Browse Items (2150 total)
Sort by:
-
Accuracy Enhancement of Portrait Segmentation by Ensembling Deep Learning Models
Portrait segmentation is widely used as a preprocessing step in multiple applications. The accuracy of portrait segmentation models indicates its reliability. In recent times, portrait segmentation using deep learning models have achieved significant success in performance and accuracy. However, these portrait segmentation models are limited to a single model. In this paper, we propose ensemble approach using multiple portrait segmentation models to improve the segmentation accuracy. The result of experiment shows that the proposed ensemble approach produces better accuracy than individual models. Accuracy of single models and proposed ensemble approach were compared with Intersection over Union (IoU) metric and false prediction rate to evaluate the accuracy performance. The result shows reduced false negative rate and false discovery rate, this reduction in false prediction has enabled ensemble approach to produce segmented images with optimized error and improved result of segmentation in portrait area of human body than individual portrait segmentation models 2020 IEEE. -
Analyzing the Performance of Canny Edge Detection on Interpolated Images
Surveillance cameras are extensively used nowadays in many commercial and domestic places to monitor theft, intrusion and other illegal activities. Typically, the cameras are placed at a high position to monitor a large area. Therefore, the captured images include background area in addition to the target objects. Under such situation, the image can be zoomed to focus on the target objects using various interpolation techniques. For further processing of the image, many techniques like edge detection, image sampling and image thresholding etc. are available. Considering edge detection to be a basic step for many application such as Object detection, Object recognition etc, in this work, we analyze the performance of the Canny Edge Detection algorithm on images interpolated using Nearest Neighbour, Bilinear and Bicubic interpolation methods. Canny Edge Detection is applied to the input image and the resultant image is saved for later comparison. The same image is upscaled using interpolation and the Canny Edge Detection algorithm is used on this upscaled image. This image is then resized to the original image size. Both the images are compared to check for their similarity using the Structural Similarity Index Method (SSIM). 2019 IEEE. -
Verification and validation of Parallel Support Vector Machine algorithm based on MapReduce Program model on Hadoop cluster
From the recent years the large volume of data is growing bigger and bigger. It is difficult to measure the total volume of structured and unstructured data that require machine-based systems and technologies in order to be fully analyzed. Efficient implementation techniques are the key to meeting the scalability and performance requirements entailed in such scientific data analysis. So for the same in this paper the Sequential Support Vector Machine in WEKA and various MapReduce Programs including Parallel Support Vector Machine on Hadoop cluster is analyzed and thus, in this way Algorithms are Verified and Validated on Hadoop Cluster using the Concept of MapReduce. In this paper, the performance of above applications has been shown with respect to execution time/training time and number of nodes. Experimental Results shows that as the number of nodes increases the execution time decreases. This experiment is basically a research study of above MapReduce applications. 2013 IEEE. -
Forecasting Prices of Black Pepper in Kerala and Karnataka using Univariate and Multivariate Recurrent Neural Networks
Our country has a high level of agricultural employment. Price swings harm the economy of our country. To combat this impact, forecasting the selling price of agricultural products has become a need. Forecasts of agricultural prices assist farmers, government officials, businesses, central banks, policymakers, and consumers. Price prediction can then assist in making better selections in this area. Black pepper, sometimes known as the "King of Spices, " is a popular spice farmed and exported in India. The largest producers of black pepper are Karnataka and Kerala. For black pepper in Kerala and Karnataka, this study provides a univariate and multivariate price prediction model using Recurrent Neural Networks (RNN), Long Short-Term Memory (LSTM), and Gated Recurrent Unit (GRU). The data is denoised using Singular Spectral Analysis (SSA). The most accurate method is the multivariate variate LSTM technique, which uses macroeconomic variables. It has a Mean Absolute Percentage Error (MAPE) of 0.012 and 0.040 for Kerala and Karnataka, respectively. Grenze Scientific Society, 2022. -
Anonymization Based Deep Privacy Preserving Convolutional Autoencoder Learning Technique for High Dimensional Data Clustering in Big Data Cloud
Data Clustering is a primary research focus in large data-driven application domains in the big data cloud as performance of the clustering dynamic data with high dimensionality is highly challenging due to major concern in the effectiveness and efficiency on data representation. Machine learning is a conventional approach to distribute the data into soft partition still it leads to increasing sparsity of data and increasing difficulties in distinguishing distance between data points. In addition, securing the personnel and confidential information of the user is also becoming vital. In order to tackle those issues, a new anonymization based deep privacy preserving learning paradigm has been presented in this paper. The proposed model is represented as deep privacy preserving convolutional auto encoder learning architecture for secure high dimensional data clustering on inferring the distribution of the data over time. Initially dimensionality reduction and feature extraction is carried out and those extracted feature has been taken for clustering on generation of objective function to produce maximum margin cluster. Those clusters are further fine tuned to feature refinement on the hyper parameter of various layers of deep learning model network to establish the minimum reconstruction error by feature refinement. Softmax layer minimizes the intra cluster similarity and inter cluster variation in the feature space for cluster assignment. Hyper parameter tuning using stochastic gradient descent has been enabled in the output layer to make the data instance in the cluster to be close to each other by determining the affinity of the data on new representation. It results significant increase in the clustering performance on the discriminative informations. Detailed experimental analysis has been performed on benchmarks datasets to compute the proposed model performance with conventional approaches. The performance outcome represents that anonymization based deep privacy preserving clustering learning architecture can produce good scalability and effectiveness on high dimensional data. 2023 American Institute of Physics Inc.. All rights reserved. -
FADA: Flooding Attack Defense AODV Protocol to counter Flooding Attack in MANET
The intrinsic nature of a Mobile Ad hoc Network (MANET) makes it difficult to provide security and it is more vulnerable to network attacks. Denial of Service (DoS) attack can be executed using Flooding attack, that has the potential to bring down the entire network. This attack works by delivering an excessive number of unwanted packets that consumes too much battery life, storage space, and bandwidth, that eventually lowers the system's performance. In order to flood the network, the attacker injects fake packets into it. Both Control Packet flooding and Data flooding attacks are taken into account in this study. FADA (Flooding Attack Defense AODV) protocol is proposed to counter flooding attack that promotes greater utilization of existing resources. This research identifies the attack-causing node, isolates it and protects the network against flooding attack. Attack Detection Rate, Attack Detection Accuracy, End-to-end delay and Throughput are few metrics used for evaluation of the proposed model. NS-2.35 is used to demonstrate the efficiency of the suggested protocol and the results prove that the proposed model increases system's throughput while decreasing attack. The simulation results have shown that the proposed FADA protocol performs better than the existing models taken into consideration. 2023 IEEE. -
Investigation on Preserving Privacy of Electronic Medical Record using Split Learning
Artificial Intelligence is deployed in multiple areas, including healthcare. Utmost research is done in AI enabled healthcare industry because of the demands like accurate result, data security, exact prediction, huge volume of data, etc. In conventional deep learning models, the training happens with the dataset that are stored in a single device. This requires a huge storage space and highly efficient machines to train the data. Usage of big data, demands for innovative models that can be deployed and used in confined storage. Split learning is one such collaborative distributed deep learning model that allows the data to be stored in a split fashion. Split learning supports desirable features like less storage, more privacy to raw data, ability to work with resource constraints, etc., making it suitable for storing electronic medical record of patients. This paper discusses the advantages of using split learning for healthcare, the possible configurations of split learning that supports data privacy in healthcare and finally discusses the open research challenges in implementing split learning for healthcare. 2024 The Authors. Published by Elsevier B.V. -
Efficient Routing Strategies for Energy Management in Wireless Sensor Network
Wireless Sensor Network (WSN) refers to a group of distributed sensors that are used to examine and record the physical circumstances of the environment and coordinate the collected data at the centre of the location. This WSN plays a significant role in providing the needs of routing protocols. One of the important aspects of routing protocol in accordance with Wireless Sensor Network is that they should be efficient in the consumption of energy and have a prolonged life for the network. In modern times, routing protocol, which is efficient in energy consumption, is used for Wireless Sensor Network. The routing protocol that is efficient in energy consumption is categorized into four main steps: CM Communication Model, Reliable Routing, Topology-Based Routing and NS Network Structure. The network structure can be further classified as flat/hierarchical. The communication model can be further classified as query, coherent/non-coherent, negotiation-based routing protocol system. The topology-based protocol can be further classified as mobile or location-based. Reliable routing can be further classified as QoS (Quality of service) or multiple-path based. A survey on routing protocol that is energy-efficient on Wireless Sensor Network has also been provided in this research. The Author(s), under exclusive license to Springer Nature Switzerland AG 2022. -
Twitter Hate Speech Detection using Stacked Weighted Ensemble (SWE) Model
Online Social Media has expanded the freedom of expression in the internet, which has become a disturbing problem if it has an impact on the situation or the interest of a country. Hate speech refers to the use of hostile, abusive or offensive language, directed at a certain group of people who share common property, whether it is their gender, ethnicity or race (i.e. racism), faith and religion. Therefore, auto detection of hate speeches has an increased importance in Online Social Media for filtering any message that has hatred language before posting it to the network. In this paper, a Stacked Weighted Ensemble (SWE) model is proposed for the detection of hate speeches. The model ensembles five standalone classifiers: Linear Regression, Nae Bayes', Random Forest, Hard Voting and Soft Voting. The experimental results on a Twitter dataset has shown an accuracy of 95.54% in binary classification of tweets into hateful speech and an improved performance is noted compared to the standalone classifiers. 2020 IEEE. -
Curvature Ductility of Reinforced Masonry Walls and Reinforced Concrete Walls
Research conducted in this work proposes an equation to evaluate and compares the curvature ductility of reinforced masonry (RM) and reinforced concrete (RC) walls. The curvature ductilities are measured at varying levels of axial stresses for walls for aspect ratio (l/h) of 0.5, 1.0 and 1.5. The percentage of reinforcement is increased from 0.25% (minimum reinforcement for RC walls as per IS-13920) to 1.00%. The curvature ductilities are evaluated by plotting flexural strength (M) versus curvature (?) for the walls. The stressstrain curves of masonry, concrete and reinforcing steel are all adopted from existing literature. The compressive strength of masonry and concrete has been chosen as 10MPa and 25MPa, respectively. The yield strength of the steel is fixed as 415MPa. The height and thickness of the wall are 3000 and 230mm, respectively, and the length of the wall is varied to obtain different aspect ratios. Results obtained from this paper imply due to increase curvature ductility, RM walls provide a better alternative for the construction of structural walls compared to RC walls in regions of significant seismicity. 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. -
Lateral Load Behavior of Unreinforced Masonry Spandrels
Spandrels, are usually classified as secondary elements and even though their behaviour has not received adequate focus unlike piers, they significantly affect the seismic capacity of the structure. Masonry spandrels are often damaged and the first structural components that crack within Unreinforced Masonry structures. Despite this, existing analytical methods typically consider a limit case in which the strength of spandrels is either neglected, considered to be infinitely rigid and strong or treated as rotated piers. It is clearly evident that such an assumption is not plausible. Hence, reliable predictive strength models are required. This thesis attempts to re-examine the flexural behaviour of spandrels and proposes an analytical model. The model is based on the interlocking phenomena of the joints at the end-sections of the spandrel and the contiguous masonry. The proposed analytical model is incorporated within a simplified approach to account for the influence of spandrel response on global capacity estimate of URM buildings. 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG. -
Efficient Disease Detection in Wheat Crops: A Hybrid Deep Learning Solution
Wheat rust disease poses a significant danger to global food security and requires rapid, precise diagnosis to be effectively managed. Using a hybrid deep learning (DL) model consisting of a convolutional neural network (CNN) and a decision tree (DT), a new method for classifying wheat rust illness across six magnitude scales has been described in the proposed study. For training and assessing the model, a dataset of 50,000 wheat leaf photos representing a wide range of disease magnitude has been amazing. The suggested work developed a hybrid CNN-DT model with an amazing overall accuracy of 93.47% by carefully analyzing the data and crafting the model. The model's resilience in identifying multiple levels of disease magnitude was proved by the performance metrics for each disease magnitude class. The proposed hybrid model also outperformed state-of-the-art models in terms of accuracy, as shown by the comparisons conducted. The findings provide important new information on the potential of DL methods for wheat rust disease classification, which can then be used as a trusted resource for early disease diagnosis and smarter agricultural policymaking. In the face of agricultural diseases, the suggested model has important implications for improving crop management, reducing yield losses, and guaranteeing food security. 2023 IEEE. -
Safety of Unmanned Systems
The safety risk management process describes the systematic application of management policies, procedures and practices to the activities of communicating, consulting, establishing the context, and assessing, evaluating, treating, monitoring and reviewing risk. This process is undertaken to provide assurances that the risks associated with the operation of unmanned aircraft systems have been managed to acceptable levels. Active efforts should be made to develop rules to ensure the safe operation of unmanned aerial vehicles. For the safe integration of operations with unmanned aerial vehicles, it is important to take into account the influence of different levels of control and autonomous capabilities, as well as the source of movement monitoring in the system. This article discusses the security issues of unmanned systems, the main directions of ensuring the information security of unmanned systems, software and hardware vulnerabilities have been identified. The methods of information protection are given, the disadvantages are indicated. The Author(s), under exclusive license to Springer Nature Switzerland AG 2024. -
Digitalization of Online Classes Among Higher Secondary Students in the Emerging Shift of Post Covid-19 (Second Wave)
The second wave of COVID-19 in India has left higher secondary school students befuddled, unhappy, and unsure about their future. During the second wave of the COVID-19 epidemic, a number of factors influence the effectiveness of online learning. Hence, the main objective of this research paper is focused on understanding the factors influencing online learning among higher secondary students. Researchers identified variables such as attitude, tools and technology, and quality of teaching and social support through extensive literature review. The research study adopted snowball sampling technique and used a survey-based online questionnaire for collecting the data; responses were obtained from 394 respondents from the state of Kerala in India. PLS-SEM was used to test the proposed hypotheses. The results of the study indicate that quality of teaching is the only factor that impacts the effectiveness of online classes among higher secondary students. Attitude, technology and tools, and social support are observed to have insignificant impact on online learning effectiveness. 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. -
The Pendant Number ofLine Graphs andTotal Graphs
The parameter, pendant number of a graph G, is defined as the least number of end vertices of paths in a path decomposition of the given graph and is denoted as ? p(G). This paper determines the pendant number of regular graphs, complete r-partite graphs, line graphs, total graphs and line graphs of total graphs. We explore the bougainvillea graphs, e-pendant graphs and v-pendant graphs. If the pendant number is 2, then the number of paths in the path decomposition of the given graph is at most ? (G), the maximum degree of the graph. Hence, a large class of graphs give a more reasonable solution to Gallais conjecture on number of paths in the given path decomposition. 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. -
Research challenges in self-driving vehicle by using internet of things (IoT)
This article summarizes the benefits, safety hazards, and limitations of owning a self-driving vehicle. Finding a way to use an SDV(Self Driving Vehicle) is minimizing the risk for an accident is important for public and road safety. The actual rate of accidents for self-driving vehicles are lower than that for regular vehicles since the total number of miles of self-driving vehicles combined is nowhere close to that of regular fossil-fueled vehicles. Even though there is no proof that self-driving vehicles will not cause accidents, it is important to know that self-driving vehicles weren't the cause in all the cases they have been involved. That is, it will not be purely considered as the machine's mistake. The safety level of self-driving vehicles has been proven to be one of the best and that has led to the number of serious accident-related wounds in self-driving vehicles to remain lower than the standard level. Nevertheless, Internet of Things plays a major role in developing the self-driving vehicle concept. 2021 IEEE. -
Moderation of Income and Age on Customer Purchase Intention of Green Cosmetics in Bangalore
Cracking the code of customer purchase behaviour is a challenge for market researchers as myriad factors interfere. Marketers are puzzled as competitors position a new product category in the market to create demand. Indian public perceived cosmetics composition as blend of healthy chemical extracts. Television commercials portrayed the presence of chemicals in cosmetics as a product performance booster. People attributed chemical presence to superior product performance. Saturated markets witnessed competitors aiming at increased sales with similar commercials. Under pressure to differentiate, the idea of organic cosmetics started. Companies invested heavily on product development, marketing and branding. Expected success was not achieved as buyers measured performance of cosmetics weighing the absence of chemicals. Scepticism on organic level of the products emerged as various brand commercials claimed their respective compositions a true organic product. Fewer studies explained purchase intention of green cosmetics without focus on health consciousness and consumer innovativeness. Product diffusions were strategized on the basis of consumer innovativeness. Health consciousness captured individuals weightage on health and well-being while purchasing a product. This paper explores relationship of health consciousness and consumer innovativeness with purchase intention development conducting exploratory factor analysis, regression analysis and interaction analysis on selected independent variables using dependent variables. The study found both consumer innovativeness and health consciousness leading to development of purchase intention of green cosmetics. Age and income moderated the relationship of consumer innovativeness and purchase intention. 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. -
Impact of ESG Practices on the Firm's Performance: A Longitudinal Study on Emerging Markets
This study investigated the relationship between business performance in emerging markets (BICS countries) and ESG disclosure scores. Overall, it did not find any correlation between different performance indicators and ESG scores. It's interesting to see that higher overall ESG scores were linked to greater share prices and earnings per share (EPS). This implies that businesses with robust ESG policies may ultimately perform better than others. The study emphasises how ESG may help create value and support sustainable corporate success in emerging markets. It highlights how crucial ESG is to investors, companies, and legislators. 2024 IEEE. -
Stability Analysis of AFTI-16 Aircraft by Using LQR and LQI Algorithms
The stability analysis of the dynamical system of linearized plant model of Advanced Fighter Technology Integration (AFTI)-16 aircraft was proposed along with the optimal control methods by applying linear quadratic regulator (LQR) and linear quadratic algorithm (LQI) algorithms. The LQR and LQI algorithms results were compared with state-space model analysis results. The state-space methods like pole placement method, without using the LQR algorithm the negative feedback system were found to be unstable. By the application of LQR and LQI algorithms to the linearized plant AFTI-16 aircraft open-loop system having negative feedback found to be stable. The stability parameters were verified by using MATLAB programming software. The eigenvalues play a key role in finding closed-loop system stability analysis. MIMO dynamical system with state feedback gain matrices is calculated by using MATLAB programming software. 2021, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. -
INDIVIDUAL AND GROUP VARIATIONS IN WAYFINDING AMONG USERS IN AN EDUCATIONAL BUILDING
The effective performance of users in an Educational Building is determined by the available resources and also the environment in which they dwell. Wayfinding is a daily occurrence for every user of an academic institution and this is facilitated through the distinct articulation of different spaces and recognizable circulation systems. The user behavior in a known/unknown building varies as an individual and with a group of individuals. This variation can be observed in an enclosed space and public setting. For an individual, the psychological state could influence navigating within the building whereas, for a group of individuals, the group dynamics could influence each other to navigate. The paper uses mixed methods to understand and assess the individual and group variations in wayfinding. The study was undertaken in a recently constructed School of Architecture at CHRIST University, Bengaluru. The understanding was accomplished with elaborate literature studies and the assessment was through the field observation techniques and surveys carried out with identified users like frequent individuals, new individuals, frequent groups, and new groups.The study tells that for both individuals and groups, the parameters like architectural elements, sensorial qualities, wayfinding behavior, gender, and psychological state influence them in wayfinding. It was also noted that most of the student users prefer shortcuts rather than the formal entance and lobby to navigate the classrooms. Accomplishing easy, comfortable, and efficient wayfinding within an educational building requires effective layout planning. These findings aim to contribute to the detailed understanding of effective layout planning in an educational building and its impact on user behavior for architects and decision-makers. ZEMCH Network.