Browse Items (11810 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. -
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. -
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. -
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. -
High-Movement Human Segmentation in Video Using Adaptive N-Frames Ensemble
A wide range of camera apps and online video conferencing services support the feature of changing the background in real-time for aesthetic, privacy, and security reasons. Numerous studies show that the Deep-Learning (DL) is a suitable option for human segmentation, and the ensemble of multiple DL-based segmentation models can improve the segmentation result. However, these approaches are not as effective when directly applied to the image segmentation in a video. This paper proposes an Adaptive N-Frames Ensemble (AFE) approach for high-movement human segmentation in a video using an ensemble of multiple DL models. In contrast to an ensemble, which executes multiple DL models simultaneously for every single video frame, the proposed AFE approach executes only a single DL model upon a current video frame. It combines the segmentation outputs of previous frames for the final segmentation output when the frame difference is less than a particular threshold. Our method employs the idea of the N-Frames Ensemble (NFE) method, which uses the ensemble of the image segmentation of a current video frame and previous video frames. However, NFE is not suitable for the segmentation of fast-moving objects in a video nor a video with low frame rates. The proposed AFE approach addresses the limitations of the NFE method. Our experiment uses three human segmentation models, namely Fully Convolutional Network (FCN), DeepLabv3, and Mediapipe. We evaluated our approach using 1711 videos of the TikTok50f dataset with a single-person view. The TikTok50f dataset is a reconstructed version of the publicly available TikTok dataset by cropping, resizing and dividing it into videos having 50 frames each. This paper compares the proposed AFE with single models and the Two-Models Ensemble, as well as the NFE models. The experiment results show that the proposed AFE is suitable for low-movement as well as high-movement human segmentation in a video. 2022 Tech Science Press. All rights reserved. -
-
Powerlessness in the moral self: a social cognitive perspective on drug users
Powerlessness resides in devalued self-images of drug users. This study, drawing on social and moral psychology, examined the moral functioning of drug users compared to non-drug users. Self-reported data concerning moral identity and moral judgment on drug use were assessed and compared between groups. Drug users appeared to have significantly weaker moral identity centrality and pro-drug moral judgment than non-drug users. They also showed dissociation in the relationship between moral identity and moral judgment. As a result, the study proposed a moral identity model of drug use to better approach social cognitive powerlessness in drug users moral self. 2021 Taylor & Francis Group, LLC. -
Low-frequency pulse-jitter measurement with the uGMRT I: PSR J0437-4715
High-precision pulsar timing observations are limited in their accuracy by the jitter noise that appears in the arrival time of pulses. Therefore, it is important to systematically characterise the amplitude of the jitter noise and its variation with frequency. In this paper, we provide jitter measurements from low-frequency wideband observations of PSR J0437 4715 using data obtained as part of the Indian Pulsar Timing Array experiment. We were able to detect jitter in both the 300-500 MHz and 1 260-1 460 MHz observations of the upgraded Giant Metrewave Radio Telescope (uGMRT). The former is the first jitter measurement for this pulsar below 700 MHz, and the latter is in good agreement with results from previous studies. In addition, at 300-500 MHz, we investigated the frequency dependence of the jitter by calculating the jitter for each sub-banded arrival time of pulses. We found that the jitter amplitude increases with frequency. This trend is opposite as compared to previous studies, indicating that there is a turnover at intermediate frequencies. It will be possible to investigate this in more detail with uGMRT observations at 550-750 MHz and future high-sensitive wideband observations from next generation telescopes, such as the Square Kilometre Array. We also explored the effect of jitter on the high precision dispersion measure (DM) measurements derived from short duration observations. We find that even though the DM precision will be better at lower frequencies due to the smaller amplitude of jitter noise, it will limit the DM precision for high signal-to-noise observations, which are of short durations. This limitation can be overcome by integrating for a long enough duration optimised for a given pulsar. The Author(s), 2024. Published by Cambridge University Press on behalf of Astronomical Society of Australia. -
Design, Training, and Implementation of A New Individualized Education Plan (IEP) Format For Special Educators And Students With Intellectual Disabilities At Selected Special Schools
An individualized Education Plan (IEP) is a multidisciplinary, teamdeveloped plan required for every child receiving special education services. The researcher delved into concerns surrounding Individualized newlineEducation Programs (IEPs) for students with intellectual disabilities. Two significant hurdles were discovered: existing IEPs lacked essential intervention areas, and special education teachers felt inadequately newlineequipped to construct effective plans. newlineThe study tackled these concerns head-on through a multi-pronged approach. Firstly, a meticulous analysis of existing IEPs revealed crucial sections missing from intervention plans, hindering their effectiveness. newlineThis analysis served as the blueprint for crafting a more comprehensive IEP format that addressed the identified gaps and provided a robust framework for intervention. Next, the study focused on empowering special education teachers. Sixty special education teachers certified by the Rehabilitation Council of newlineIndia, participated in training sessions on the new format, undergoing a vital skills and knowledge upgrade in IEP development. This equipped them with the tools and understanding necessary to create more effective plans tailored to individual student needs. The theory then transitioned to practice. Students with intellectual newlinedisabilities were included in interventions based on the improved IEPs, with their progress closely tracked and evaluated. The results were highly promising. Teachers demonstrated a tangible improvement in knowledge, translating into their ability to create more effective IEPs. More importantly, students thrived with the enhanced format. Those involved in interventions using the improved IEPs exhibited significant progress in various domains, highlighting the positive impact of the new approach. The study culminated in key recommendations for further newlineimprovement. Ongoing teacher training sessions were suggested to ensure teachers remain updated on best practices and evolving methodologies. -
-
Inhibiting extracellular cathepsin d reduces hepatic steatosis in spraguedawley rats y
Dietary and lifestyle changes are leading to an increased occurrence of non-alcoholic fatty liver disease (NAFLD). Using a hyperlipidemic murine model for non-alcoholic steatohepatitis (NASH), we have previously demonstrated that the lysosomal protease cathepsin D (CTSD) is involved with lipid dysregulation and inflammation. However, despite identifying CTSD as a major player in NAFLD pathogenesis, the specific role of extracellular CTSD in NAFLD has not yet been investigated. Given that inhibition of intracellular CTSD is highly unfavorable due to its fundamental physiological function, we here investigated the impact of a highly specific and potent small-molecule inhibitor of extracellular CTSD (CTD-002) in the context of NAFLD. Treatment of bone marrow-derived macrophages with CTD-002, and incubation of hepatic HepG2 cells with a conditioned medium derived from CTD-002-treated macrophages, resulted in reduced levels of inflammation and improved cholesterol metabolism. Treatment with CTD-002 improved hepatic steatosis in high fat diet-fed rats. Additionally, plasma levels of insulin and hepatic transaminases were significantly reduced upon CTD-002 administration. Collectively, our findings demonstrate for the first time that modulation of extracellular CTSD can serve as a novel therapeutic modality for NAFLD. 2019 by the authors. -
An empirical analysis of android permission system based on user activities
In today's world there has been an exponential growth among smart-phone users which has led to the unbridled growth of smart-phone apps available in Google play store, app store etc., In case of android application, there are many free applications for which the user need not shell out a penny to use the services. Here the magic word is "free" which entices millions of pliant people into installing those apps and giving unnecessary access to their data and device control. Current studies have shown that over 70% of the apps in market, request to gather data digressive to the most functions of apps that might cause seeping of personal data or inefficient use of mobile resources. Of late, couple of malignant applications gather unobtrusive information of the user through third-party applications by increasing their permissions to high-level on the Android Operating System. Android permission system provides, the user access to the third party apps and in return based on the permissions granted by the user, an app can access the related resource from the user's mobile. A user is bound to grant or deny permits during the installation of the application. For the most part, users don't focus on the asked permissions, or sometimes users do not understand the meaning of the permission and install the app on their device. They allow a way for attackers to perform the malicious task by demanding for more than expected set of permissions. These extra permissions permit the attacker to exploit the device and also retrieve sensitive information from it. In this research paper we describe how permission system security can create an awareness among the users that would assist them in deciding on permission grants. This improved and responsible user activities in Android OS can help the users in utilizing their device securely. 2018 Ankur Rameshbhai Khunt and P. Prabu. -
Complicated Grief during COVID-19: An International Perspective
Cultures across the globe have evolved time-tested rituals to honor those who die and offer solace and support to survivors with the goal of helping them to accept the reality of the death, cope with the feelings of loss, adjust to life without the deceased, and find ways to maintain a connection to the memory of the deceased. The COVID-19 pandemic has disrupted these rituals and brought significant changes to the way we mourn. Specifically, public health responses to COVID-19 such as social distancing or isolation, delays or cancellations of traditional religious and cultural rituals, and shifts from in-person to online ceremonies have disrupted rituals and thus made it more difficult to access support and complete the psychological tasks typically associated with bereavement. This paper conceptualizes the common bereavement tasks including emotion-focused coping, maintaining a connection to the deceased, disengagement and reframing death and loss, and problemfocused coping. It provides examples of how the COVID-19 pandemic has altered mourning rituals across several cultures and religions and contributed to prolonged grief disorder as defined by the ICD-11 that includes depressive symptoms and post-traumatic stress. Early evidence suggested that the suddenness of loss, the social isolation, and the lack of social support often associated with COVID-19-related death are salient risk factors for complicated grief. As a consequence, psychological assessments, grief counseling, and mental health support are needed by families of patients who died from COVID-19. These services must be essential components of any comprehensive public health response to the pandemic. 2022 Hogrefe Publishing. -
Author profiling: Age prediction of blog authors and identifying blog sentiment
Authorship profiling is about finding out different characteristic of an author like age, gender, native languages, education background etc., by finding out the patterns in their writing. Blog authors write about a lot of topics like purchase decisions, digital advertising, personality development, fitness, technology updates etc., and these authors play an influential role on its readers. In this paper, we are categorizing the blog authors in three different age groups based on the content available from the blog. Natural Language Toolkit (NLTK) is a set of libraries used for natural language processing to distinguish among the different writing pattern of the author based on the different age groups. NLTK helps to make analysis on the words of the blogs which is an important feature in our research. We also wanted to conduct sentiment analysis on the blog in order to understand the insight on how the author feels about the blog topic. Thus, we have used Nae Bayes Classifier for doing the analysis and considered two sentiments for the same: positive and negative. An average accuracy of 66.78% was achieved in predicting the age of authors. From the sentiment analysis we figured out that elder authors tend to have more positivity in their blogs as compared to younger authors. 2019, Institute of Advanced Scientific Research, Inc.. All rights reserved. -
Gut Homeostasis; Microbial Cross Talks In Health and Disease Management
The human gut is a densely populated region comprising a diverse collection of microorganisms. The number, type and function of the diverse gut microbiota vary at different sites along the entire gastrointestinal tract. Gut microbes regulate signaling and metabolic pathways through microbial cross talks. Host and microbial interactions mutually contribute for intestinal homeostasis. Rapid shift or imbalance in the microbial community disrupts the equilibrium or homeostatic state leading to dysbiosis and causes many gastrointestinal diseases viz., Inflammatory Bowel Disease, Obesity, Type 2 diabetes, Metabolic endotoxemia, Parkinsons disease and Fatty liver disease etc. Intestinal homeostasis has been confounded by factors that disturb the balance between eubiosis and dysbiosis. This review correlates the consequences of dysbiosis with the incidence of various diseases. Impact of microbiome and its metabolites on various organs such as liver, brain, kidney, large intestine, pancreas etc are discussed. Furthermore, the role of therapeutic approaches such as ingestion of nutraceuticals (probiotics, prebiotics and synbiotics), Fecal Microbial Treatment, Phage therapy and Bacterial consortium treatment in restoring the eubiotic state is elaborately reviewed. 2021 The Author(s). Published by Enviro Research Publishers. -
Quantum Convolutional Neural Network for Medical Image Classification: A Hybrid Model
This study explores the application of Quantum Convolutional Neural Networks (QCNNs) in the realm of image classification, particularly focusing on datasets with a highly reduced number of features. We investigate the potential quantum computing holds in processing and classifying image data efficiently, even with limited feature availability. This research investigates QCNNs' application within a highly constrained feature environment, using chest X-ray images to distinguish between normal and pneumonia cases. Our findings demonstrate QCNNs' utility in classifying images from the dataset with drastically reduced feature dimensions, highlighting QCNNs' robustness and their promising future in machine learning and computer vision. Additionally, this study sheds light on the scalability of QCNNs and their adaptability across various training-test splits, emphasizing their potential to enhance computational efficiency in machine learning tasks. This suggests a possibility of paradigm shift in how we approach data-intensive challenges in the era of quantum computing. We are looking into quantum paradigms like Quantum Support Vector Machine (QSVM) going forward so that we can explore trade offs effectiveness of different classical and quantum computing techniques. 2024 IEEE. -
A study on the perception of MOOC (Massive Open Online Course) amongst the students of Christ University, Bengaluru /
Massive open online courses (MOOC) are a recent innovation in the field of online learning. Several top-tier universities around the world have started offering MOOC programmes in a wide array of professional, technical as well as creative fields. Top MOOC providers such as Coursera, Udacity and edX have a student fellowship from all across the world, pursuing one or more from the thousands of courses offered by these MOOC giants. -
Context Driven Software Development
The Context-Driven Software Development (CDSD) is a novel software development approach with an ability to thrive upon challenges of 21st century digital and disruptive technologies by using its innovative practices and implementation prowess. CDSD is a coherent set of multidisciplinary innovative and best practices like context-aware and self-adaptive system modelling, human-computer interaction, quality engineering, software development-testing-and continuous deployment frameworks, open-source tools-technology-and end-to-end automation, software governance, engaging stakeholders, adaptive solutioning, design thinking, and group creativity. Implementation prowess of CDSD approach stems from its three unique characteristics, namely, its principles, Contextualize-Build-Validate-Evolve (CBVE) product development element, and iterative and lean CDSD life cycle with Profiling, Contextualizing, Modelling, Transforming, and Deploying phases with in-process and phase-end Governance and Compliances. CDSD approach helps to address issues like complexity, software ageing, risks related to internal and external ecosystem, user diversity, and process-related issues including cost, documentation, and delay. 2021, Springer Nature Switzerland AG. -
The Capital structure puzzle
International Journal of Research in Commerce & Management Vol.4, No.03, pp.134-136 ISSN No. 0976-2183 -
Implementation of Supervised Pre-Training Methods for Univariate Time Series Forecasting
There has been a recent deep learning revolution in Computer Vision and Natural Language Processing. One of the biggest reasons for this has been the availability of large-scale datasets to pre-train on. One can argue that the Time Series domain has been left out of the aforementioned revolution. The lack of large scale pretrained models could be one of the reasons for this.While there have been prior experiments using pre-trained models for time series forecasting, the scale of the dataset has been relatively small. One of the few time series problems with large scale data available for pre-training is the financial domain. Therefore, this paper takes advantage of this and pretrains a ID CNN using a dataset of 728 US Stock Daily Closing Price Data in total, 2,533,901 rows. Then, we fine-tune and evaluate a dataset of the NIFTY 200 stocks' Closing Prices, in total 166,379 rows. Our results show a 32% improvement in RMSE and a 36% improvement in convergence speed when compared to a baseline non pre trained model. 2023 IEEE.