Browse Items (7684 total)
Sort by:
-
Flow of nanofluid past a stretching cylinder subject to Thompson and Troian slip in the presence of gyrotactic microorganisms
Incorporating the Thompson and Troian slip condition, this work studies the bioconvective flow of a nanofluid past a vertically stretching cylinder. The Thompson and Troian slip deals with the molecular scale interactions at the solidfluid interface, which plays a pivotal role in the fluid flow analysis. This study helps in understanding the behaviours of fluid flow in the presence of non-linear slip past a vertically stretching cylinder. The corresponding partial differential equations (PDEs) for momentum, energy, concentration of nanoparticles, and concentration of microbes are developed using Buongiornos model. A suitable similarity transformation is then applied to these PDEs, converting them into a set of ordinary differential equations (ODEs). The RungeKuttaFehlberg (RKF-45) method is utilized to calculate the numerical solution of the resulting ODE problem. The results demonstrate that the interaction of slip conditions, viscous dissipation, heat source, and bioconvection causes complex flow patterns and heat transfer characteristics. These observations are extremely relevant for applications including better oil recovery procedures, biomedical engineering, and microfluidic devices where exact control over nanofluid behaviour is necessary. Some of the major observations of the study include the enhancement of the temperature in the nanofluid for higher Eckert numbers, control of fluid flow through an external magnetic field, and Peclet number significantly decreased the motile density in the nanofluid. The Author(s), under exclusive licence to Springer Nature Switzerland AG 2024. -
FEDGE: FEDerated learning at the EDGE on space platforms using deep neural network architectures
We introduce FEDGE: FEDerated Learning at the EDGE, a framework designed for efficient AI deployment in resource-constrained satellite constellations. FEDGE integrates federated learning with edge computing to address communication overhead and latency challenges in distributed space environments. The framework features a novel edge-enhanced ground station protocol that dynamically schedules model aggregation based on satellite-provided metadata, combined with local stochastic gradient descent training at satellite edge devices and gradient compression via quantization. Experimental validation on MNIST and EuroSAT datasets demonstrates the practical viability of the approach. On MNIST, FEDGE achieved 94.33% training accuracy with 0.21 loss and 90.05% test accuracy with 0.24 loss. On EuroSAT, the framework reached 93.47% training accuracy with 0.18 loss and 91.51% test accuracy with 0.21 loss. Gradient quantization reduces data exchange by up to 14 with approximately 4% impact on test loss. These results validate FEDGE as a communication-efficient solution for decentralized AI deployment in satellite systems, enabling autonomous spacecraft intelligence and addressing the unique constraints of space-based computing platforms. The Author(s) 2025. -
Segmentation of overlapping leukemic cells in histopathological images using HSV- based watershed transformation
Accurate segmentation of white blood cells (WBCs) is essential for computer-aided diagnosis, as overlapping and densely clustered cells often present significant challenges. This work introduces a hybrid framework for segmentation that proposes a fusion of hue and saturation in the Hue Saturation Value (HSV) domain. Gaussian smoothing, Otsu thresholding, and Morphological refinement is employed to enhance cell contrast and eliminate noise. A marker-based watershed algorithm is subsequently applied for accurate separation of overlapping WBCs. Evaluation on the ALL-IDB2 dataset confirms the methods capability through achieving a Dice Similarity Coefficient(DSC) of 0.8929 and an Intersection over Union (IoU) of 0.8099 to produce well-defined cellular boundaries. The novelty of this study lies in the integrated hue-saturation fusion and marker-based watershed strategy, offering improved boundary localization and reliable segmentation of overlapping WBCs. Bharati Vidyapeeth's Institute of Computer Applications and Management 2025. -
Pediatric brain tumor segmentation and classification framework using SGC-U-NET and ARC-DEEP-CNN
Timely and precise pediatric Brain Tumor (BT) classification is challenging in the prevailing studies owing to the lack of growth rate calculation. Therefore, this paper proposes a growth rate-aware intelligent BT classification using child Magnetic Resonance Imaging (MRI) based on Arcsin Deep Convolutional Neural Network (Arc-Deep-CNN). Initially, the child's MRI is collected and then pre-processed for angle correction, resolution improvement, skull removal, and edge sharpening to improve the image quality. Meanwhile, the binary image dilation is done in the postpre-processing for accurate tumor location identification using the Central Limit Theorem-based Battle Royale Optimization Algorithm (CLT-BROA). From the pre-processed images, the wavelet features are extracted to improve the detection rate. Based on the tumor-identified images, pre-processed images, and extracted features, a robust Shuffled Group Convolutional layer added U-Net (SGC-U-Net) significantly segments the normal brain, benign, core, and malignant tumors affected brain. Then, the 3D tumor reconstruction is done by performing splitting, feature extraction, and growth rate calculation. Finally, a novel Arc-Deep-CNN proficiently classifies the BT into Medulloblastoma, Glioma, and Meningioma tumors with respect to the growth rate. The proposed Arc-Deep-CNN achieved maximum accuracy and minimum training time of 98.77% and 52136ms, respectively, showing impressive performance in pediatric BT classification. Bharati Vidyapeeth's Institute of Computer Applications and Management 2025. -
An intelligent black-box testing model for isolating logical flaws and anomalies in applications using GTMRM
Web Applications (WAs) are becoming more vulnerable to attacks as they are more popular. Nevertheless, the conventional testing methodologies didnt differentiate the Logical Flaws (LFs) and anomalies in WAs, thereby increasing the misclassification rate. Hence, in this paper, a novel black-box testing framework that incorporates an advanced technique called Gated Transformer Memorized transferred Recurrent Mishswish unit (GTMRM) is proposed for distinguishing between LFs and other vulnerabilities, thus enhancing the reliability of WAs. Initially, the user registration is carried out, followed by Hash-based Message Authentication Code Hash-based Message Authentication Code (HMAC) creation. Afterward, the registered users log into the application to request a Uniform Resource Locator (URL) for access. In the meantime, to authenticate the user, the HMAC verification is performed. Once the authentication is successful, the user is granted for accessing the functionalities. Thereafter, the black-box-centric LF and anomaly identification is done; here, the raw dataset is initially pre-processed. Subsequently, concerning a similar domain, the pre-processed data is clustered. Next, the features are extracted, followed by feature selection. Then, from the grouped data, the graph is constructed. The pattern labelling is carried out centered on the graph features. Lastly, the Logical Flaws (LF), anomaly, and legitimate access are proficiently classified by the proposed GTMRM. A compensation measure is applied in the case of a LF. After that, the data is securely stored in the cloud server with an accuracy of 99.14%. Bharati Vidyapeeth's Institute of Computer Applications and Management 2025. -
A hybrid GNNvanilla vision transformer model for IoT-based soil and crop forecasting
In this work, we propose a Graph?Neural Network (GNN) and Vanilla Transformer-based hybrid model for IoT driven soil and crop prediction. Conventional forecasting approaches are unable to model complicated spatial and temporal inter-dependencies and are not?very effective. The given paper solves this problem by using GNNs to learn the spatial relationships among the IoT sensor nodes and vanilla transformer model to?learn the temporal dependencies in crop and weather data. Vanilla vision transformer is able to recover missing contextual information during training. It is trained on data from IoT sensors that monitor soil moisture, temperature, humidity and a variety of other environmental factors as?well as historical crop yield and weather related information. The hybrid model can enable the real-time accurate prediction for crop?yield production and soil health status, which enables a smarter agriculture decision. The experimental results show that the proposed work achieves the lowest root mean square error (RMSE 2.1) and the highest crop accuracy (92%) for?short-term and long-term forecasts. Bharati Vidyapeeth's Institute of Computer Applications and Management 2025. -
A statistically guided hybrid machine learning framework for predicting supply chain resilience in complex operational environments
This study proposes a hybrid machine learning framework to predict supply chain resilience by integrating principal component analysis, K-Means clustering, and ensemble learning models. The approach captures firm-level heterogeneity, enabling context-specific resilience prediction and interpretability using SHAP values. The findings demonstrate that ensemble models, particularly XGBoost, outperform traditional regression models, and reveal distinct resilience drivers across operational clusters. The framework offers actionable insights for improving resilience strategies and contributes a scalable, explainable approach for data-driven supply chain risk management. Bharati Vidyapeeth's Institute of Computer Applications and Management 2025. -
AI-driven deep learning framework for energy-efficient optimization in IoT-enabled wireless networks
Artificial intelligence (AI) and Internet of Things (IoT)-enabled wireless sensor networks (WSNs) have revolutionized industries by providing automation, real-time monitoring, and analytics that are predictive. WSNs still face significant obstacles such data security, network flexibility, and energy limitations in spite of these developments. In order to optimize energy use in Internet of Things (IoT)-based WSNs, this study introduces a novel Reinforcement Learning-based Energy-Efficient Communication Protocol (RL-EECP) to optimize the lifetime of networks and guarantee effective data transmission. The suggested protocol integrates sleep scheduling, reinforcement learning, and data fusion techniques. Also, an adaptive prioritization approach is introduced that assesses nodes according to the surroundings, significance, and energy consumption. Experiments show that RL- EECP performs better than existing studies in extending node lifetime and preserving excellent network performance. Bharati Vidyapeeth's Institute of Computer Applications and Management 2025. -
A novel deep learning based multimedia video retrieval framework using may fly optimization
Developing a video retrieval framework in multimedia management is a main challenge due to the massive growth of video content on the internet. A major drawback of video retrieval is its long search response time and low accuracy. To tackle these issues, this paper introduces a novel deep learning-based Multimedia video retrieval system (DL-MVR) to minimize the search response time with high accuracy. The collected video is initially converted into key frames and pre-processed with contrast adaptive histogram equalization to remove noise artifacts thereby improving image quality. After pre-processing, the images are fed to Efficient Net to extract patch features. Finally, to retrieve the similar video, matching is done using may fly optimization (MFO), that compares the query frame features to the video database. Several performance metrics are analysed to measure the effectiveness of the proposed strategy in terms of accuracy and response time. Experimental results indicate that the proposed system has a search response time of 0.71s, which is lower than existing methods. The proposed DL-MVR method achieves 99.26% of accuracy. The proposed method improves the overall accuracy by 9.32%, 22.04%, and 19.40% which is better than CNN-AlexNet (convolutional neural network), Pyramid regional graph network and CBVR respectively. Bharati Vidyapeeth's Institute of Computer Applications and Management 2025. -
Computing isogeny on Edwards curves for quantum safe cryptography
In recent years, cryptographic research has seen a surge of interest in post-quantum cryptography driven by the potential threat that quantum computers pose to traditional public-key cryptosystems. Isogeny-based cryptography is a promising method in post-quantum cryptography, relying on the computational challenge of calculating isogenies, which are specific mappings between elliptic curves. The efficiency of isogeny computations is vital for real-world cryptographic applications. However, computing isogenies, especially with large parameters, can be very resource intensive. To overcome this challenge, we purpose an efficient method for computing odd-degree isogenies on certain form of an elliptic curves by employing an auxiliary coordinate. Our work appears to bridge the gap in computational efficiency for odd-degree isogenies, especially in terms of reducing the complexity of the isogeny computations when compared to traditional affine and projective methods. The derived formula is more efficient than affine and projective cases. We also analyse the algebraic complexity of these calculations and compare them to alternative formulae. Additionally, we evaluate the runtimes for isogeny computation across different prime numbers and compare them with other elliptic curve model to check the performance. At last, we suggest potential avenues for future work. Bharati Vidyapeeth's Institute of Computer Applications and Management 2025. -
Trusted explainable AI based implementation for detection of neurodegenerative disorders (ND)
The potential of explainable artificial intelligence (XAI) in detection of neurodegenerative disorders (ND) holds great promise in the field of healthcare. These diseases interfere with the daily functioning and independence of a person. The current studies lack in highlighting the aspect of explainability in their predictions and the various algorithms cannot provide any plausible explanations for their predictions making it difficult for medical professionals to place trust in their findings. Thus, the proposed framework aims to bridge this gap by exploring the development of a trustworthy framework for XAI-based ND detection, focusing on key aspects that can significantly impact its effectiveness and acceptance. The framework makes use of Trust-based SHAP (SHapley Additive exPlanations) values in classification. By computing trust values, the framework ensures more reliable predictions and increases interpretability, instilling confidence in clinicians and patients. The results show that with the inclusion of the trust-driven framework, the accuracy of the algorithm increased from 93.33% in the normal circumstances to 98.21%, highlighting the efficacy of the framework as compared to the other works. This shows that a trustworthy framework for XAI-driven ND detection can reshape care by enabling early detection, personalized treatment plans and enhancing decision-making process. Bharati Vidyapeeth's Institute of Computer Applications and Management 2024. -
Energy efficient heterogeneous clustering scheme using improved golden eagle optimization algorithmfor WSN-based IoT
In the Internet of Things, Wireless Sensor Networks (WSNs) are networks of interconnected sensors that wirelessly collect and transmit information about the environment. Using IoT sensors, IoT applications can remotely monitor and control physical environments. Clustering in WSNs involves organizing sensor nodes into groups called clusters with one or more CHs for efficient data integration, communication and management, improving network performance and resource utilization. In WSNs, achieving energy efficiency is critical to extend network lifetime and ensure stable operation. An important aspect contributing to energy optimization is the selection of CHs. However, the lack of an efficient cluster head selection mechanism remains a significant challenge. Therefore, this study introduces an optimized multivariate cluster head selection method that leverages the Improved Golden Eagle Optimization Algorithm (IGEOA). With this approach, the selection of CHs is optimized, combining multiple objective functions designed for energy efficiency. By using this algorithm, clusters are formed based on the selected CHs. In addition, a cluster maintenance phase is integrated to supervise the post-establishment clustering of the network, which ensures optimal cluster performance and resource utilization in WSN. Evaluation through simulation illustrates that the proposed method significantly improves both performance and energy efficiency in a WSN environment. Bharati Vidyapeeth's Institute of Computer Applications and Management 2024. -
Block chain-based security and authentication for forensics application using consensus proof of work and zero knowledge protocol
The technique that checks the origin, integrity, Zero-Knowledge authenticity of photographs is known as image authentication. Numerous studies on image authentication have revealed numerous trade-offs between four desirable features, namely robustness, security, flexibility, and efficiency. This study demonstrated a high-security Forensic Image (FI) as well as an authentication mechanism. Initially, the FI considered image registration with features for the Consensus method (CM) to generate blocks on each feature using a hypothesis test-based similarity measure. Because Proof-of-Work (PoW) blockchain technology is widely used, maintaining the Consensus PoW(CPoW) requires a massive amount of computing power. ZKP authentication is a critical cryptographic mechanism that authenticates network nodes without revealing the users identity or any other data given by the user. The blockchain stores the secret information, as well as the hash value of the original FI. This allows for the tracking of all medical pictures exchanged through the proposed blockchain network. The blockchain stores the private information as well as the hash value of the original medical image. The experimental results indicate the utility of the proposed approach with performance measures in contrast to established security analysis methods. Bharati Vidyapeeth's Institute of Computer Applications and Management 2024. -
Nonlinear Dynamics and Control of Driven Climate Variability and Ocean Heat Feedbacks
Abstract: Earths climate system is a highly complex and interconnected network governed by nonlinear interactions among the atmosphere, oceans, land, ice, and biosphere, where energy exchanges and feedback mechanisms play a dominant role. In recent decades, anthropogenic greenhouse gas emissions, especially carbon dioxide (), have significantly disrupted this balance, resulting in accelerated ocean heat uptake and persistent temperature anomalies. Determining the long-term dynamics of these interactions remains a critical challenge for accurate climate prediction and mitigation planning. This paper examines the combined dynamics of temperature anomaly, atmospheric concentration, and ocean heat content (OHC) using a novel mathematical approach. By employing the Caputo derivative to describe the model as a fractional-order dynamical system, hereditary effects and long-term dependencies that are inherent in climatic processes can be incorporated. Boundedness, existence and uniqueness of solutions, and both local and global stability are among the fundamental qualitative characteristics of the system that are investigated. To further illustrate stability behavior, streamline graphs are plotted. To ensure an accurate approximation of the fractional dynamics, numerical simulations are conducted using the Adams Bashforth Moulton (ABM) predictorcorrector method. Bifurcation analysis and computations of the Lyapunov exponent are performed to investigate the nonlinear properties of the system, exposing parameter regimes that behave chaotically for different fractional orders. Phase portraits in 2D and 3D show the intricate history of the climate variables. Additionally, to control chaotic oscillations, a sliding mode control approach is used. The findings highlight the promise of control theoretic techniques in climate dynamics by showing that the system is stabilized and chattering is successfully eliminated with the right control parameters. The results demonstrate that the fractional-order formulation provides enhanced capability in capturing long-term dependencies and nonlinear feedback mechanisms inherent in climate dynamics. The overall results show the models robustness as a theoretical framework for climate analysis and offer quantitative insights into the coupled climate systems long-term behavior. The models incorporation of nonlinear interactions among important variables improves the models interpretability and gives a more accurate picture of climate dynamics, which strengthens the foundation for assessing the effects of emissions and guiding the formulation of climate policy. King Abdulaziz University and Springer Nature Switzerland AG 2026. -
Neuro-fuzzy model optimization for laser sensor-based quality control for robotic welding of AISI 1030 steel
Robotic welding demonstrates considerable potential in the automation of metal joining processes, resulting in enhanced consistency. This study proposes a methodology for evaluating weld quality by utilizing a laser sensor in conjunction with a hybrid neuro-fuzzy model. The system, designed for AISI 1030 mild steel, utilizes a Design of Experimentation (DOE) methodology to collect empirical data and train the model. A MOTOMAN MA1440 robotic arm, integrated with an AccuFast-II laser sensor, was utilized to acquire real-time weld characteristics. The proposed model integrates fuzzy logic with artificial neural networks (ANNs) for predicting weld quality and is subsequently optimized using the Class Topper Optimization (CTO) algorithm. The model exhibited a high level of prediction accuracy, as indicated by R-squared values of 1.0, 0.99677, 0.99851, and 0.97561 for the training, testing, validation, and overall WQCI datasets, respectively. The process parameters obtained from the CTO analysis yielded a WQCI of 0.824, exceeding the highest experimental value of 0.808, which reflects a 1.98% enhancement in weld quality. The system demonstrated strong performance on both straight and curved weld paths, achieving a positional error of less than 0.29 mm, which falls within the acceptable weld gap range of 11.6 mm. This study emphasizes the practical implementation of a neuro-fuzzy prediction system integrated with an innovative metaheuristic for quality control in robotic arc welding. The integration improves weld consistency, minimizes defects, and increases production efficiency, representing a notable advancement in intelligent manufacturing. The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd. 2026. -
Harnessing MnMoO4 nanoparticles for eco-conscious effluent degradation and catalytic applications
The increasing need for green technology solutions to reduce water pollution and enhance sustainable catalysis has prompted the research for efficient photocatalysts. In this research, a green synthesis method was adopted to synthesize MnMoO4 NPs using solution combustion route followed by calcination. Synthesized MnMoO4 showed superior photocatalytic performance under visible light, with 91% degradation of Rose Bengal dye in aqueous medium indicates its potentiality for wastewater treatment. The material also showed catalytic efficiency in the coupling reaction of aniline and dimedone as model substrates for the synthesis of ?-enaminone derivatives, displaying its usability in organic catalysis. The work highlights the dual functional ability of MnMoO4 both an organic catalyst and a photo-catalyst, providing a green path for synthetic and environmental applications. The dual functionality combined with green fabrication process exemplifies the novelty and applicability of this article. The Author(s), under exclusive licence to Springer Nature Switzerland AG 2025. -
Conflict and Coexistence of Human Rights: An Exploratory Study with Reference to Intellectual Property Rights
Human rights and Intellectual Property Rights (IPRs) have developed independently. Human rights are inalienable rights associated with human dignity while IPRs are the rights with the goal of promoting innovations and the interests of select communities to further economic and technological growth. The economic and personal interests of the individual have received prime attention under the international intellectual property law. Economic growth is given priority over human rights in the international criteria for IPRs in global trade. Whereas, it has a significant impact on the implementation of human rights for both individuals and communities, including the rights to adequate food, health, environment, and education. IPRs are gravely at odds with human rights, even though a connection between the two rights can be found in General Comment No. 17 on Article 15(1)(c) of the International Covenant on Economic, Social, and Cultural Rights (ICESCR) and Article 27 of the Universal Declaration of Human Rights (UDHR). According to the UDHR, intellectual property is a human right in and of itself, but its enforcement often infringes other human rights. In light of the above perspective, the authors explore the interrelationship between IPRs and human rights and also analyze the evolving IPRs, in different fields of its application, causing adversarial impacts on several other human rights. The Author(s), under exclusive licence to Springer Nature Switzerland AG 2025. -
Effect of alkali treated palmyra fibers on strength and durability properties of binary blended concrete
The exponential growth in urban as well as industrial development has led to growing interest in waste management and utilizing industrial byproducts. The palmyra fiber is an abundantly available fiber extracted from the palmyra palm tree whose potential as reinforcing material is less explored. This study investigates the influence of alkali treated palmyra fibers on the strength and durability properties of binary blended concrete with 80% cement and 20% ground granulated blast-furnace slag. The alkali treated palmyra fibers of 50mm length were added in three different proportions of 0.5%, 1% and 1.5% by mass of binder materials to produce M30 grade of concrete. The workability of binary blended concrete was reduced with the addition of alkali treated palmyra fibers. Comprehensive investigations were carried out on both mechanical (compressive, split tensile, and flexural strength) and durability properties (sorptivity, resistance to sulphate and acid attack). Additionally, the performance under impact loading was also evaluated. The results reveal that compressive strength nominally reduced by 312% with the addition of fibers, while tensile strength and flexural strength increased with every increment in fiber content. The inclusion of palmyra fibers considerably increased impact resistance, ranging from 300 to 600% compared to conventional concrete. Also, palmyra fiber reinforced concrete exhibited better resistance to sulphate attack. Springer Nature Switzerland AG 2025. -
A novel discrete slash family of distributions with application to epidemiology informatics data
This study puts forward a new class of discrete distribution that can be used by the epidemiologists and medical scientists to model data relating to epidemiology informatics. The proposed distribution is superior to traditional discrete modeling alternatives, viz., discrete Weibull and geometric distributions in terms of its model fit and flexibility to handle heavy-tailed dataset. It is a flexible three-parameter discrete distribution, grounded in the slash family and can be considered as a refined extension to the geometric distribution. We explored the diverse properties of this novel distribution thoroughly by evaluating the mathematical properties. The models parameters are estimated using the maximum likelihood estimation method, where the methodology validity is confirmed through an extensive simulation study. Furthermore, the practical utility of the distribution to model epidemiology informatics was examined with the help of eight different datasets representing three different dimensions of the epidemiology informatics, viz., mortality, infection and medication statistics. The Author(s), under exclusive licence to Springer Nature Switzerland AG 2024. -
Policies and metrics for schedulers in cloud data-centers using CloudSim simulator
Todays cloud technology consumers must address escalating computing and storage demands for services and applications. However, decision-making on provisioning and scheduling is challenging due to varying workflow demands within Infrastructure as a Service (IaaS). This study formulates an optimization problem with multiple objectives to identify optimal policies, employing heuristic metrics through cloud simulation similar to AWS EC2 instances. Experiments involve two task scheduler types, time-shared and space-shared, aimed at minimizing execution time and cost. The study introduces two novel algorithms, SLB and MinMax, for comparison with standard algorithms. It emphasizes the importance of precise quantification of uncertainty in cloud storage allocation and highlights the state-of-the-art policies and metrics achieved through virtualization techniques. The studys novelty lies in simulating both policies at two levels and proposing a novel algorithm for multi-objective optimization while providing cost and time measurements. Contributions include experimenting with various combinations, applying heuristics to entire data center entities, proposing a novel algorithm, and offering cost and time measurements for the optimizations. The Author(s), under exclusive licence to Springer Nature Switzerland AG 2023.