to achieve complex and life-saving applications thanks to high data rates and developing wireless communication standards. The comfort and ease with which humans live have been enhanced by modern electronic communication devices that take advantage of the Internet of things (IoT). IoT helps to solve various patient monitoring issues together with updates to implantable medical devices. The antenna is the main part of IoT and implantable devices. Due to shifting standards, novel antenna design is always needed for these applications. The innovative guidelines place strict criteria on antenna design elements such as operating frequency, gain, and size. Additionally, the majority of IoT and medical applications require that antenna performance be unaffected by antenna bending. Recently, due to emerging flexible substrates, a lot of sophisticated applications in the field of IoT and medical areas have been reported. The optimal substrate for the flexible antenna has a low dielectric loss, a small coefficient of thermal expansion, low relative primitivity, and high thermal conductivity. The substrateʼs dielectric constant must be high to reduce the antennaʼs size. The trade-off between efficiency and antenna design size is that increasing antenna efficiency will cost more with larger antenna sizes. To enable wireless communication, which is widely desired by todayʼs information-oriented culture, flexible antennas operating in specified frequency bands must be integrated into flexible electronic systems. For instance, industrial, scientific, and medical (ISM) frequency bands are used for biomedical antennas. The flexible antenna design in this frequency band is attracting a lot of researchers in the industry as well as academics. This chapter addresses progress in flexible antenna design, simulation, various novel flexible substrates, measurement, and suitability for various applications. Furthermore, various challenges that need to be addressed by the researchers have been presented.
Voice Recognition Using Natural Language Processing Big Data Management in Sensing Applications in AI and Iot, 2021
Smart IoT Assistant for Government Schemes and Policies Using Natural Language Processing Big Data Management in Sensing Applications in AI and Iot, 2021
An investigation on the performance of hybrid features for feed forward neural network based english handwritten character recognition system Wseas Transactions on Signal Processing, 2014
Multi-Scale Deep Transfer Learning and Attention-Guided Feature Fusion for Robust Cervical Cell Categorization J Pradeep, M Harikrishnan 2026 Second International Conference on Multi-Agent Systems for … , 2026 2026
Radar Cross Section Reduction Using Metasurfaces: Methods, Metrics, and Future Scope M Karthikeyan, S Vidhya, N Rajesh, J Pradeep, M Harikrishnan Terahertz Technology in Microwave and Photonics for Effective Communications … , 2025 2025 Citations: 1
Energy-Efficient GNRFET-Based Ternary Full Adder for Next-Generation Integrated Circuits J Pradeep, B Ramu Journal of VLSI Circuits and Systems 7 (1), 94-105 , 2025 2025
Frequency-Adaptive Low-Light Image Enhancement Using Laplacian Pyramid and Blending Scheme R Kurinjimalar, J Pradeep, M Harikrishnan 2025 International Conference on Emerging Technologies in Engineering … , 2025 2025 Citations: 1
Low-Power Non-Volatile Flip-Flop Design Using NAND-Spin Logic for Energy Efficient Systems J Pradeep, S Radhika 2025 International Conference on Emerging Technologies in Engineering … , 2025 2025
Lora Based Remote Monitoring Emergency System for Underground Miners TSK Priya, J Pradeep 2025 International Conference on Emerging Technologies in Engineering … , 2025 2025
Absorbers as biosensors: leveraging absorption phenomena for enhanced biosensing M Karthikeyan, J Pradeep, M Harikrishnan, R Sitharthan, M Rajesh, ... Advances in All-optical Communication, 11-1-11-15 , 2024 2024 Citations: 1
Underwater Image Enhancement Using Gaussian Pyramid, Laplacian Pyramid and Contrast Limited Adaptive Histogram Equalization R Kurinjimalar, J Pradeep, M Harikrishnan 2024 IEEE 3rd World Conference on Applied Intelligence and Computing (AIC … , 2024 2024 Citations: 6
Diabetic Retinopathy Detection Using Spiking Neural Network J Pradeep, M Harikrishnan International Conference on Data Science and Applications, 91-102 , 2024 2024
Recent developments in flexible and printed reconfigurable antennas for medical and Internet of things applications M Karthikeyan, J Pradeep, M Harikrishnan, R Sitharthan, N Mishra, ... Advances in Flexible and Printed Electronics: Materials, fabrication, and … , 2023 2023 Citations: 5
LIDAR & Thermal Camera Based Braking and Life Saving System With Actuator Shield for Wheels R Kurinjimalar, J Pradeep, K Jayapreethi, M Madhumitha, S Narmadha 2023 International Conference on System, Computation, Automation and … , 2023 2023 Citations: 1
Fault Detection in Windmills Using Augmented Reality HMML Arunagiri P, Pradeep J International Journal of Sensors, Wireless Communications and Control 13 (4) , 2023 2023
Enhanced Recognition system for Diabetic Retinopathy using Machine Learning with Deep Learning Approach J Pradeep, NE Jeffery, M Saranraj, JN Hussain Journal of Population Therapeutics and Clinical Pharmacology 30 (11), 452-466 , 2023 2023 Citations: 9
Tunable optimal dual band metamaterial absorber for high sensitivity THz refractive index sensing M Karthikeyan, P Jayabala, S Ramachandran, SS Dhanabalan, ... Nanomaterials 12 (15), 2693 , 2022 2022 Citations: 84
Automatic railway detection and tracking inspecting system J Pradeep, M Harikrishnan, K Vijayakumar Proceedings of International Conference on Data Science and Applications … , 2021 2021 Citations: 2
Smart IoT Assistant for Government Schemes and Policies Using Natural Language Processing J Pradeep, K Manojkiran, VP Gopi, B Jayakumar Big Data Management in Sensing: Applications in AI and IoT , 2021 2021 Citations: 1
Voice Recognition Using Natural Language Processing J Pradeep, K Vijayakumar, M Harikrishnan Big Data Management in Sensing: Applications in AI and IoT , River … , 2021 2021 Citations: 2
Efficient vehicle detection system using OCR and REST API J Pradeep, A Muthukumaran, G Nareshraman, R Krishnan, ... International Journal of Engineering Research & Technology (IJERT) 9 (7 … , 2020 2020 Citations: 3
Efficient Vehicle Detection System using OCR and REST API RKBSRP J. Pradeep, A. Muthukumaran, G. Nareshraman INTERNATIONAL JOURNAL OF ENGINEERING RESEARCH & TECHNOLOGY (IJERT) Volume 09 … , 2020 2020
Distributed entropy energy-efficient clustering algorithm for heterogeneous wireless sensor network based chaotic firefly algorithm cluster head selection J Pradeep, MPM Kumar Journal of Critical Reviews 7 (8), 1208-1215 , 2020 2020 Citations: 5
MOST CITED SCHOLAR PUBLICATIONS
Diagonal based feature extraction for handwritten character recognition system using neural network J Pradeep, E Srinivasan, S Himavathi 2011 3rd international conference on electronics computer technology 4, 364-368 , 2011 2011.0 Citations: 341
Neural network based handwritten character recognition system without feature extraction J Pradeep, E Srinivasan, S Himavathi 2011 international conference on computer, communication and electrical … , 2011 2011.0 Citations: 85
Tunable optimal dual band metamaterial absorber for high sensitivity THz refractive index sensing M Karthikeyan, P Jayabala, S Ramachandran, SS Dhanabalan, ... Nanomaterials 12 (15), 2693 , 2022 2022.0 Citations: 84
Neural network based recognition system integrating feature extraction and classification for english handwritten J Pradeep, E Srinivasan, S Himavathi International journal of Engineering 25 (2), 99-106 , 2012 2012.0 Citations: 84
Diagonal feature extraction based handwritten character system using neural network J Pradeep, E Srinivasan, S Himavathi International Journal of Computer Applications 8 (9), 17-22 , 2010 2010.0 Citations: 60
Path Planning of Autonomous Mobile Robots: A Survey and Comparison P Victerpaul, D Saravanan, S Janakiraman, J Pradeep Journal of Advanced Research in Dynamical and Control Systems 12, 1535-1565 , 2017 2017.0 Citations: 49
Performance Analysis of Hybrid Feature Extraction Technique for Recognizing English Handwritten Characters J Pradeep, E Srinivasan, S Himavathi Proceedings of Second World Congress on Information and Communication … , 2012 2012.0 Citations: 25
Design and Implementation of Gesture Controlled Robotic Arm for Industrial Applications J PRADEEP, PV PAUL International Journal of Advanced Scientific Research & Development 3 (4 … , 2016 2016.0 Citations: 23
An investigation on the performance of hybrid features for feed forward neural network based English handwritten character recognition system J Pradeep, E Srinivasan, S Himavathi WSEAS transactions on signal processing 10 (1), 21-29 , 2014 2014.0 Citations: 14
Enhanced Recognition system for Diabetic Retinopathy using Machine Learning with Deep Learning Approach J Pradeep, NE Jeffery, M Saranraj, JN Hussain Journal of Population Therapeutics and Clinical Pharmacology 30 (11), 452-466 , 2023 2023.0 Citations: 9
Underwater Image Enhancement Using Gaussian Pyramid, Laplacian Pyramid and Contrast Limited Adaptive Histogram Equalization R Kurinjimalar, J Pradeep, M Harikrishnan 2024 IEEE 3rd World Conference on Applied Intelligence and Computing (AIC … , 2024 2024.0 Citations: 6
Recent developments in flexible and printed reconfigurable antennas for medical and Internet of things applications M Karthikeyan, J Pradeep, M Harikrishnan, R Sitharthan, N Mishra, ... Advances in Flexible and Printed Electronics: Materials, fabrication, and … , 2023 2023.0 Citations: 5
Distributed entropy energy-efficient clustering algorithm for heterogeneous wireless sensor network based chaotic firefly algorithm cluster head selection J Pradeep, MPM Kumar Journal of Critical Reviews 7 (8), 1208-1215 , 2020 2020.0 Citations: 5
Comparison of CAN and Flexray Protocol for Automotive Application RD J.Pradeep, S. Richerd Sebasteen International Journal of Pure and Applied Mathematics, 119 (14), 1739-1745 , 2018 2018.0 Citations: 5
Efficient vehicle detection system using OCR and REST API J Pradeep, A Muthukumaran, G Nareshraman, R Krishnan, ... International Journal of Engineering Research & Technology (IJERT) 9 (7 … , 2020 2020.0 Citations: 3
Performance Analysis of Wireless Mesh Network in Hospital Environment D Prabakar, J Pradeep, GA Elango, N Lakshmy IJCA Special Issue on International Conference on Electronics, Communication … , 2012 2012.0 Citations: 3
E, Shrinivasan, S. Himavathi “Diagonal based feature extraction for handwritten character recognition system using neural network” J Pradeep IEEE , 0 Citations: 3
Automatic railway detection and tracking inspecting system J Pradeep, M Harikrishnan, K Vijayakumar Proceedings of International Conference on Data Science and Applications … , 2021 2021.0 Citations: 2
Voice Recognition Using Natural Language Processing J Pradeep, K Vijayakumar, M Harikrishnan Big Data Management in Sensing: Applications in AI and IoT , River … , 2021 2021.0 Citations: 2
Design of FlexRay communication controller protocol for an automotive application R Radhiga, J Pradeep 2015 IEEE 9th International Conference on Intelligent Systems and Control … , 2015 2015.0 Citations: 2