Browse Items (9795 total)
Sort by:
-
Portrait segmentation using ensemble of heterogeneous deep-learning models
Image segmentation plays a central role in a broad range of applications, such as medical image analysis, autonomous vehicles, video surveillance and augmented reality. Portrait segmenta-tion, which is a subset of semantic image segmentation, is widely used as a preprocessing step in multiple applications such as security systems, entertainment applications, video conferences, etc. A substantial amount of deep learning-based portrait segmentation approaches have been developed, since the performance and accuracy of semantic image segmentation have improved significantly due to the recent introduction of deep learning technology. However, these approaches are limited to a single portrait segmentation model. In this paper, we propose a novel approach using an ensemble method by combining multiple heterogeneous deep-learning based portrait segmentation models to improve the segmentation performance. The Two-Models ensemble and Three-Models ensemble, using a simple soft voting method and weighted soft voting method, were experimented. Intersection over Union (IoU) metric, IoU standard deviation and false prediction rate were used to evaluate the performance. Cost efficiency was calculated to analyze the efficiency of segmentation. The experiment results show that the proposed ensemble approach can perform with higher accuracy and lower errors than single deep-learning-based portrait segmentation models. The results also show that the ensemble of deep-learning models typically increases the use of memory and computing power, although it also shows that the ensemble of deep-learning models can perform more efficiently than a single model with higher accuracy using less memory and less computing power. 2021 by the authors. Licensee MDPI, Basel, Switzerland. -
Selfie Segmentation in Video Using N-Frames Ensemble
Many camera apps and online video conference solutions support instant selfie segmentation or virtual background function for entertainment, aesthetic, privacy, and security reasons. A good number of studies show that Deep-Learning based segmentation model (DSM) is a reasonable choice for selfie segmentation, and the ensemble of multiple DSMs can improve the precision of the segmentation result. However, it is not fit well when we apply these approaches directly to the image segmentation in a video. This paper proposes an N-Frames (NF) ensemble approach for a selfie segmentation in a video using an ensemble of multiple DSMs to achieve a high-performance automatic segmentation. Unlike the N-Models (NM) ensemble which executes multiple DSMs at once for every single video frame, the proposed NF ensemble executes only one DSM upon a current video frame and combines segmentation results of previous frames to produce the final result. For the experiment, we use four state-of-the-art image segmentation models to make an ensemble. We evaluated the proposed approach using 81 videos dataset with a single-person view collected from publicly available websites. To measure the performance of segmentation models, Intersection over Union (IoU), IoU standard deviation, false prediction rate, Memory Efficiency Rate and Computing power Efficiency Rate parameters were considered. The average IoU values of the Two-Models NM ensemble, Two-Frames NF ensemble, Three-Models NM ensemble and Three-Frames NF ensemble were 95.1868%, 95.1253%, 95.3667% and 95.1734% each, whereas the average IoU value of single models was 92.9653%. The result shows that the proposed NF ensemble approach improves the accuracy of selfie segmentation by more than 2% on average. The result of cost efficiency measurement shows that the proposed method consumes less computing power like single models. 2021 IEEE. -
A Study on the Effect of Canny Edge Detection on Downscaled Images
Abstract: Nowadays user devices such as phones, tablets etc. allows processing the images with help of high-end applications and softwares developed. Most of the times, the images are downscaled to make them compatible with these end devices. This leads to the loss of image quality. This loss of information on downscaling an image results in distortion of edges and while zoomed in results into a blurred image. As the edge detection is a basic step for many image processing applications such as object detection, object segmentation, object recognition, etc. It is necessary to know the impact of edge detection on downscaled image. In this paper, we are using Canny Edge detection method to detect the edges. The original images are downscaled using different interpolation methods. Canny Edge detection is applied on original images and downscaled images to compare the distortion in the edges. We used Structural Similarity Index Method for comparison. We are also comparing execution time taken by Canny Edge Detection on different interpolation methods to check for optimal interpolation method. We observed that the distortion in edges and time efficiency differ for different interpolation methods which are detailed below in the result section. As blurring is also a disadvantage of downscaling, we are applying Gaussian Blur on the images to compare the blurring due to Gaussian blur technique and blurring due to downscaling. 2020, Pleiades Publishing, Ltd. -
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. -
Customers satisfaction towards online banking services of public sector banks
At present the banking industry around the world has been undergoing a rapid transformation. The deepening of information technology has facilitated better tracking and fulfillment of commitments, multiple delivery channels for online customers and faster resolution of issues. Customer satisfaction is important criteria for banks sustenance, now banks are offers online banking services according to the customer needs and requirements. This study analysed customers satisfactions towards online banking services of public sector banks in Tiruchirappalli district. It is understand from the present study that bank websites and technology platforms has to offer various knowledge features on financial services. To retain the existing customers, banks has to conduct regular surveys on the customer satisfaction. The results of the study shows that variables like prompt response, security and Website design and ease of use are top three factors affected customer satisfaction. IJSTR 2020. -
The role of business development management consultants in construction industry in India
The Indian construction industry is in boom in terms of the growth and development taking place in the field of infrastructure and the service sector and subsequently many public owned design, architecture, construction and engineering companies were set up in India. The current study emphasises on the role of business development management consultants in the overall performance of construction companies and the existing review of literature focuses on the various factors which affects the overall business performance of the construction companies which includes different variables considering all the existing studies this study identified the gap in terms of intervention of BDMC in the construction field. The methodology adopted for the study is quantitative exploratory research and the results of the study was very positive that the promoters believe that the business development management consultants contribute a lot in the overall performance of the construction company with their expertise in the field and professional way of dealing with the other stake holders of the company helps them to perform very effectively and efficiently. This study helped in getting the practical exposure to the working of construction companies and understanding the role of business development management consultants. The Author(s) 2019. -
Effectiveness of information and communication technologies on the learning outcomes of management graduates in public and private sector institutes
In most of the countries, information and communication technology (ICT) has a made a strong impact on the development of educational sector and it is considered to be the strongest pedagogy in delivering the communication effectively. This is the era of Information and Communication and to increase the efficiency and effectiveness of Education sector the role of ICT is indispensable. The performance of the students can be made better with the help of information and communication technology (ICT). ICT will help the students to gather the information very effectively and help in in developing their knowledge skills as well as to improve their learning skills. To understand the effectiveness and also the impact of ICT in Education sector especially in management education, for the current study the data is been collected from 158 respondents who are MBA graduates from 15 B-schools from Bangalore city, the study is being conducted with the help of convenient sampling from various B-schools from Bangalore city. The study has shown that the learning and teaching through ICT improves the knowledge and learning skills of students. This indicates that existence of ICT is improving the educational efficiency among the management graduates in Bangalore city. 2019, Institute of Advanced Scientific Research, Inc. All rights reserved. -
Optimization of Biodiesel Production from Waste Cooking Oil by Box Behnken Design Using Response Surface Methodology
Interest in Biodiesel production has grown over the years due to concerns related to the environment, and the solutions include deriving energy from waste as the replacement for diesel, a petroleum-derived fuel. Biodiesel has been accepted as a "green fuel" as it is a renewable, non-toxic, safe and biodegradable energy material. The utilisation of waste cooking oil (WCO) by converting it into biodiesel is one of the promising alternatives to diesel. An attempt to optimise the biodiesel production from WCO (a waste material) has been made via this study. The process adopted was Trans-esterification of pretreated WCO, and the optimization of biodiesel production was carried out by Box-Behnken method using a response surface methodology. The variations between the analytical and experimental results were within acceptable limits. The response surface methodology resulted in an optimum yield of 96.88% (analytical), which was validated through an experiment within an acceptable error of 0.58%. 2021,International Journal Of Renewable Energy Research.All rights reserved. -
Maximised bioethanol extraction from bamboo biomass through alkali pretreatment and enzymatic saccharification by application of ANN-NSGA-II-based optimisation method
The demand for alternative fuels is growing due to the depletion of fossil fuel resources. Non-edible resources are explored as alternatives, and a bamboo is an up-and-coming option for producing ethanol. The extraction process for bioethanol from bamboo involves alkali pretreatment, enzymatic saccharification, and ethanol production. The bamboo biomass is treated with alkali at high temperatures and pressure. This treatment helps break the lignin bonds that hinder the reaction between cellulose and enzymes. As a result, the pretreated biomass contains 40% less lignin than its raw form. Next, the air-dried pretreated biomass undergoes saccharification using Supercut Acid Cellulose. The saccharification process is optimised to achieve the shortest possible time, determined through prediction models based on artificial neural networks and optimisation techniques like Non-dominated Sorting Genetic Algorithm-II. The optimised process involves specific biomass and enzyme loading, producing reducing sugars estimated using the DNS method. Following enzymatic Saccharification, the hydrolysate is fermented using Saccharomyces cerevisiae, a type of yeast. This fermentation process yields ethanol with a 1614.26mg/kg concentration. 2023, The Author(s), under exclusive licence to Springer Nature B.V. -
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. -
Value co-creation through search efforts and customer involvement impacting purchase intention of smart phones
A marketing strategy which successfully involves its customer helps in stimulating purchase intentions. Understanding the behavioral aspects of customers become pertinent in formulating such strategies. The aim of this paper is to explore the underlying factors of customer involvement in value co-creation and discover how it affects the purchase intention of the customers towards smartphones. The study also tries to understand the contribution of search efforts towards customer involvement and how it affects purchase intention. The data for the study has been collected through a validated questionnaire from 233 respondents. Extensive literatures are reviewed to identify research gap and identify the variables for the study. The study can help marketers to identify the factors of customer involvement so that they can understand the customer purchase behaviour better and hence forecast on customer purchase intention to improve their sales of smartphones. BEIESP. -
Evaluating forces associated with sentient drivers over the purchase intention of organic food products
The study proposes to find out the factors which influence awareness among the consumers towards purchasing organic food product. The study is based on primary data by using tools Chi-square test, Cronbach alpha, KMO, and Bartlett's test, ANOVA, regression, correlation, and cross-tabulation. The study found that awareness driver's nutritional information, price, certification, brand name, and logos have an essential influence on the purchase intention of the product of organic food. However, labeling and food standards do not show a noteworthy rapport between labeling and organic food products' purchase plans. The core commitment and flow to explore are to analyze purchasers with respect to organic guarantee systems (accreditation, guidelines, logo, imprints, and confirmation) so we can distinguish the genuine organic products. The independent factors of awareness like organic buying preference and buying frequency, have a significant influence on the purchase intention of organic food. The research provided evidence of consumer awareness and purchase intention of organic food that would help the organic food industry to promote their products according to the attribute of customers. 2020 Asian Economic and Social Society. All rights reserved. -
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. -
Alpha Decay Favoured Isotopes of Some Superheavy Nuclei: Spontaneous Fission Versus Alpha Decay
Romanian Journal of Physics, Vol-57 (9-10), pp. 1335-1345. -
Evolution of primordial dark matter planets in the early Universe
In a recent paper we had discussed possibility of DM at high redshifts forming primordial planets composed entirely of DM to be one of the reasons for not detecting DM (as the flux of ambient DM particles would be consequently reduced). In this paper we discuss the evolution of these DM objects as the Universe expands. As Universe expands there will be accretion of DM, helium and hydrogen layers (discussed in detail) on these objects. As they accumulate more and more mass, the layers get heated up leading to nuclear reactions which burn H and He when a critical thickness is reached. In the case of heavier masses of these DM objects, matter can be ejected explosively. It is found that the time scale of ejection is smaller than those from other compact objects like neutron stars (that lead to x-ray bursts). These flashes of energy could be a possible observational signature for these dense DM objects. 2021 COSPAR -
Alpha decay favoured isotopes of some superheavy nuclei: Spontaneous fission versus alpha decay
Spontaneous fission and alpha decay are the main decay modes for superheavy nuclei. The superheavy nuclei which have small alpha decay half-life compared to spontaneous fission half-life will survive fission and can be detected in the laboratory through alpha decay. We have studied the alpha decay half-life and spontaneous half-life of some superheavy elements in the atomic range Z = 100-130. Spontaneous fission half-lives of superheavy nuclei have been calculated using the phenomenological formula and the alpha decay half-lives using Viola-Seaborg-Sobiczewski formula (Sobiczewski et al. 1989), semi empirical relation of Brown (1992) and formula based on generalized liquid drop model proposed by Dasgupta-Schubert and Reyes (2007). The results are reported here. -
Primordial Planets with an Admixture of Dark Matter Particles and Baryonic Matter
It has been suggested that primordial planets could have formed in the early universe and the missing baryons in the universe could be explained by primordial free-floating planets of solid hydrogen. Many such planets were recently discovered around the old and metal-poor stars, and such planets could have formed in early epochs. Another possibility for missing baryons in the universe could be that these baryons are admixed with DM particles inside the primordial planets. Here, we discuss the possibility of the admixture of baryons in the DM primordial planets discussed earlier. We consider gravitationally bound DM objects with the DM particles constituting them varying in mass from 20 to100 GeV. Different fractions of DM particles mixed with baryonic matter in forming the primordial planets are discussed. For the different mass range of DM particles forming DM planets, we have estimated the radius and density of these planets with different fractions of DM and baryonic particles. It is found that for heavier-mass DM particles with the admixture of certain fractions of baryonic particles, the mass of the planet increases and can reach or even substantially exceed Jupiter mass. The energy released during the process of merger of such primordial planets is discussed. The energy required for the tidal breakup of such an object in the vicinity of a black hole is also discussed. 2023 by the authors. -
Is asteroid 33 Polyhymnia a dark matter (DM) degenerate object?
Polyhymnia (33 Polyhymnia) is a main belt asteroid in our solar system with a diameter around 54km. The density of asteroid 33 Polyhymnia, located in the main asteroid belt, is calculated to be 75g/cc. Researchers have speculated the possibility that Polyhymnia could be composed of high-density superheavy elements near atomic number 164. Here, we propose that Polyhymnia could be an asteroid composed of degenerate dark matter (DM) and there could be many such asteroids in our solar system. (This is following our earlier work suggesting that Planet Nine could be such an object.) The Author(s), under exclusive licence to SocietItaliana di Fisica and Springer-Verlag GmbH Germany, part of Springer Nature 2024. -
Calculator using brain computer interface
This paper is undertaken with an aspiration to provide a new way to calculate that can be availed by exploiting BCI. Often it's said things moved at the blink of eye, now the time has come to make it true. This project is developed to ease the efforts in two different ways. First a mind controlled image viewer is build which can be used to change images at the blink of an eye. Second is a simple single digit calculator which lets the user choose the number and the operators just by focusing.Brain-Computer Interface aims to improve the detection and decoding of brain signals acquired by electroencephalogram (EEG). IAEME Publication.