Local Extrema Gabor Patterns: A Hybrid Approach for Enhanced Palmprint Recognition Shefali Sharma, Gaurav Saxena, Anurag Mahajan Recent Advances in Electrical and Electronic Engineering, 2026 Introduction: Palmprint recognition is an emerging field within biometric-based personal identification, offering greater security than traditional ID cards or password-based systems. While biometric methods using traits like fingerprints and facial features are well-studied, palmprint identification remains relatively underexplored despite its strong potential. This paper introduces a novel feature descriptor, Local Extrema Gabor Patterns (LEGP), for palmprint recognition. Methods: The proposed method combines the Gabor Transform (GT) with local extrema, effectively capturing the directional extrema relationships between a reference pixel and its surrounding neighbors. Results: The results demonstrate that the LEGP outperforms state-of-the-art techniques like LBP, LDP, LTP, and their Gabor-enhanced variants (GLBP, GLDP, GLTP), achieving a significantly higher recognition rate (RR). Discussion: Existing approaches primarily use Gabor Transform (GT) features from the transform domain and Local Binary Patterns (LBP) from the spatial domain. In contrast, our proposed method combines both domains by introducing LEGP, a new spatial feature. While LBP captures pixel relationships, LEGP extracts directional extrema from Gabor transform responses. Conclusion: The proposed Local Extrema Gabor Patterns (LEGP) descriptor enhances palmprint recognition by integrating the Gabor Transform with local extrema analysis to capture directional features. Evaluated on the HKPU Palmprint database, LEGP outperforms existing pattern-based methods in terms of recognition rate, achieving 99.53% accuracy with a processing time of 76.2 ms.
MOR Framework: IoT-Driven Dynamic Decision- Making for Fleet Optimization and Maintenance Prioritization Preeti Sharma, Meenakshi Malik, Chander Prabha, Durgesh Nandan, Snehal Bhosale, Anurag Mahajan Recent Patents on Engineering, 2026 Introduction: Vehicular Internet of Things–Wireless Sensor Networks (IoT-WSNs) employ interconnected sensors such as GPS, accelerometers, cameras, and temperature monitors to enhance safety, performance, and efficiency. These networks collect data on speed, engine health, tire pressure, and road conditions, enabling real-time analytics for predictive maintenance and routing. However, most existing models are vulnerable to noise and outliers, limiting decision accuracy. Methods: This study introduces the Middle-Order Ranking (MOR) framework, which integrates IoT sensor data to optimize vehicle routing and insurance settlement. Using data from 50 simulated vehicles, the model calculates Feature Ranking Scores (FRS) through the Median of Sensor readings (MOS) and Mean Absolute Deviation (MAD). The Middle-Order Routing Score (MORS) is then derived to allocate vehicles to routes across five complexity levels. Simulations were implemented in Python 3.x with NumPy, Pandas, and Matplotlib libraries. Results: Vehicles with MORS ≥0.70 were consistently assigned to high-complexity routes such as highways or emergency response, while vehicles with MORS ≤0.50 were restricted to low-priority routes or excluded. Brake usage and engine temperature emerged as the most reliable features, achieving the highest FRS values. Discussion: The MOR framework enhances robustness by minimizing the influence of sensor anomalies and noise. Compared to conventional max/min methods, MOS–MAD pairing ensures stable, interpretable, and data-driven vehicle prioritization suitable for insurance and fleet operations. conclusion: Conclusion: Through the optimal prioritization of features for enhanced vehicle selection and risk management efficiency, the MOR model was able to make accurate routing decisions over 10 simulation rounds. MOR exhibits the potential to develop into a productive data-based systems for automated insurance management and future vehicle systems. Conclusion: MOR demonstrates strong potential for intelligent fleet management, predictive maintenance, and automated insurance systems. Future work will extend this model to real-world datasets for greater applicability.
Person identification using novel local triangular binary pattern-based texture descriptor Arti Tekade, T. Vijayan, B. Karthik, Anurag Mahajan Eurasip Journal on Advances in Signal Processing, 2025 Human authentication is a crucial part of most computer vision automation systems. Conventional fingerprint, iris, face, or palm print-based systems cannot identify individuals when their external biometric components are destroyed, such as by severe burns, rashes, or wounds. The main elements of any person authentication system are non-forgery, security, resilience, and privacy. The local texture descriptor is vital in describing hand radiographic images' texture. This paper presents the novel local triangular binary pattern based texture descriptor to provide a local texture description of the hand radiographic images. The performance of the proposed descriptor is assessed using different machine learning classifiers such as K-nearest neighbor (KNN), support vector machine (SVM), radial basis function-SVM (RBF-SVM), classification tree (CT), and random forest (RF) for authentication of the 20 users based on hand radiographs. The suggested system provides an overall accuracy of 84.17% for KNN, 90% for SVM, 91.35% for RBF-SVM, 92.50% for CT, and 96.67% for RF for the 20 users for the In-house hand radiographic dataset.
A multi-class SVM based CBIR system using Forest Optimization algorithm and Firefly algorithm Kishore Dannina, Durgesh Nandan, K Meenakshi, Anurag Mahajan Engineering Research Express, 2025 Content based image retrieval (CBIR) system extracts the images that are similar to the query image (QI) from database. In this work, a new CBIR system is introduced using feature extraction techniques, Forest Optimization Algorithm, Multi-Class Support Vector Machine (MC-SVM) and Firefly Algorithm. In this method, HSV colour model for colour feature extraction, CS-SCHT for texture feature extraction and Modified Exponent Fourier Moments for shape feature extraction are used for extracting the features which are related to query and images in the database. Forest Optimization Algorithm (FOA) is used for feature subset selection. Later, MC-SVM is used to classify the selected features into different classes. Lastly, the classification accuracy improved by using firefly algorithm (FA). The proposed method performance is evaluated on datasets Corel-1k, Corel-5k, Corel-10k, GHIM-10k, and CALTECH 101 databases and the results emphasise the proposed model has shown better results compared to the existing works.
Smart Grid Data Compression and Reconstruction by Discrete Wavelet Transform and A. Mahajan International Journal of Electrical and Electronic Engineering and Telecommunications, 2025 A smart grid offers safe, reliable, and useful electricity. Phasor measuring units, smart meters, and other monitoring and measurement devices in the smart grid keep track of statuses at every grid level. As a result, utilities, control centers, and customers must exchange and retain an enormous amount of data in real time. As a result, data storage and communication require efficient data compression. To accurately reflect status of the system and regenerate nearly flawlessly on receiving side, compression should preserve all of the data’s critical information. This work uses the discrete wavelet transform to recover compressed voltage sag signals. With reduced data, the disruptions can be communicated more quickly. Because the proposed system uses lesser filters and few decomposition layers, it is simpler than the previous design. The results of the simulation demonstrate improvement in the reconstruction error and compression ratio by minimizing their values. This design is time-saving and simple to use.
Hybrid MNLTP Texture Descriptor and PDCNN-Based OCT Image Classification for Retinal Disease Detection Anurag Mahajan, Jahida Subhedar, Shabana Urooj, Neeraj Kumar Shukla Computers Materials and Continua, 2025 : Retinal Optical Coherence Tomography (OCT) images, a non-invasive imaging technique, have become a standard retinal disease detection tool. Due to disease, there are morphological and textural changes in the layers of the retina. Classifying OCT images is challenging, as the morphological manifestations of different diseases may be similar. The OCT images capture the reflectivity characteristics of the retinal tissues. Retinal diseases change the reflectivity property of retinal tissues, resulting in texture variations in OCT images. We propose a hybrid approach to OCT image classification in which the Convolution Neural Network (CNN) model is trained using Multiple Neighborhood Local Ternary Pattern (MNLTP) texture descriptors of the OCT images dataset for a robust disease prediction system. Parallel deep CNN (PDCNN) is proposed to improve feature representation and generalizability. The MNLTP-PDCNN model is tested on two publicly available datasets. The parameter values Accuracy, Precision, Recall, and F1-Score are calculated. The best accuracy obtained specifying the model’s overall performance is 93.98% and 99% for the NEH and OCT2017 datasets, respectively. With the proposed architecture, comparable performance is obtained with a subset of the original OCT2017 data set and a comparatively smaller number of trainable parameters (1.6 million, 1.8 million, and 2.3 million for a single CNN branch, two parallel CNN branches, and three parallel network branches, respectively), compared to off-the-shelf CNN models. Hence, the proposed approach is suitable for real-time OCT image classification systems with fast training of the CNN model and reduced memory requirement for computations.
Power Efficient Counter Design using CNTFET with AI Integration Imran Ahmed Khan, Owais Ahmad Shah, Durgesh Nandan, Amrita Rai, Anurag Mahajan Recent Advances in Electrical and Electronic Engineering, 2025 Background: Reducing power consumption in digital circuits can be achieved by minimizing the number of transitions, and Gray code provides a binary numeral system optimized for this purpose. Traditional CMOS-based counters face limitations in power efficiency and performance at nanoscale levels. This research presents a novel design of a Gray code counter utilizing Carbon Nanotube Field-Effect Transistors (CNTFETs) as a high-performance alternative to CMOS technology. Methods: The CNTFET-based Gray code counter was evaluated across a range of temperatures (25°C to 100°C), input voltages (0.7V to 1.3V), and clock frequencies (200 MHz to 800 MHz). Supervised machine learning was employed to predict and analyze key performance metrics, including propagation delay, power consumption, and Power-Delay Product (PDP), for both CMOS and CNTFET Gray code counters under varying conditions. Results: The results demonstrate that the CNTFET-based Gray code counter exhibits significantly lower power dissipation, faster operation, and a minimum PDP compared to its CMOS counterpart across the tested temperature, voltage, and frequency variations. The machine learning predictions aligned closely with simulation results, confirming the accuracy of this approach in optimizing the design. Conclusion: The study validates the CNTFET Gray code counter as a highly efficient, low-power solution suited for high-performance applications. Its superior performance characteristics suggest that CNTFET technology, coupled with AI-driven optimization, holds promise for advanced lowpower VLSI circuit designs.
Smart grid data compression and reconstruction by wavelet packet transform Rakhi Jadhav, Anurag Mahajan Methodsx, 2024 A smart grid is a power network from generation to consumers and provides beneficial, steady, and safe electricity. It utilizes smart meters for billing, phasor measurement units to check the system's health., etc. As a result, it contains enormous volumes of real-time data that may be shared and stored by users, control centers, and services in a smart grid. It weakens the smart grid's communication networks. The size of the data in a smart grid will grow extremely in the future. As a result, it must reduce distortion in data compression and denoise while minimizing the demand on storage and communication networks. The goal of data compression and denoising should be to maximally conserve the useful data while accurately reflecting the state of the system and providing sufficient data regeneration at the receiving end. This paper has used lower-order different wavelets to represent a design to compress and reconstruct data at level three using wavelet Packet Transform. It works on the phasor measurement unit's current magnitude and voltage sag signals.•The proposed design has a better compression ratio.•Low reconstruction error.•This design is easy to access, systematic, profitable, and not time-consuming.
Business Analytics for farmers in Crop Yield Prediction Naman Chaudhary, Gyanendra Pratap Singh, Vaibhav Dagula, Anurag Mahajan Proceedings of 2023 IEEE Technology and Engineering Management Conference Asia Pacific Temscon Aspac 2023, 2023
A Brief Bibliometric Survey on Flexible and Wearable Microstrip Patch Antennas Library Philosophy and Practice, 2021
Liver Segmentation and Liver Cancer Detection Based on Deep Convolutional Neural Network: A Brief Bibliometric Survey Library Philosophy and Practice, 2021
Night Vision Bot using Dynamic IR and Object Detection Devesh Abhyankar, Gurumoorty Suresh, Hrithik Sambhaji Karjule, Parth Bhardwaj, Harish Muleva, Anurag Mahajan 2021 5th International Conference on Electrical Electronics Communication Computer Technologies and Optimization Techniques Iceeccot 2021 Proceedings, 2021
A Brief Bibliometric Survey on Night Vision Bot using Dynamic IR and Object Detection Library Philosophy and Practice, 2021
Local Extrema Gabor Patterns: A Hybrid Approach for Enhanced Palmprint Recognition S Sharma, G Saxena, A Mahajan Recent Advances in Electrical & Electronic Engineering 19 (1), E23520965407832 , 2026 2026
Power efficient counter design using CNTFET with ai integration IA Khan, OA Shah, D Nandan, A Rai, A Mahajan Recent Advances in Electrical & Electronic Engineering 18 (10), 2069-2082 , 2025 2025 Citations: 3
A multi-class SVM based CBIR system using Forest Optimization algorithm and Firefly algorithm K Dannina, D Nandan, K Meenakshi, A Mahajan Engineering Research Express 7 (3), 035239 , 2025 2025 Citations: 1
Autism spectrum disorder detection using parallel DCNN with improved teaching learning optimization feature selection scheme T Dhamale, S Bhandari, V Harpale, P Sakhi, K Napte, A Mahajan SAIEE Africa Research Journal 116 (3), 89-100 , 2025 2025 Citations: 5
Person identification using novel local triangular binary pattern-based texture descriptor A Tekade, T Vijayan, B Karthik, A Mahajan EURASIP Journal on Advances in Signal Processing 2025 (1), 7 , 2025 2025 Citations: 2
Hybrid MNLTP Texture Descriptor and PDCNN-Based OCT Image Classification for Retinal Disease Detection J Subhedar, A Mahajan, S Urooj, NK Shukla Computers, Materials & Continua 82 (2), 2831-2847 , 2025 2025 Citations: 4
Smart grid data compression and reconstruction by wavelet packet transform R Jadhav, A Mahajan MethodsX 13, 102872 , 2024 2024 Citations: 6
Behavioural Analysis of Hatchetman Attacker for Detection & Prevention in Low Power and Lossy IoT Network. G Soni, K Chandravanshi, AS Kaurav, A Saxena, D Nandan, A Mahajan Instrumentation, Mesures, Métrologies 23 (5) , 2024 2024 Citations: 6
Strip-radiator and reflector based multi-layered CPW-fed antenna for tracking application TN Pawase, A Mahajan, A Malhotra Progress In Electromagnetics Research C 146, 163-175 , 2024 2024 Citations: 1
An error-efficient Gaussian filter for image processing by using the expanded operand decomposition logarithm multiplication D Nandan, J Kanungo, A Mahajan Journal of ambient intelligence and humanized computing 15 (1), 1045-1052 , 2024 2024 Citations: 73
Compact hybrid EBG microstrip antenna for wearable applications T Pawase, A Malhotra, A Mahajan Frequenz 77 (11-12), 557-566 , 2023 2023 Citations: 5
Automatic liver cancer detection using deep convolution neural network KM Napte, A Mahajan, S Urooj IEEE Access 11, 94852-94862 , 2023 2023 Citations: 21
Smart Grid Data Denoising and Compression Using Wavelet Packet Transform. R Jadhav, A Mahajan Mathematical Modelling of Engineering Problems 10 (4) , 2023 2023 Citations: 1
Liver segmentation using marker controlled watershed transform. KM Napte, A Mahajan International Journal of Electrical & Computer Engineering (2088-8708) 13 (2) , 2023 2023 Citations: 7
A review on recent work on oct image classification for disease detection J Subhedar, A Mahajan 2022 OPJU International Technology Conference on Emerging Technologies for … , 2023 2023 Citations: 11
ESP-UNet: Encoder-decoder convolutional neural network with edge-enhanced features for liver segmentation K Napte, A Mahajan, S Urooj Traitement du Signal 40 (5), 2275-2281 , 2023 2023 Citations: 4
Deep learning based liver segmentation: A review K Napte, A Mahajan Revue d'Intelligence Artificielle 36 (6), 979 , 2022 2022 Citations: 8
Review on Data Compression Methods of Smart Grid Power System Using Wavelet Transform R Jadhav, A Mahajan Smart Energy and Advancement in Power Technologies: Select Proceedings of … , 2022 2022 Citations: 2
Data compression and noise reduction in smart grid using discrete wavelet transform RY Jadhav, A Mahajan Traitement du Signal 39 (5), 1857 , 2022 2022 Citations: 4
Circularly Polarized Flexible Dual-Band Microstrip Antenna Using Kapton Material. TN Pawase, A Malhotra, A Mahajan International Journal of Microwave & Optical Technology 17 (2) , 2022 2022 Citations: 1
MOST CITED SCHOLAR PUBLICATIONS
Area-and power-efficient architecture for high-throughput implementation of lifting 2-D DWT BK Mohanty, A Mahajan, PK Meher IEEE transactions on circuits and systems ii: express briefs 59 (7), 434-438 , 2012 2012 Citations: 85
An efficient VLSI architecture design for logarithmic multiplication by using the improved operand decomposition D Nandan, J Kanungo, A Mahajan Integration 58, 134-141 , 2017 2017 Citations: 74
An error-efficient Gaussian filter for image processing by using the expanded operand decomposition logarithm multiplication D Nandan, J Kanungo, A Mahajan Journal of ambient intelligence and humanized computing 15 (1), 1045-1052 , 2024 2024 Citations: 73
Polycystic Ovarian Syndrome Detection by Using Two-Stage Image Denoising. SB Choubey, A Choubey, D Nandan, A Mahajan Traitement du signal 38 (4) , 2021 2021 Citations: 24
Automatic liver cancer detection using deep convolution neural network KM Napte, A Mahajan, S Urooj IEEE Access 11, 94852-94862 , 2023 2023 Citations: 21
An efficient VLSI architecture for iterative logarithmic multiplier D Nandan, J Kanungo, A Mahajan 2017 4th international conference on signal processing and integrated … , 2017 2017 Citations: 20
An efficient antilogarithmic converter by using 11-regions error correction scheme D Nandan, A Mahajan, J Kanungo 2017 4th international conference on signal processing, computing and … , 2017 2017 Citations: 16
An Efficient VLSI architecture design for antilogarithmic converter by using the error correction scheme D Nandan, J Kanungo, A Mahajan International Conference on Signal Processing(ICSP 2016), 1-5 , 2016 2016 Citations: 15
Liver segmentation and liver cancer detection based on deep convolutional neural network: a brief bibliometric survey KM Napte, A Mahajan Library Philosophy and Practice, 1-27 , 2021 2021 Citations: 13
Area-delay efficient and low-power carry skip adder for high performance computing systems S Patel, B Garg, A Mahajan, S Rai 2019 IEEE International Symposium on Smart Electronic Systems (iSES … , 2019 2019 Citations: 13
An efficient architecture of iterative logarithm multiplier D Nandan, J Kanungo, A Mahajan International Journal of Engineering & Technology 7 (2.16), 24-28 , 2018 2018 Citations: 13
A review on recent work on oct image classification for disease detection J Subhedar, A Mahajan 2022 OPJU International Technology Conference on Emerging Technologies for … , 2023 2023 Citations: 11
Efficient VLSI architecture for implementation of 1-D discrete wavelet transform based on distributed arithmetic A Mahajan, BK Mohanty 2010 IEEE Asia Pacific Conference on Circuits and Systems, 1195-1198 , 2010 2010 Citations: 9
Deep learning based liver segmentation: A review K Napte, A Mahajan Revue d'Intelligence Artificielle 36 (6), 979 , 2022 2022 Citations: 8
Development of signal processing algorithm for optical coherence tomography K Patil, A Mahajan, S Balamurugan, P Arulmozhivarman, R Makkar 2020 International Conference on Communication and Signal Processing (ICCSP … , 2020 2020 Citations: 8
Liver segmentation using marker controlled watershed transform. KM Napte, A Mahajan International Journal of Electrical & Computer Engineering (2088-8708) 13 (2) , 2023 2023 Citations: 7
Smart grid data compression and reconstruction by wavelet packet transform R Jadhav, A Mahajan MethodsX 13, 102872 , 2024 2024 Citations: 6
Behavioural Analysis of Hatchetman Attacker for Detection & Prevention in Low Power and Lossy IoT Network. G Soni, K Chandravanshi, AS Kaurav, A Saxena, D Nandan, A Mahajan Instrumentation, Mesures, Métrologies 23 (5) , 2024 2024 Citations: 6
An efficient architecture of leading one detector D Nandan, J Kanungo, A Mahajan International Journal of Pure and Applied Mathematics 118 (14), 267-272 , 2018 2018 Citations: 6
Autism spectrum disorder detection using parallel DCNN with improved teaching learning optimization feature selection scheme T Dhamale, S Bhandari, V Harpale, P Sakhi, K Napte, A Mahajan SAIEE Africa Research Journal 116 (3), 89-100 , 2025 2025 Citations: 5
