About
Dr. Ramakrishna Hegde is a distinguished academician, researcher, and academic leader currently serving as Professor and Associate Dean, School of Engineering, RV University, Mysuru. He previously served as Professor in the Department of Computer Science and Engineering at SJCE, JSS Science and Technology University, Mysuru. With extensive experience in engineering education, he has made significant contributions to teaching, research, curriculum development, academic administration, and institutional growth.
His research interests include Cyber Security, Artificial Intelligence, Internet of Things (IoT), Machine Learning, Deep Learning, Quantum Computing, Smart Systems, and Computer Science & Engineering. Dr. Hegde has published over 44 research papers in reputed national and international journals and conferences, authored two books and two book chapters, been granted one patent, published 15 patents and is currently supervising three research scholars. He actively promotes collaborative research and industry-academia partnerships.
An accomplished educator, Dr. Hegde has played a key role in developing outcome-based curricula, adopting innovative teaching practices, and mentoring undergraduate, postgraduate, and doctoral students. He has successfully organized international conferences, faculty development programmes, workshops, hackathons, expert lectures, and industry interactions that foster innovation and experiential learning.
Dr. Hegde is an active IEEE volunteer and professional leader. He has served as an organizer, program coordinator, session chair, track chair, and technical committee member for several IEEE international conferences. He currently serves as Chair of the IEEE Systems Council Bangalore Chapter (since 2025) and previously served as Secretary of the IEEE Computer Society Bangalore Chapter (2021, 2022).
A passionate advocate of emerging technologies, Dr. Hegde has delivered keynote addresses, invited talks, and technical sessions in India and internationally, including Singapore, Edinburgh, and STEKOM University, Indonesia. At RV University, he continues to advance academic excellence by strengthening research, promoting interdisciplinary collaboration, and preparing industry-ready graduates for the digital future.
Enhanced non-contrast computed tomography images for early acute stroke detection using machine learning approach.
A machine learning-based approach was developed for early acute stroke detection using enhanced non-contrast CT images. Hybrid preprocessing, active contour segmentation, LBP, Gabor, and DWT features were employed. Dingo optimization selected optimal features, while HBO optimized parameters. The proposed XGBoost model achieved 97% accuracy and 1.5% FPR, outperforming existing models.
Accident detection using Automotive Smart Black-Box based Monitoring system. The Automotive Monitoring using
The proposed black-box system authenticates drivers and continuously monitors vehicle parameters, including speed, temperature, humidity, gas levels, location, and accidents. Data is stored locally and in the cloud for live monitoring, accident investigation, and insurance claims. Automatic alerts enable rapid medical assistance. The system demonstrates higher accuracy than RFID, SVM, CNN, and RNN methods.
Self-diagnosis platform via IOT-based privacy preserving medical data.
The proposed Self-Diagnosis Platform (SDP) enhances IoT healthcare security using EdDSA encryption to protect patient data and preserve identity privacy. Experimental results demonstrate superior performance over LDQN, SE-AC, PMDA, and EPPDA, achieving significantly lower execution time and end-to-end delay. SDP offers an effective, secure, privacy-preserving, and versatile solution for real-world healthcare applications.
Advancements in Climate Modeling and Prediction through Machine Learning Techniques.
Machine learning is increasingly transforming weather and climate modelling, improving resolution, prediction, parameterization, and complete climate models, with recent advances emphasizing critical evaluation of ML applications.
A Method for Predicting Cardiovascular Disorder using Machine Learning Techniques.
The proposed system predicts cardiovascular disease risk using patients’ live health and lifestyle data. Among Random Forest, Linear Regression, KNN, and XGBoost, XGBoost achieved highest accuracy with minimal Type-I error.
A System to Track Fabricated Content in Social media using Machine Learning Approach.
The model uses web-extracted fictitious datasets to detect fraudulent website information. It applies Random Forest, Logistic Regression, Decision Tree, KNN, and Gradient Boosting classifiers for accurate classification.
An Hand Gesture Smart Translation System for Differently Abled People.
This research objective is to develop a deaf communication system.
There are two sections to this system. (1)Sign language conversion from the audio messages (2) Converting pictures and videos into speech or text. In the first section we will examine, audio is taken as input, the recorded spoken message is converted into text, and Hindi Sign language graphics or GIFs are displayed. The hearing and the deaf can communicate more easily when this method is used. In the second group, we gather the photographs, use CNN to train them, and display the outcomes.
Intrusion Detection System Using Machine Learning Based on NSL KDD Dataset.
IThis study evaluates machine-learning techniques for intrusion detection using the NSL-KDD dataset. Results indicate Random Forest achieved the highest accuracy, while additional algorithms could further improve IDS performance.
An Efficient Method for Credit Card Fraud Detection using Machine Learning
The study evaluates hybrid machine-learning models for credit card fraud detection. Among 66 models, KNN-CatBoost achieved superior AUC, recall, and F1-score, outperforming conventional approaches and demonstrating strong detection performance.
Face Recognition-Based Video Surveillance and Tracking System
In this research, the proposed a system focuses on video camera face identification and recognition, surveillance systems that play a major role in situational control. These technologies convert video surveillance from a data-gathering instrument
to a system for gathering intelligence and information. Surveillance systems can respond to an activity in real time and obtain pertinent information at a much greater resolution by using real-time video analysis.
Code-Ids: Convolutional Neural Network Based Intrusion Detection System Using Deep Learning
This paper proposes a novel CODE-IDS framework using deep learning Network to improve network security by precise cyber threat identification and mitigation. While allowing secure access to attacker, the system records the actions of possible attackers. The technique successfully differentiates between typical, DDoS, MiTM, and probe attack traffic by using Deep learning. PCA is used for feature extraction and the Adaptive Weighted Particle Swarm Optimization is applied to retain the most important classification for feature selection. The parameters, enhancing its accuracy and performance the overall accuracy of the suggested model is 98.78% and methods achieving a low accuracy of 93.85% 96.8% and 96.8% respectively.
ACERNET: Automated Cervical Cancer Detection System Via Hybrid Resnet And Random Forest
The proposed ACERNET model enables automated cervical cancer detection using deep learning. Cervicography images are preprocessed with Gaussian filters, followed by ResNet50-based feature extraction and Random Forest classification of adenocarcinoma and squamous cell carcinoma. Achieving 98.88% accuracy, ACERNET outperforms CerviFormer, CYENET, and C3Net, demonstrating improved reliability for early diagnosis.
Prediction of efficient driving mode for vehicles using machine learning
This study automates driving-mode selection in autonomous vehicles using real-time camera data and machine learning. CNN, MobileNetV2-based CNN, and decision trees are evaluated for adaptive road-condition recognition.
Detection of mental health issues and solution to mental toxicity using machine learning
Machine learning can support early detection of mental health concerns by analyzing social media and online content for patterns associated with depression, anxiety, and stress. It can also identify toxic content, including hate speech and cyberbullying, enabling real-time moderation. These applications promote mental wellbeing, reduce online toxicity, and create safer digital environments.
Social Engineering Threat: Phishing Detection using Machine Learning Approach
This paper explores machine learning approaches for detecting social engineering and phishing attacks. Phishing deceives users through emails, messages, or social media to steal sensitive information or install malware. The study evaluates various machine learning techniques, focusing on data preprocessing, feature extraction, model development, and URL-based phishing detection to improve cybersecurity and protection.
Real Time Patient Monitoring System Using BLYNK
The proposed IoT-based health monitoring system integrates five sensors—pulse, body temperature, MQ-2, MQ-135, and room temperature—to monitor patients and their environment. ESP32 processes real-time sensor data and transmits results via Wi-Fi to the Blynk Android application. The system provides affordable, efficient, and remote health monitoring for patients, doctors, and caregivers.
Decentralized Blockchain-Based Infrastructure for Numerous IoT Setup
This study proposes a modular blockchain architecture for scalable IoT device reconfiguration. REST-based publish/subscribe communication and smart contracts enable efficient upgrades, demonstrating reliable performance with minimal resource overhead.
Smart Translation for Physically Challenged People Using Machine Learning
This research work develops a two-way sign language communication system, converting audio into Indian Sign Language visuals and translating sign images into text or speech using CNN-based image recognition.
A survey on Development Smart Healthcare Monitoring System in IoT Environment
IoT-based healthcare monitoring systems provide convenient and efficient remote health monitoring using sensors to track patients’ vital signs and daily health activities. Collected data can be stored and transmitted through the internet to doctors, nurses, and caregivers. This paper reviews various smart healthcare technologies and approaches for continuous patient health monitoring.
A Review on Data Mining and Machine Learning Methods for Student Scholarship Prediction
This review explores machine learning and data mining techniques for scholarship prediction. Naïve Bayes, Decision Tree, and k-NN identify deserving candidates, comparing algorithm accuracy for reliable scholarship decisions.
Vulnerability Management using Machine Learning Techniques
The paper investigates machine learning techniques for identifying cybersecurity threats, attack vectors, and software vulnerabilities. It proposes the Cybersecurity Vulnerability Ontology (CVO) for structured vulnerability representation and a Cyber Intelligence Alert (CIA) framework for generating advanced security alerts. The study evaluates their accuracy, performance, and usefulness in strengthening Web Application Firewalls and cybersecurity.
An Efficient Method for Credit Card Fraud Detection using Machine Learning
This study investigates supervised machine-learning techniques for credit card fraud detection, addressing imbalanced and evolving fraud patterns. SVM, XGBoost, and AdaBoost are evaluated for effective and adaptable detection.
English Text to Indian Sign Language Translation systems
The proposed system translates English sentences into Indian Sign Language (ISL) using parsing, sentence reordering, word elimination, and lemmatization. Each processed word is mapped to corresponding ISL sign videos, which are concatenated for sequential display. The system facilitates communication for hearing-impaired individuals and serves as an effective educational aid.
Novel Technique for Securing IoT Systems by using Multiple ECC and Ceaser Cipher Cryptography
The paper examines cryptographic techniques for securing IoT systems, including AES, Diffie-Hellman, and Elliptic Curve Cryptography. It highlights challenges in implementing these methods on resource-constrained devices and proposes a novel Caesar cipher for lightweight encryption. The study emphasizes developing flexible cryptographic suites combining encryption and authentication to enhance IoT security.
A Secrete Hiding Method with Edge Based Classifier for Image Steganography using LSB with Enhanced Ceaser Cipher Cryptography
This work proposes a fast video steganalysis method using improved Canny edge detection and Fuzzy Random Forest classification to detect LSB-based hidden messages, achieving secure and efficient embedding performance.
An Optimal Modified Matrix Encoding Technique For Secret Writing In Mpeg Video Using Ecc
This work proposes a secure video steganography technique combining Multi-curve ECC encryption with Optimized Modified Matrix Encoding (OMME). Artificial Bee Colony optimization selects pixels to minimize distortion while embedding encrypted data into H.264 video. The approach improves security, robustness, embedding efficiency, and carrier capacity against attacks compared with existing methods.
Design And Implementation Of Image Steganography By Using Lsb Replacement Algorithm And Pseudo Random Encoding Technique
This paper proposes secure image steganography using Least Significant Bit (LSB) and Pseudo Random Encoding techniques to conceal secret data within a cover image. The embedded image remains visually indistinguishable from the original. During retrieval, reverse techniques extract the hidden information. PSNR is evaluated to compare the performance and effectiveness of both methods.
Conceptual Design Of Edge Adaptive Steganography Scheme Based On Advanced Lsb Algorithm
This work proposes adaptive edge-based image steganography using LSBMR, selecting embedding regions according to secret-message size. The approach improves embedding capacity while preserving high visual quality of stego images.
Design and Implementation of Data Hiding Technique by Using MPEG Video with constant bit rate
Block chain Integrated Smart City Governance Models for Secure Transparent and Decentralized Decision Making
Blockchain can transform smart city governance by enabling secure, transparent, and decentralized decision-making. Smart contracts and distributed ledgers improve accountability, automate public services, support real-time resource distribution, and encourage citizen participation. Applications include energy trading, waste management, service automation, and participatory budgeting. Key challenges involve citizen engagement, legal compliance, and technological integration.
Predictive Analytics Using AI for Smart Policing Crime Prevention and Public Safety Management in Urban Areas.
AI and IoT are transforming smart policing through predictive analytics, real-time surveillance, automated decision-making, and optimized resource allocation. These technologies enhance crime prevention, public safety, and inter-jurisdictional collaboration. However, challenges include privacy, algorithmic bias, surveillance, and civil liberties. Effective data integration, collaboration, ethical standards, and regulatory frameworks are essential for equitable implementation.
Machine Learning Based Scholarship and Credit pre-Assessment System
The study proposes a Naïve Bayes classifier to predict students’ scholarship eligibility based on academic performance, communication skills, income, regularity, time management, and other factors. The system helps identify scholarship potential early and supports student development. Experimental results achieved 96.7% accuracy with only 3.3% error, providing reliable scholarship prediction.
Acute Stroke Detection: Machine learning is used with enhanced CT images to improve early acute stroke detection.
Automotive Accident Detection: A smart black-box system monitors vehicle parameters and automatically detects accidents and locations.
Privacy-Preserving Healthcare: An IoT-based self-diagnosis platform uses EdDSA encryption to protect medical data and improve privacy.
Climate Prediction: Machine learning techniques are increasingly being applied to weather and climate modelling and prediction.
Cardiovascular Disease Prediction: ML classifiers such as Random Forest, Linear Regression, KNN and XGBoost are applied to predict cardiovascular disorders.
Social Media Fake Content Detection: ML models including Random Forest, Logistic Regression, Decision Trees, KNN and Gradient Boosting help identify fabricated online content.
Sign Language Translation: AI-based systems translate speech into sign language and images/videos into speech or text to support differently abled people.
Intrusion Detection: Machine learning is applied to cybersecurity intrusion detection, with Random Forest showing strong performance on the NSL-KDD dataset.
Credit Card Fraud Detection: Supervised ML techniques such as XGBoost, SVM and AdaBoost are used to detect fraudulent transactions.
Face Recognition and Surveillance: Real-time video analysis and facial recognition can support intelligent surveillance and situational monitoring.
Deep Learning for Cybersecurity: CODE-IDS uses CNN-based deep learning, PCA and optimization techniques for detecting cyberattacks.
Cervical Cancer Detection: ACERNET combines ResNet50 and Random Forest to classify cervical cancer from medical images.
Intelligent Driving: ML and image recognition can automatically select suitable driving modes based on real-time vehicle and camera data.
Mental Health and Online Toxicity: ML can analyze online content to identify indicators of mental health issues and detect toxic content such as cyberbullying.
Phishing Detection: Machine learning is used to identify phishing websites, URLs and social-engineering attacks.
Real-Time Patient Monitoring: IoT sensors connected to an ESP32 and Blynk application enable remote monitoring of vital and environmental parameters.
Blockchain for IoT: Blockchain can support secure, modular and scalable management of large numbers of IoT devices through smart contracts and publish/subscribe mechanisms.
Smart Healthcare: IoT healthcare systems use sensors to collect, store and transmit patients' vital information to healthcare professionals.
Steganography and Cryptography: Several approaches combine LSB techniques, ECC, Caesar cipher, optimization and video/image steganography for secure data hiding.
AI, Blockchain and Smart Governance: AI, IoT and blockchain are explored for smart-city governance, predictive policing, public safety, transparent decision-making and scholarship prediction
Research Interests
Artificial Intelligence and Machine Learning
Deep Learning and Computer Vision
AI for Healthcare and Medical Informatics
Cybersecurity and Machine Learning
Cybersecurity, Cryptography and Steganography
Internet of Things (IoT) and IoT Security
Blockchain and Decentralized Systems
AI and IoT for Smart Cities
Intelligent Transportation and Automotive Systems
Natural Language Processing and Assistive AI
Indian Sign Language and AI-Based Assistive Technologies
Financial Fraud Detection and Predictive Analytics
AI for Education and Student Analytics
AI for Climate and Environmental Prediction
Quantum Computing
Resource Person at the IEEE Global System Council Summit 2025, Singapore.
Served as IEEE International Conference Organising Chair – AICDMB 2025, VVCE, Mysuru.
Received the prestigious “Shikshaka Ratna” Award at Vishwa Havyaka Sammelana 2024.
Elected to an IEEE leadership position.
Elected as Chair – IEEE System Council Bangalore Section Chapter
Received Best Research Paper Award at IEEE ICRASET 2024 International Conference.
Served as Resource Person at DRDO, Ministry of Defence, Government of India.
Received felicitation from the Director, DRDO-DIFL, Ministry of Defence, Government of India.
Participated and contributed in the launch of Karnataka Cyber Security Policy 2024 at Vidhana Soudha/Vikasa Soudha, Bengaluru.
Received Best Project Award at KSCST, VTU Belagavi.
Associated with the IEEE CS Outstanding Student Branch Award.
Awarded as Best Paper at International Conference in Edinburgh, United Kingdom.
Delivered an awareness program broadcast on All India Radio.
Panel Discussion member on Cyber Security at IIT Madras.
Awareness on Cyber Security in Suvarna News.
Participated in Cyber Edge 2024 at IIT Madras.
Served as Session Chair at ICDDS International Conference, BMS College of Engineering, Bengaluru.
Served as Session Chair at the 3rd IEEE MysuruCon 2023 International Conference, MCE Hassan.
Served as EXECOM Member, IEEE Computer Society Bangalore Chapter – 2023.
Received an Award for Best Performing Faculty during the year 2014
Delivered programs on Cyber Security and Career guidance in All India Radio.
Contributed to the PMKVY Government of India Skill Development Programme
Received a Best Student Project Award @SDMIT, Ujire
Mentor for National level Hackathon Winners.
Certified as a Cyber Crime Intervention Officer by National Security Database.
Recognized as Senior Member of IEEE.
Served as Secretary, IEEE Computer Society Bangalore Chapter.
Served as Session Chair at ETCST International Conference, VTU Mysuru.
Delivered training programmes at MIRA Management School, Mysuru.
Successfully reached the World’s Highest Motorable Road in the Himalayas.
Featured in/recorded a programme at Mysore Akashavani Kendra.
Patents (16 patents published and 1 patent granted).
Book publications : 2 books published
Regularly contributing articles to local newspapers.
Successfully completed more than 5 trekking (10 days or more) in Uttarakand, Jammu Kashmir and Ladak.
Professor and Associate Dean
School of Engineering
-
Mysuru