Digital Eprints Services at Vignan's Foundation for Science, Technology & Research
Not a member yet
721 research outputs found
Sort by
Transfer learning-based deep ensemble neural network for plant leaf disease detection
No full textDeep learning is quickly becoming the standard technique for image classification. The main problem facing the automatic identification of plant diseases using this strategy is the lack of image databases capable of representing the wide variety of conditions and symptom characteristics found in practice. Data augmentation techniques decrease the impact of this problem, but those cannot reproduce most of the practical diversity. This
paper explores the use of individual lesions and spots for the task, rather than considering the entire leaf. Since each region has its own characteristics, the variability of the data is
increased without the need for additional images. This also allows the identification of multiple diseases affecting the same leaf. On the other hand, suitable symptom segmentation still needs to be done manually, preventing full automation. The accuracies obtained using this approach were, in average, 12% higher than those achieved using the original images. Additionally, no crop had accuracies below 75%, even when as many as 10 diseases were considered. Although the database does not cover the entire range of practical possibilities, these results indicate that, as long as enough data is available, deep learning techniques are effective for plant disease detection and recognitio
Anomaly Detection in Credit Card Transaction using Deep Learning Techniques
Get PDFIDS (intrusion detection systems) use analysis of
network traffic patterns to detect incidents of hacking. It is
essential to do feature extraction in order to minimize the
computational cost associated with processing raw data in the
IDS. Feature extraction decreases the number of features,
which decreases the time it takes to train and increases
accuracy. This research employs a simple LSTM autoencoder
and a Random Forest to recognize intrusion attempts by IDSs.
By activating and disabling various characteristics, the extent
to which this feature extraction function can enhance accuracy
is examined. To find out if detection algorithms are effective
after feature extraction, the NSL-KDD dataset has been
employed. Autoencoder hyperparameters contain the two
activation functions. The loss and activation functions of the
ReLU and the SoftMax have the greatest accuracy rating of
any function. The use of a Long Short-Term Memory
Autoencoder (LSTMAE) and a Random Forest (RF) for
identifying the best features is a goal of this study. According
to preliminary experimental data, classifiers that employ these
variables have a prediction rate of 94.74 percent
Significance of Blockchain Technologies in Industry
Get PDFBlockchain technology is set to extremely affect a wide assortment of enterprises, extending from capital markets to the music business. While some uti-lization cases may appear glaringly evident, the innovation is as yet encircled by a
lot of promotion and vulnerability. As a chief, in what manner would it be a good idea for you to move toward the subject, and when would it be a good idea for you to get the ball rolling and effectively expect to execute blockchain innovation?
As indicated by Juniper Research, six of ten enormous enterprises are either effectively considering or during the time spent sending the blockchain revolution. Among organizations that have arrived at the proof of concept stage, 66% (66 per-
cent) expected blockchain to be incorporated into their frameworks before the fnish of 2018. The examination guaranteed that those organizations that would proft most from blockchain incorporate those with the requirement for (1) straightfor-wardness in exchanges, (2) current reliance inheritance stockpiling frameworks, and (3) a high volume of transmitted data. Taking a gander at the explanations behind actualizing blockchain, there is a characteristic hazard that supervisors anx-ious to investigate new advancements from a hasty opinion without investigating elective alternatives. As per the exploration, foundational change, as opposed to
innovation, may give both better and less expensive answers for the current issue. For some organizations, the go-to way to deal with examining potential use cases for blockchain is to search for wasteful aspects in ebb and fow processes. This
approach is ensured to give a few outcomes; however, frequently the arrangement is to genuinely restructure inheritance procedures to ft an advanced world as opposed
to investigating new and obscure advances. One motivation behind why blockchain frequently develops as a response to numerous issues is that it is anything but diff-cult to envision signifcant level use instances of blockchain innovation. Be that as it may, as we adventure under the outside of such use cases, applying blockchai
An Intelligent IoT Framework for Handling Multidimensional Data Generated by IoT Gadgets: Applications and Use Cases
No full textIn recent years, a series of real-life problems are being solved by the leading role of sensors and the Internet. Smart towns, smart health structures, smart construction, smart landscapes, and smart transport are all part of the applications. However, the IoT sensor data involves a variety of problems in real time, including dilution of unhygienic sensor data and extraordinary resource costs. In addition to normal clinical practices, information and communications technology (ICT) that enables Internet of Things for the development of mechanisms to control elderly behavior allows geriatrics to detect changes in behavior related to such conditions early on. The data capture layer is a discreet low-cost infrastructure that sums up physical system heterogeneity, while data processing capacity handles the huge amount and semantics of sensed knowledge easily. Details are accessible with wired or wireless Internet access. These create enormous amounts of fresh, organized, unstructured, real-time, and big data. The IoT data is very comprehensive and nuanced, with information on the circumstances and the environment. The IoT will never remain idle, as thousands of Internet articles become information collectors and produce huge data. Today’s bulk of large data consists of IoT devices and grows exponentially per year. The analysis of such data needs innovative IoT techniques and data processing. IoT requires a specific abundance of facts, however, which continuously flow from various objects. Conventional technology is also critical. IoT data are very detailed and nuanced, capable of delivering in real time information on actual events or the environment. A smart IOT system has been developed in this manuscript to handle multidimensional data generated by IOT sensors. The model suggested showed that the accuracy and speed of data handling are high when compared to traditional models and to current models
Analysis of Indian and American poetry using topic modeling and Deep learning
Get PDFText classification is a supervised machine learning technique that assigns a set of predefined categories or classes to the given text corpora based on the content of the processed text using Natural language processing techniques. Text classification is widely used in numerus applications such as categorizing
the sentiment of the tweets and reviews, classification of news and web pages into multiple categories and automatic classification of emails in to spam or not spam. Under the text categories poetry is a lit- erary text and it is special when compared with the regular prose text. A very less focus is given to the task of classification of poetry by the research community. In this context, this work aimed to classify
poetry using machine learning and deep learning models and to analyze the performance of the algo-rithms. To perform this task, poetry corpus is categorized into multiple classes using Latent Dirichlet Allocation a topic modeling technique. The classification task is carried using Multinomial Bayesian,
SGD models under machine learning methods and LSTM, Bi-LSTM and CNN models under the deep learn-ing methods. The results are evaluated with parameter accuracy. As a result of this experiment the best classification accuracy is achieved using CNN model with 87% by outperforming other models. This shows that for literary text classification CNN can be considered as a best classifier in comparative with the LSTM and Bi-LSTM models
An Epidemic Graph's Modeling Application to the COVID‐19 Outbreak: Concepts, Methodologies, Tools and Applications
No full textThe furious disease named COVID-19 is an outbreak in the current scenario. To control the spreading of this disease, new models were developed which utilized established methodologies to analyze how different containment strategies will influence the spread of the virus. It presents a novel machine learning approach that can estimate any epidemiological model's parameters based on two types of information: either static or dynamic. It primarily utilizes the Graph model using deep learning approaches and Long-term memories (LSTMs) to obtain mobility data's spatial and temporal properties of SIR and SIRD models. It runs and simulates using data on the Italian COVID dynamics and compares the model predictions to previously observed epidemics
Cadmium chloride elicitation of Abutilon indicum cell suspension cultures for enhanced stigmasterol production
No full textAbutilon indicum (Malvaceae), a therapeutically valuable shrub can act as a continuous source of stigmasterol, accredited with pharmacological significance. In the present study, the content of stigmasterol when analyzed in both in vivo and in vitro plants was found to be 13.89 ± 1.43 and 20.50 ± 2.34 µg/gFW, respectively. The callus obtained from the in vitro plants of A. indicum was found to contain 10.78 ± 0.19 µg/gFW of stigmasterol and was used for initiation of suspension cultures. In comparison to the calli, suspension cultures of A. indicum accumulated considerable amounts of stigmasterol (16.08 ± 1.92 µg/gFW) on the 12th day, i.e., end of log phase. The suspensions on further elicitation with Cadmium Chloride have shown a significant increase (2.59-fold) in the amount of stigmasterol compared to the initial calli, reaching 41.73 ± 3.77 µg/gFW. Thus, cell suspensions of A. indicum offer a unique advantage for large-scale production of stigmasterol under in vitro conditions, by retaining its natural essence and safety in human consumption
A prospective removal approach of Reactive Yellow 14 (RGB) dye using Spongomorpha indica
No full textThe current study focuses on the removal of Reactive Yellow 14 (RY 14) (RGB) from synthetic medium using Spongomorpha indica. Batch experiments were conducted, and the data were applied to isotherms, kinetic, thermodynamic studies. FTIR, SEM/EDX analysis was used to characterize geochemical and surface properties of Spongomorpha indica . Langmuir isotherm and Pseudo second order kinetic model have been observed to be the best-fitted models from the experimental batch adsorption study. The maximum adsorption capacity was observed as 87.71 mg g -1
by Spongomorpha indica for the removal of RY14 dye. For negative values of ΔG° and ΔH°, the thermodynamic assessment of the batch removal process indicates the spontaneity along with exothermic nature of the system. Subsequently, Spongomorpha indica showed higher potential for removing RY 14 from aqueous
medium
Ionic Mass Transfer at Point Electrodes Located at Cathode Support Plate in an Electrorefining Cell in Presence of Rectangular Turbulent Promoters
No full textCurrent density plays a major role in deciding the plant size, current efficiency, and energy consumption in electrorefining cells. In general, operating current density will be 40% of the limiting current density. Forced circulation of the electrolyte in the presence of promoters improves the mass transfer coefficient. In the present study, rectangular turbulence promoters are fitted at the bottom side of the cell to improve the mass transfer coefficient at the cathode support plate. The limiting current density technique is used to measure the mass transfer coefficient. The variables covered in the present study are the effects of flow rate, promoter height, and spacing among the promoters. The electrolyte consists of copper sulfate and sulphuric acid. At a regulated flow rate, the electrolyte is pumped from the recirculation tank to the cell through an intermediate overhead tank. The limiting current density increased with an increasing flow rate in the presence of promoters, and thus the overall mass transfer coefficient on the cathode support plate also improved. With an increase in the flow rate of the electrolyte from 6.67 × 10−6 to 153.33 m3/s, limiting current density increased from 356.8 to 488.8 A/m2 for spacing of 0.30 m, with a promoter height of 0.01 m. However, it is noteworthy that when the promoter height is increased from 0.01 to 0.07 m, the overall mass transfer coefficient is found to increase up to 60%, but with the further increase in the promoter height to 0.30 m the mass transfer coefficient starts to decrease. Therefore, the optimized cell parameters are established in this work. The current sustainable concept of employing rectangular turbulence promoters will bring benefits to any precious metal refining or electrowinning tank house electrolytes
Influence of processing variables on tensile strength and water absorption of natural fibres hybrid composites
Get PDFNatural fibers are gaining popularity among researchers and academics for use in polymer composites due to their environmental friendliness and long-term viability.A new series of green composites using chicken feather fiber (CFF) and jute as reinforcing materials in a polypropylene resin-based polymer matrixwere used to fabricate the hybrid composites. This paper examines the impact of molding pressure, temperature, and time on the tensile strength and water absorption characteristics of compression molded hybrid composites. The findings of experiments undertaken to evaluate hybrid composites produced with different fiber fractions and processing parameters are discussed in this paper.When the composites’ overall tensile strength was compared, the 100% Jute fiber composite had the highest tensile strength when the maximum pressure, temperature, and time were kept constant. Temperature and pressure have a considerable impact on composite sample tensile strength, whereas time has little influence.As predicted, raising the jute weight percentage in composites increases water absorption