Browse Items (2150 total)
Sort by:
-
Smart Skin Cancer Diagnosis: Integrating SCA-RELM Method for Enhanced Accuracy
One out of three cancers now is skin cancer, a figure that has exploded in the previous several decades. Melanoma is the worst kind of skin cancer and occurs in 4% of cases. It is also the most aggressive type. The health and economic impact of skin cancer is substantial, especially given its rising incidence and fatality rates. However, with early detection and treatment, the 5-year survival rate for skin cancer patients is much improved. As a result, a lot of money has gone into studying the disease and developing methods for early diagnosis in the hopes of stopping the rising tide of cancer cases and deaths, particularly melanoma. In order to enhance non-invasive skin cancer diagnosis, this research examines a range of optical modalities that have been utilized in recent years. The suggested system uses image processing to identify, remove, and categorize lesions from dermoscopy images; this system will greatly aid in the detection of melanoma, a type of skin cancer. A median filter is employed for preprocessing. Using watershed and clever edge detector, it can segment objects. The BOF plus SURF method is employed for feature extraction. It employs SCA-RELM, which performs better than the other two conventional approaches, to train the model. 2024 IEEE. -
Melanoma Skin Cancer Detection using a CNN-Regularized Extreme Learning Machine (RELM) based Model
Recent years have brought a heightened awareness of skin cancer as a potentially fatal type of human disease. While all three forms of skin cancer - Melanoma, Basal, and Squamous are terrifying, Melanoma is the most erratic. Melanoma cancer is curable if caught at an early stage. Multiple current systems have demonstrated that computer vision can play a significant role in medical image diagnosis. This study suggests a new approach to picture categorization that can help convolutional neural networks train more quickly (CNN). CNN has seen widespread use in multiclass image classification datasets, but its poor learning performance for huge volumes of data has limited its usefulness. On the other hand, whereas Regularized Extreme Learning Machine (RELM) are capable of rapid learning and have strong generalizability to improve their recognized accuracy quickly. This study introduces a novel CNN-RELM, a novel classifier that integrates convolutional neural networks with regularized extreme learning machines. CNN-RELM begins by training a Convolutional Neural Network (CNN) through the gradient descent technique until the desired learning and target accuracy is achieved. This approach outperforms the CNN and RELM model with an accuracy of around 98.6%. 2023 IEEE. -
Should Crypto Integrate Micro-Finance option?
Purpose - The purpose of the study is to identify the readiness or acceptance by the younger population specifically, the school and university students towards the investment in cryptocurrency if the micro-finance option is included in such new asset investments. Further to this the research also focusses on the mediating factor as trustworthiness to identify the impact or influence of the variable towards the acceptance of the new asset investment.Design/methodology/approach - The research conducted through relevant literature with sufficient variables measuring with five point Likert's scale. The research also tested with hypothesis on the relationship with variables. A total of 293 valid respondents data were collected and analysed through Structural Equation model.Findings - The analysis and results suggested that the perception, awareness and trustworthiness has positive impact towards the readiness towards crypto investments. Whereas, the investment behaviour has complex acceptability towards the readiness as it failed to accept the hypothesis.Research limitations/implications - the research is limited with the younger population however the research did not focusses on the economically challenged population as they may not be afford to invest in such platforms. The future studies can also be focussed on the same area with more towards the other factors that influence the economically challenged population and identify solution their economic growth. Furthermore, the study may be game changer for the policy makers in legalising the crypto investments in the country.Originality/value - According the wider background study and with substantial literature the research is of first in its kind as per the author's knowledge to integrate the micro finance concept in crypto investments to promote the investment habit among the younger population. 2024 IEEE. -
A Survey on Feature Selection, Classification, and Optimization Techniques for EEG-Based BrainComputer Interface
In braincomputer interface (BCI) systems, the electroencephalography (EEG) signal is extensively utilized, as the recording of EEG brain signals is having relatively low cost, the potentiality for user mobility, high time resolution, and non-invasive nature. The EEG features are extracted by the BCI to execute commands. In the feature set obtained, the computational complexity increases, and poor classifier generalization can be caused by the utilization of a lot of overlapping features. The irrelevant features accumulation could be avoided with the feature selection procedures application. The feature selection algorithms are utilized to select diverse features for each classifier. Classifiers are the algorithms that are run to attain the classification. The researchers have examined diverse classifier implementation techniques to identify the feature vectors class. A review of EEG-BCI techniques available in the literature for feature selection, classifiers, and optimization algorithms is presented in this work. The research challenges, gaps, and limitations are identified in this paper. The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024. -
A Deep Learning Method for Classification in Brain-Computer Interface
Neural activity is the controlling signal used in enabling BCI to have direct communication with a computer. An array of EEG signals aid in the selection of the neural signal. The feature extractors and classifiers have a specific pattern of EEG control for a given BCI protocol, which is tailor-made and limited to that specific signal. Although a single protocol is applied in the deep neural networks used in EEG-based brain-computer interfaces, which are being used in the feature extraction and classification of speech recognition and computer vision, it is unclear how these architectures find themselves generalized in other area and prototypes. The deep learning approach used in transferring knowledge acquired from the source tasks to the target tasks is called transfer learning. Conventional machine learning algorithms have been surpassed by deep neural networks while solving problems concerning the real world. However, the best deep neural networks were identified by considering the knowledge of the problem domain. A significant amount of time and computational resources have to be spent to validate this approach. This work presents a deep learning neural network architecture based on Visual Geometry Group Network (VGGNet), Residual Network (ResNet), and inception network methods. Experimental results show that the proposed method achieves better performance than other methods. 2023 IEEE. -
Efficiency of Indian Banks with Non-Performing Assets as Undesirable Outputs
The performance evaluation of any banks is of utmost importance for bank management, investors, and policymakers. Due to globalization, all the banks are working in a competitive environment. Several risk factors affect the operational efficiency of banking system. This study aims to evaluate the efficiency of Indian banks with NPAs as uncontrolled variables. Due to the nature of NPAs, these are assumed as undesirable outputs in the DEA modelling. The results reveal that public sector banks experienced more input losses due to NPAs compared to private banks. The private banks experienced more loss in inputs due to the scale of operation. The Wilcoxon Signed-Rank test shown that the impact of NPAs and scale of operation are statistically significant at 0.05 level. 2023 American Institute of Physics Inc.. All rights reserved. -
The impact of Covid-19 on global upstream and downstream supply chain management activities
COVID-19 pandemic has affected thousands of people worldwide; with significant economic changes in the past and to the changes to be made for future. Many organisations especially; The Intergovernmental economic organisation (OECD - The Organisation for Economic Co-operation and Development) warned the companies and industries on the global economic cut, the corona virus will be boarding. The global economy and international markets pitched down with the spread of corona virus spreading from China which is the world's second largest economy to other countries including Asia; Europe; Australia; Europe; America and the Middle East. Many economies came up with many policies to prevent the further spread of this virus; including restrictions on travel and quarantines; which has disrupted international supply chains affecting a lot of business operations and dwindle revenues. About 75 percent of business including Wholesale; Manufacturing; Retail and Services in China and about 51,000 companies have this impact at a global level according to data from Dun and Bradstreet. The success or failure of every Business depends on how well they manage their supply chain management activities. The impact of corona virus on supply chain activities is twofold. One is; Upstream Supply chain management where companies should monitor the backward integrated activities in procuring the inventory; which has accommodated a loss in the production because of closure of factories and a slowdown in the economy. Second is; Downstream Supply chain management where the intermediaries and middlemen face a lot of problems because of scarcity in inventory and many quarantine measures taken by many economies. Many disruptions in both Upstream and Downstream Supply chains lead to severe scarcity of inventory which was experienced globally by all the economies. This situation has made many economies to think of the inter connectivity and inter dependency among global nations in terms of supply chain. This article is aimed to highlight the effects and changes COVID-19 pandemic has brought in the supply chain industry from both Upstream and Downstream perspective. 2022 Author(s). -
Artificial Intelligence & Automation: Opportunities and Challenges
Artificial Intelligence (AI) and Automation innovation are growing at a steady rate that are changing organizations and bringing efficiency and adding to the economic development. The utilization of AI and robotization will likewise help improve different areas from wellbeing to horticulture. Furthermore, utilizing Automation and Artificial Intelligence would, follow the schedule, transform the idea of work and the working environment itself. For sure, machines will actually do large numbers of the undertakings typically done by people, just as supplement manual work and play out certain errands that an individual wouldn't have the option to do. Consequently, AI and mechanization have a great deal to bring to organizations and enterprises worldwide. This research paper comes up with a rundown through the blooming of Artificial Intelligence and Automation. We explored the existing potentiality of cognitive emerging technologies. This paper outlines the discussion about artificial intelligence and automation technologies and an overview of the applications. 2023 American Institute of Physics Inc.. All rights reserved. -
Predictive Modeling of Solar Energy Production: A Comparative Analysis of Machine Learning and Time Series Approaches
In this study, we dive into the world of renewable energy, specifically focusing on predicting solar energy output, which is a crucial part of managing renewable energy resources. We recognize that solar energy production is heavily influenced by a range of environmental factors. To effectively manage energy usage and the power grid, it's vital to have accurate forecasting methods. Our main goal here is to delve into various predictive modeling techniques, encompassing both machine learning and time series analysis, and evaluate their effectiveness in forecasting solar energy production. Our study seeks to address this by developing robust models capable of capturing these complex dynamics and providing dependable forecasts. We took a comparative route in this research, putting three different models to the test: Random Forest Regressor, a streamlined version of XGBoost, and ARIMA. Our findings revealed that both the Random Forest and XGBoost models showed similar levels of performance, with XGBoost having a slight edge in terms of RMSE.. By providing a comprehensive comparison of these different modeling techniques, our research makes a significant contribution to the field of renewable energy forecasting. We believe this study will be immensely helpful for professionals and researchers in picking the most suitable models for solar energy prediction, given their unique strengths and limitations. 2024 IEEE. -
Advancements in Medical Imaging: Detecting Kidney Stones in CT Scans using a ELM-I AdaBoost-RT Model
Kidney stones have been more common in recent years, leading many to believe that the condition is common. The condition's strong relationship with other terrible diseases makes it a major threat to public health. The development of instruments and procedures that facilitate the diagnosis and treatment of this ailment has the potential to enhance the effectiveness and efficiency of health care. Preprocessing, feature extraction, level set segmentation, and model training are the four steps that make up this approach. Part of the preprocessing includes eliminating the skeletal skeleton and soft-organs. Level set segmentation is commonly used for object tracking, motion segmentation, and image segmentation. An extremely effective feature extraction method called Gray level co-occurrence matrix (GLCM) is suggested for extracting the necessary characteristics from the segmented image. That ELM-I-AdaBoost-RT was used all during training. This cutting-edge technique achieves an average accuracy of 95.83%, surpassing both ELM and AdaBoost. 2024 IEEE. -
On Automatic Target Recognition (ATR) using Inverse Synthetic Aperture Radar Images
Inverse Synthetic Aperture Radar (ISAR) is used to image sea surface targets during day/night and all-weather capabilities for applications such as coastal surveillance, ship self-defense, suppression of drug trafficking etc. Hence automating classification of ships by means of machine learning methods has become more significant. Typical classification approaches consist of pre-processing, feature extraction and processing by classifiers. Image processing techniques are applied for pre-processing ISAR images. Transformation invariant features are then extracted to which classifiers such as SVM, Neural Networks (NNs) are applied. The performance of these algorithms depend on the manually chosen features and is trained to perform well for different target profiles. The target image (profile of target) varies depending on the target type, aspect angle and motion introduced due to different sea states. In addition, Deep learning methods are also being explored for classification of ships. The challenge is to classify ships for different sea conditions and image acquisition parameters with limited database and processing resource. 2023 IEEE. -
A Review of Smart Grid Management Systems Using Machine Learning Algorithms for Efficient Energy Distribution
The smart grid is an intelligent electricity network that uses digital technology to improve the efficiency, reliability, and sustainability of power delivery. Machine learning is a type of artificial intelligence that can be used to analyze data and learn from it. This makes it a valuable tool for the smart grid, as it can be used to solve a variety of problems, such asorecasting energy demand, detecting, and preventing outages, optimizing power flows, managing distributed energy resources, ensuring grid security.In this article, we will review the use of machine learning in the smart grid. We will discuss the different machine learning algorithms that are being used, the challenges that need to be addressed, and the future of machine learning in the smart grid.. The Authors, published by EDP Sciences, 2024. -
A Secure Data Encryption Mechanism in Cloud Using Elliptic Curve Cryptography
Cloud computing is undergoing continuous evolution and is widely regarded as the next generation architecture for computing. Cloud computing technology allows users to store their data and applications on a remote server infrastructure known as the cloud. Cloud service providers, such Amazon, Rackspace, VMware, iCloud, Dropbox, Google's Application, and Microsoft Azure, provide customers the opportunity to create and deploy their own applications inside a cloud-based environment. These providers also grant users the ability to access and use these applications from any location worldwide. The subject of security poses significant challenges in contemporary times. The primary objective of cloud security is to establish a sense of confidence between cloud service providers and data owners inside the cloud environment. The cloud service provider is responsible for ensuring user data's security and integrity. Therefore, the use of several encryption techniques may effectively ensure cloud security. Data encryption is a commonly used procedure utilised to ensure the security of data. This study analyses the Elliptic Curve Cryptography method, focusing on its implementation in the context of encryption and digital signature processes. The objective is to enhance the security of cloud applications. Elliptic curve cryptography is a very effective and robust encryption system due to its ability to provide reduced key sizes, decreased CPU time requirements, and lower memory utilisation. 2024 IEEE. -
Designing a Precision Seed Sowing Machine for Enhanced Crop Productivity
A seed sowing machine is a valuable agricultural device that facilitates the precise and efficient sowing of seeds in fields. When designing and optimizing such a machine, several crucial factors need consideration including seed size, seed rate, soil type, and field conditions. The primary objective is to achieve uniform seed distribution and optimal seed-to-soil contact, which can be accomplished by incorporating a seed metering mechanism to control the seed rate accurately. Versatility is another important aspect of the machine's design, as it should be able to handle different seed sizes, types, soil conditions, and field variations. To achieve this, utilizing advanced technologies such as sensors, automation, and precision farming techniques can significantly enhance the machine's performance and efficiency while also reducing costs and minimizing environmental impact. The optimization of a seed sowing machine plays a crucial role in ensuring successful crop production. By implementing cutting-edge technologies and precision farming techniques, farmers can increase their yields and decrease the amount of seed and fertilizer needed for a specific area. Ultimately, this leads to improved productivity, increased profitability, and a more sustainable approach to agriculture. 2024 E3S Web of Conferences -
Multiplier-free Realization of High throughout Transpose Form FIR Filter
This paper presents a multiplier-free realization of the block finite impulse response (FIR) filter in transpose form configuration using binary constant shifts method (BCSM). The proposed architecture is synthesized using Xilinx Vivado and Cadence RTL Encounter compiler for the area and power analysis and is compared with the existing works in the literature. The comparison highlights the advantages of the proposed architecture in terms of power, hardware complexity and throughput for realizing reconfigurable high throughput block FIR filters. 2020 IEEE. -
A Comparative Study in Predictive Analytic Frameworks in Big Data
Every information processing sector uses predictive analytic framework in terms of distributed datasets through a variety of applications. These analytic frameworks are effectively used for various analyses of data, parameter, and attributes. Leveraging data to make insightful decisions for maximizing the effectiveness requires the determination of the best predictive framework for any organization. Even a retail unit which wants to scale up its production rely on multiple parameters. These parameters must be analyzed for effective quality control in any domain. Since there are diversities in every domain the data will be in varied form, and these are accumulated as Big Data. These analyses are done using machine learning frameworks. The strategy involved would differ from one domain to another such as in the health care sector the framework might predict the magnitude of patients admitted to the urgent care facility over the upcoming days whereas in the production industry the framework would align quality control measures. This article analyses a few domains and their deployed machine learning impacts in a strategic way. 2023 American Institute of Physics Inc.. All rights reserved. -
A microstructure exploration and compressive strength determination of red mud bricks prepared using industrial wastes
The consensual view among researchers concerning building with industrial by-products is that the utilization of by-products represents green technology and sustainable development. The current investigation focuses on the utilization of an assortment of by-products for the production of bricks. The by-products include Red Mud (RM), Fly ash (FA), and Ground Granulated Blast Furnace Slag (GGBS) combined in different proportions with lime. The Red Mud employed ranged from 100% to 60% with a decrement of 10%, whereas Fly ash and GGBS varied between10% and 40% with an increment of 10%. Bricks produced from two methods namely, ambient curing and firing methods, were tested as per IS standards/ASTM norms, on both the materials and the composites of bricks. XRD, XRF, and SEM focused on both the raw materials and the composites. Because geopolymer materials are partially amorphous materials with complex composition, understanding the structural characteristics of geopolymers is opined as intricate. The results of the investigation show that the compressive strength of the bricks increased with the increment in the percentage of Fly ash and GGBS. The compressive strength of Red Mud-GGBS fired bricks attained maximum strength of 7.56 MPa. 2021 Elsevier Ltd. All rights reserved. -
Experimental investigation of boundary shear stress in meandering channels
Laboratory experimentation for bed shear stress distribution has been carried out in two sets of meandering channels. The channels have cross-over angles of 110 and 60 constructed by 'sine-generated' curves over a flume of 4?m width. Variations in bed roughness were studied for the meandering main channel. Bed shear stress distribution across a meandering length for the 110 and 60 channels was examined for different sinuosities and roughnesses. The boundary shear stress study illustrated the position of maximum shear along the apex section and across the meandering path. These variations were observed for different flow depths. A comparison of the bed shear among the three experimental channels was conducted, and the results were analyzed. 2023 Author(s). -
An Efficient HOG-Centroid Descriptor for Human Gait Recognition
Automatic recognition of human gait have gained much attention nowadays. Histogram of Oriented Gradient (HOG) is a widely adopted descriptor for object's shape analysis. In this paper, combination of HOG descriptor with silhouette centroid for human gait recognition is proposed. The resultant descriptor, namely HOG-Centroid, achieves better recognition performance on comparison with HOG descriptor individually as well as other existing gait recognition methods. Experiments are carried out with CASIA gait dataset B and cumulative matching scores of 95.3%, 98.1% and 99.2% are obtained for rank 1, rank 5 and rank 10 respectively. 2019 IEEE. -
Discriminative Gait Features Based on Signal Properties of Silhouette Centroids
Among the biometric recognition systems, gait recognition plays an important role due to its attractive advantages over other biometric systems. One of the crucial tasks in gait recognition research is the extraction of discriminative features. In this paper, a novel and efficient discriminative feature vector using the signal characteristics of motion of centroids across video frames is proposed. These centroid based features are obtained from the upper and lower regions of the gait silhouette frames in a gait cycle. Since gait cycle contains the sequence of motion pattern and this pattern possesses uniqueness over individuals, extracting the centroid features can better represent the dynamic variations. These variations can be viewed as a signal and therefore the signal properties obtained from the centroid features contains more discriminant information of an individual. Experiments are carried out with CASIA gait dataset B and the proposed feature achieves 97.3% of accuracy using SVM classifier. 2019, Springer Nature Singapore Pte Ltd.