A deep learning model with inception vision transformer and harmonic fusion network for solar panel fault detection and classification Gangu Dharmaraju, Manoj Kumar Jena, Venkataramana Attada Computers and Electrical Engineering, 2026 A solar panel is made up of silicon-based solar cells and uses the Photovoltaic (PV) effect to convert solar energy into electrical energy. The conventional fault detection models failed to identify the exact fault types, thereby requiring additional tools for classification. Thus, an innovative Inception Vision Transformer and Harmonic fusion Network (IncepViTH-Net) approach is developed for automated solar panel fault detection and classification. At first, the solar panel images are collected from a specific dataset. Then, sourced images are filtered by pixel brightness correction methodology. Later, the affected fault regions of solar panels are segmented by U-Net architecture. Moreover, features, like Convolutional Neural Network (CNN), histogram features, Angular Second Moment (ASM), and inverse difference moment are extracted for further analysis. Then, the solar panel fault detection and classification is executed by IncepViTH-Net model, which fuses strengths of Inception Version 3 Network (InceptionV3-Net), Vision Transformer (ViT), and harmonic analysis. The outcomes achieved by IncepViTH-Net are an accuracy of 96.988 %, True Positive Rate (TPR) of 97.988 %, and True Negative Rate (TNR) of 96.577 % for K Value 8.
Advanced skin lesion diagnosis with efficientnet-b7 feature extraction and SVM classification V Manjula, Pala Pooja Ratnam, Dr Golagani Prasanna Priya, Gangu Dharmaraju, Regidi Suneetha, Nagamalli Arasavalli, Mirtipati Satish Kumar International Journal of Basic and Applied Sciences, 2025 Skin cancer is the most common form of cancer globally. Timely detection is crucial, since failure to identify it in the first stage may result in grave consequences. Notwithstanding its apparent visibility, significant intra-class heterogeneity and inter-class homogeneity complicate its identification. Current AI methodologies for detecting skin cancer are hindered by their reliance on convolutional neural networks, resulting in a lack of interpretability and sluggish processing speeds. To address the issue, the study proposes a comprehensive pipeline that integrates deep learning and machine learning methodologies to enhance detection accuracy in identifying skin lesions. The dataset under con-consideration is the International Skin Imaging Collaboration (ISIC) 2020. Initially, we pre-process the photos to guarantee precise training and categorization. The EfficientNet-B7 deep learning model is employed for feature extraction and fed into a support vector machine (SVM) classifier. The assessment of parameters such as accuracy, precision, recall, and F1 score yielded an accuracy of 97.52% and an F1 score of 98.61%. The proposed model demonstrates superior results relative to other current models.
Image Based Multi Class Fault Detection in PV Systems using Deep Learning Model Gangu Dharmaraju, Manoj Kumar Jena, Attada Venkataramana 2025 IEEE 7th International Conference on Computing Communication and Automation Iccca 2025, 2025 Solar energy is one of the most environmentally friendly power sources that is the most sustainable, utilizing solar panel energy. The fault on solar panels reduces with loss in production efficiency. This proposed methodology observes multi class fault on the solar panel using a ViT model. Unlike the convolutional neural networks (CNN), such as R2CNN, vision transformers compute interconnections using the self-attention mechanism, which are long enough to capture the extended interconnections in the input images, and hence vision transformers are efficient for the fault identification and classification in solar panels. The construction of dataset with faults including snow cover, dust accumulation, physical damage, bird droppings and electrical damage is part of our methodology. Thus, the proposed model was trained and fine-tuned on this dataset and classifies the faults better than the conventional deep learning model with an accuracy of 98 %. We also improve false negatives and false positives in the evaluation and our approach improves accuracy.
Image Based Multi Class Fault Detection in PV Systems using Deep Learning Model Gangu Dharmaraju, Manoj Kumar Jena, A V Ramana 7th IEEE International Conference on Emerging Electronics Icee 2025, 2025 Solar energy is one of the most environmentally friendly energy sources that is the most sustainable, utilizing solar panel energy. The fault on solar panels attenuates with loss in production efficiency. This proposed methodology observes multi class fault on the solar panel using a ViT model. Unlike the convolutional neural networks (CNN), such as R2CNN, vision transformers compute interdependencies using the self-attention mechanism, which are long enough to capture the extended interdependencies in the input images, and hence vision transformers are efficient for the fault detection and classification in solar panels. The construction of such a large dataset of faults including dust accumulation, physical damage, snow cover, bird droppings and electrical damage is part of our methodology. Thus, the proposed model was trained and fine-tuned on this dataset and classifies the faults better than the conventional deep learning model with an accuracy of 98 %. We also improve false positives and false negatives in the evaluation and our approach improves accuracy.
A Hybrid Approach for Malicious URL Detection Using ML Classifiers and Graph Neural Networks Bimanna Alaladinni, Gangu Dharmaraju, Vadada Yamuna, Senige Rajasekhar Reddy, A. Lakshmanarao, G. Charles Babu 2025 6th International Conference on Data Intelligence and Cognitive Informatics Icdici 2025, 2025 Cyber threats such as phishing attacks often exploit malicious URLs to deceive users and compromise sensitive information. Traditional feature-based machine learning methods, while effective, may struggle to capture the intricate relationships among URL components. In this research, a hybrid framework is proposed combining Graph Neural Networks (GNN) and Random Forest Classifier (RFC) to detect malicious URLs with high accuracy. The methodology applies extensive text preprocessing to tokenize URLs into subdomains, domains, and path elements. For graph-based learning, each URL is converted into a token graph where nodes represent URL components and edges capture relational structures. Simultaneously, handcrafted features such as URL length, HTTPS presence, and special character counts are extracted for RFC training. The GNN model learns structural dependencies through graph convolution operations, while the RFC model leverages decision tree ensembles for robust feature-based classification. Experiments on a large-scale Kaggle phishing URL dataset demonstrate that both models achieve excellent detection performance, with RFC slightly outperforming GNN in precision and speed. The hybrid framework provides a scalable and effective solution for real-time phishing detection, combining the strengths of structural graph analysis and traditional machine learning.
Phishing Website Detection through Ensemble Machine Learning Techniques Gangu Dharmaraju, Tatapudi Nirosh Kumar, P.PattabhiRama Mohan, Raja Rao Pbv, A. Lakshmanarao 2024 2nd International Conference on Computer Communication and Control Ic4 2024, 2024 Phishing attacks have become increasingly sophisticated, posing a significant threat to individuals and organizations. The ability to detect phishing websites is crucial for mitigating potential risks and safeguarding sensitive information. Traditional methods of detecting phishing websites often struggle to keep pace with the evolving tactics employed by cybercriminals. As a result, there is a pressing need for innovative and adaptive solutions to identify and combat this pervasive threat. This paper proposed an advanced phishing detection system by leveraging the power of machine learning ensemble algorithms. A dataset from Kaggle was collected. Initially, four ML classifiers namely Random /forest, Extra tree classifier, Gradient boosting classifier and logistic regression classifier applied for phishing website detection. Later ensemble of these four ML algorithms with different ensemble method is develop for phishing website detection. In ensemble, two types of ensemblesapproach namely stacking ensemble and voting ensemble applied. Experimental results showcased the potential of this ensemble approach to improve accuracy and adaptability in the prediction of phishing websites.
Exploring Deep Learning Approaches for News Classification with CNNs, RNNs and Transformers Gowripushpa Geddam, Gangu Dharmaraju, Gottala Parameswara Kumar, Mahesh Babu Ketha, A. Lakshmanarao 2024 1st International Conference on Innovations in Communications Electrical and Computer Engineering Icicec 2024, 2024 In the realm of text classification, particularly news classification, advanced deep learning models have demonstrated substantial potential for enhancing accuracy and performance. This paper investigates the application of several prominent deep learning architectures: CNNs, RNNs and Transformer-based BERT model. CNNs were employed to capture local patterns and hierarchical features within text data, contributing to effective categorization of news articles. RNNs, including LSTM networks, were utilized to model sequential dependencies and contextual relationships in news text, addressing the challenge of understanding temporal aspects. Additionally, Transformer-based models such as BERT (Bidirectional Encoder Representations from Transformers) were leveraged for their superior contextual and semantic comprehension, achieving notable performance improvements. The study provides a comprehensive analysis of these models' effectiveness, comparing their strengths and limitations based on accuracy results. BERT and its hybrid combinations, particularly with LSTM, achieved the highest accuracy. The paper highlights the most effective approaches for leveraging deep learning in text classification and offers insights into future research directions in this domain.
RECENT SCHOLAR PUBLICATIONS
An Efficient Intrusion Detection Model with Novel Machine Learning Stacking Ensemble Techniques TVS Krishna, PS Lakshmi, M Prathyusha, M Swapna, J Kanimozhi, ... Embracing the Digital Horizon: Advanced AI and Intelligent Systems for … , 2026 2026 Citations: 3
A deep learning model with inception vision transformer and harmonic fusion network for solar panel fault detection and classification G Dharmaraju, MK Jena, V Attada Computers and Electrical Engineering 129, 110819 , 2026 2026 Citations: 1
Hybrid Autoencoder and Ensemble Deep Learning Model for Stroke Prediction S Thirunagari, AVL Devi, PS Deepthi, VD Babu, VV Kalyani, ... 2025 6th International Conference on IoT Based Control Networks and … , 2025 2025
Attention-Enhanced Deep CNN Framework for Efficient Brain Tumor Classification G Tayaru, KR Kumar, NMA NM, G Prasanthi, JDS Kumar, ... 2025 6th International Conference on IoT Based Control Networks and … , 2025 2025
A Deep Multi-Modal Framework for Cyberbullying Detection using Textual and Behavioural Features on Social Media CV Ramana, V Sumathy, N Vullam, G Prasanthi, A Lakshmanarao, ... 2025 6th International Conference on IoT Based Control Networks and … , 2025 2025
Image Based Multi Class Fault Detection in PV Systems using Deep Learning Model G Dharmaraju, MK Jena, AV Ramana 2025 IEEE 7th International Conference on Emerging Electronics (ICEE), 1-6 , 2025 2025
Hybrid Deep Learning Framework for Fake News Detection using BERT Embeddings and Context-Aware Metadata Fusion KR Kumar, P Sujatha, M Sivajyothi, VD Babu, A Lakshmanarao, PV Rao 2025 International Conference on Sustainable Communication Networks and … , 2025 2025
Secure key based cloud security utilizing three-way protection with martino homomorphic encryption for preventing unauthorized data access G Ramanjaiah, TSR Kiran, A Srisaila, A Lakshmanarao, ... Swarm and Evolutionary Computation 98, 102131 , 2025 2025 Citations: 6
Enhanced Detection of Autism Spectrum Disorder using Machine Learning and Deep Neural Models with ADASYN and ENN Techniques PP Rani, K Nagaraju, KN Rao, V Srivallidevi, A Lakshmanarao, LB Devi 2025 5th International Conference on Soft Computing for Security … , 2025 2025
ADVANCED LUNG CANCER DETECTION MODEL WITH U-NET-BASED SEGMENTATION, ARI-TFMOA FEATURE OPTIMIZATION AND MACHINE LEARNING CLASSIFICATION S PATNALA, A LAKSHMANARAO, SB VADDE Journal of Theoretical and Applied Information Technology 103 (14) , 2025 2025
Enhanced Migraine Diagnosis using ADASYN-based Balancing and Cluster-Assisted Feature Augmentation GNR Devi, M Swetha, A Lakshmanarao, PA Kumar, M Kalidas, NR Babu 2025 8th International Conference on Computing Methodologies and … , 2025 2025
Email Spam Classification Using Novel Fusion of Machine Learning and Feed Forward Neural Network Approaches K Sreekala, M Venkatesh, MVR Murthy, SV Meena, S Rathula, ... Protecting and Mitigating Against Cyber Threats: Deploying Artificial … , 2025 2025
A Hybrid Approach for Malicious URL Detection Using ML Classifiers and Graph Neural Networks B Alaladinni, G Dharmaraju, V Yamuna, SR Reddy, A Lakshmanarao, ... 2025 6th International Conference on Data Intelligence and Cognitive … , 2025 2025 Citations: 1
An Optimized Capsule Network Architecture for Brain Tumor Classification with Comparative Insights from Transfer Learning Models KR Kumar, BV Varma, N Vullam, KS Rupa, A Lakshmanarao, PG Krishna 2025 6th International Conference on Data Intelligence and Cognitive … , 2025 2025
SqueezeNet-based key generation for secure device-to-device group authentication in 5G wireless networks KP Chandran, G Dharmaraju, A Misra, A Ali Wireless Networks 31 (5), 3571-3589 , 2025 2025 Citations: 2
A hybrid convolutional neural network-recurrent neural network approach for breast cancer detection through Mask R-CNN and ARI-TFMOA optimization. K Sreekala, S Yalamati, A Lakshmanarao, G Kumari, TM Kumari, ... International Journal of Electrical & Computer Engineering (2088-8708) 15 (3) , 2025 2025 Citations: 2
Optimized Transfer Learning and Machine Learning Hybrid Approach for Early Lung Cancer Diagnosis R Alladi, P Sumithabhashini, B Mopuru, VD Babu, KN Neeraja, ... 2025 Third International Conference on Augmented Intelligence and … , 2025 2025
Android malware detection through opcode sequences using deep learning LSTM and GRU networks A Lakshmanarao, JS Mantena, KK Thota, PS Chandaka, CVM Krishna, ... Indonesian Journal of Electrical Engineering and Computer Science 38 (2), 1106 , 2025 2025 Citations: 2
Food Classification Using a Hybrid Framework with Transfer Learning and Machine Learning Models B Panduri, N Vullam, RSSR Battula, VD Babu, VS Desanamukula, ... 2025 International Conference on Advances in Modern Age Technologies for … , 2025 2025 Citations: 1
Hybrid Approach Using RoBERTa and Neural Networks for Airline Twitter Sentiment Analysis R Sukrutha, K Sreedevi, LK Pithani, BA Ram, DC Mouli, A Lakshmanarao 2025 International Conference on Advances in Modern Age Technologies for … , 2025 2025
MOST CITED SCHOLAR PUBLICATIONS
Machine learning techniques for heart disease prediction A Lakshmanarao, Y Swathi, PSS Sundareswar Forest 95 (99), 97 , 2019 2019 Citations: 120
Plant disease prediction and classification using deep learning ConvNets A Lakshmanarao, MR Babu, TSR Kiran 2021 International Conference on Artificial Intelligence and Machine Vision … , 2021 2021 Citations: 112
SMS spam detection using machine learning and deep learning techniques S Gadde, A Lakshmanarao, S Satyanarayana 2021 7th international conference on advanced computing and communication … , 2021 2021 Citations: 108
Heart disease prediction using feature selection and ensemble learning techniques A Lakshmanarao, A Srisaila, TSR Kiran 2021 Third International Conference on Intelligent Communication … , 2021 2021 Citations: 72
Phishing website detection using novel machine learning fusion approach A Lakshmanarao, PSP Rao, MMB Krishna 2021 international conference on artificial intelligence and smart systems … , 2021 2021 Citations: 59
Malicious URL detection using NLP, machine learning and FLASK A Lakshmanarao, MR Babu, MMB Krishna 2021 international conference on innovative computing, intelligent … , 2021 2021 Citations: 54
Android Malware Detection with Deep Learning using RNN from Opcode Sequences. A Lakshmanarao, M Shashi International Journal of Interactive Mobile Technologies 16 (1) , 2022 2022 Citations: 37
An effecient fake news detection system using machine learning A Lakshmanarao, Y Swathi, TSR Kiran International Journal of Innovative Technology and Exploring Engineering 8 … , 2019 2019 Citations: 37
Enhancing Cloud Data Privacy with a Scalable Hybrid Approach: HE-DPSMC. J Singh, AM Reddy, V Bande, A Lakshmanarao, GS Rao, K Samunnisa journal of electrical systems 19 (4) , 2023 2023 Citations: 20
An efficient Android malware detection model using ConvNets and resnet models A Lakshmanarao, PB Madhuri, K Dasari, KA Babu, SR Sulthana 2024 International Conference on Intelligent Algorithms for Computational … , 2024 2024 Citations: 19
Phishing website detection through ensemble machine learning techniques G Dharmaraju, TN Kumar, PPR Mohan, RR Pbv, A Lakshmanarao 2024 2nd International Conference on Computer, Communication and Control … , 2024 2024 Citations: 18
A survey on machine learning for cyber security A Lakshmanarao, M Shashi International Journal of Scientific & Technology Research 9 (01), 499-502 , 2020 2020 Citations: 18
Plant disease prediction using transfer learning techniques A Lakshmanarao, N Supriya, A Arulmurugan 2022 Second International Conference on Advances in Electrical, Computing … , 2022 2022 Citations: 17
A hybrid Kmeans and ML classification approach for credit card fraud detection K Sreekala, R Sridivya, NKK Rao, RK Mandal, GJ Moses, ... 2024 3rd International Conference for Innovation in Technology (INOCON), 1-5 , 2024 2024 Citations: 15
Machine learning and deep learning framework with feature selection for intrusion detection A Lakshmanarao, A Srisaila, TSR Kiran 2022 International Conference on Communication, Computing and Internet of … , 2022 2022 Citations: 15
An Efficient Deep Learning framework with CNN and RBM for Native Speech to Text Translation S Neelima, K Dasari, A Lakshmanarao, PJ Rao, MK Jetty 2024 3rd International Conference for Advancement in Technology (ICONAT), 1-6 , 2024 2024 Citations: 14
Machine learning approach for diabetes prediction using genetic algorithm based feature selection TSR Kiran, A Srisaila, GS Shankar, B Sowjanya, A Lakshmanarao 2024 3rd International Conference for Innovation in Technology (INOCON), 1-5 , 2024 2024 Citations: 14
Advancing heart disease detection by using ensemble meta-features integration K Hymavathi, SH Mehanoor, G Srinivas, A Lakshmanarao, J Jalla 2024 IEEE International Conference on Computing, Power and Communication … , 2024 2024 Citations: 14
Fake news detection using ML and DL approaches G Srinivas, A Lakshmanarao, S Sushma, MV Krishna, S Neelima 2023 International Conference on Circuit Power and Computing Technologies … , 2023 2023 Citations: 14
SMS spam detection using NLP and deep learning recurrent neural network variants VV Kalyani, MVR Sundari, S Neelima, PSS Prasad, PPR Mohan, ... 2024 International Conference on Cognitive Robotics and Intelligent Systems … , 2024 2024 Citations: 12
