检索结果-内蒙古大学图书馆

EPDTNet + -EM: Advanced Transfer Learning and SubNet Architecture for Medical Image Diagnosis

Cognitive Computation 2025年第2期17卷 1-22页

作者： K, Dhivya K, Sangamithrai S, Indra Priyadharshini M, Vedaraj Department of Electronics and Communication Engineering SRM Institute of Science and Technology Tamilnadu Kattankulathur India Department of Computer Science and Engineering Vel Tech Rangarajan Dr. Sagunthala R & d Institute of Science and Technology Tamilnadu Chennai India School of Computer Science and Engineering Vellore Institute of Technology Chennai India Department of Computer Science and Engineering R.M.D. Engineering College RSM Nagar Tamilnadu Kavaraipettai Thiruvallur601206 India

This paper presents a novel medical imaging framework, Efficient Parallel Deep Transfer SubNet+-based Explainable Model (EPDTNet + -EM), designed to improve the detection and classification of abnormalities in medical images. The model integrates Enhanced Exponential Transfer Learning (En-ETL) and the Parallel SubNet (PSNet +) architecture, addressing challenges like overfitting, high computational cost, and limited generalization across complex datasets. En-ETL accelerates learning and enhances precision, while PSNet + balances computational efficiency and performance through parallel convolutional layers and an axis attention mechanism. The proposed framework preprocesses medical images using resizing, noise removal, and contrast enhancement, ensuring high-quality inputs for analysis. For classification, the PSNet + model categorizes medical images into normal and abnormal cases. Explainable AI (XAI) is applied to highlight key features in classified images, improving interpretability and clinical decision-making. Experimental results validate the framework on various datasets, including MRI, CT, X-ray, and mammography images. The proposed model achieves a classification accuracy of 98.83%, a false positive rate of 2%, and an execution time of 5.3 ms. Comparative Analysis show superior performance over existing approaches in classification accuracy and efficiency. Ablation studies confirmed the significant contributions of the proposed model using each component in enhancing performance. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2025.

关键词： Transfer learning

来源：评论

学校读者我要写书评

暂无评论

Incorporation of Two-Fold Trust in Group Recommender System to Handle Popularity Bias

引用

SN computer science 2024年第2期5卷 259页

作者： Richa Bedi, Punam Gupta, Shivani Kumar, Ranjeet Department of Computer Science and Engineering Birla Institute of Technology Mesra Jharkhand Ranchi India Department of Computer Science University of Delhi Delhi New Delhi India School of Computer Science Engineering and Technology Vellore Institute of Technology Tamilnadu Chennai India School of Electronics Engineering Vellore Institute of Technology Tamilnadu Chennai India

In recent years, there has been a rise in work which focuses on various issues of recommender systems other than accuracy. Popularity bias is one factor that causes the list of recommendations to deviate from the user’s expectation and causes biased list to generate. This paper proposed an approach that handles the popularity bias in trust aware group recommender system using trust propagation. The proposed work considers two factors, novelty and serendipity, to evaluate the presence of bias. By incorporating trust, along with novelty, diversity, and unexpectedness, the diverse list of recommendation that maintains accuracy should be generated. Trust propagation combines trustworthy user’s opinion indirectly involves enhancing the accuracy. The trust propagation and randomness in trust propagation ensures the trade-off between the accuracy and bias of the recommendations. A prototype is developed using the tourism dataset and compared using traditional approach of recommendation using various accuracy measures and coverage. © The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd 2024.

关键词： Diversity Novelty Popularity bias Recommender system Relevancy

来源：评论

学校读者我要写书评

暂无评论

Exploring News-Feed Credibility using Emerging Machine Learning and Deep Learning Models

引用

Journal of engineering science and technology Review 2022年第3期15卷 31-37页

作者： Aju, D. Kumar, Kakelli Anil Lal, Anisha M School of Computer Science and Engineering Vellore Institute of Technology TN Vellore India

In the recent past, the phenomenal growth, availability, and access of information on social media have made it perplexing to discern between real and fake information. The faster and easier dissemination of information through various means has accelerated the explosive growth of its falsehood. At the same time, the newsfeeds and their credibility in the social networks are in danger since the fake news is alarmingly disseminating very fast. Henceforth, the credibility of the newsfeeds or any information has become a real research challenge to cross-check the respective news or information. The cross verification can be performed concerning its source, the exact content, and the respective publisher to catalog it into fact or fake. Despite a few constraints, machine learning plays a crucial and significant role in classifying the respective news feeds. Various machine learning methodologies such as BLR (Bilinear logistic regression), NB (Naive Bayes), SVM (Support Vector Machine), and RF (Random Forest) have been reviewed and experimented with for detecting the fact and fake news feeds. After the experimentation, the limitations of the respective machine learning methods were explored and noted. Henceforth, a deep learning method called BERT is implemented and it is observed that BERT provides better efficiency than the machine learning algorithms. Furthermore, it is ascertained that the deep learning method provided the best accuracy of 99.79 % with the available dataset. © 2022. School of science, IHU. All rights reserved.

关键词： Support vector machines

来源：评论

学校读者我要写书评

暂无评论

Fused YOLO and Traditional Features for Emotion Recognition From Facial Images of Tamil and Russian Speaking Children: A Cross-Cultural Study

引用

IEEE Access 2025年 13卷 86828-86840页

作者： Mekala, A. Mary Varalakshmi, M. Gowda, C. P. Achyutha Kumar, Leti Manish Lyakso, Elena E. Frolova, Olga Nersisson, Ruban Vellore Institute of Technology School of Computer Science and Engineering Vellore632014 India Bms Institute of Technology and Management Karnataka Bengaluru560064 India Saint Petersburg State University Department of Higher Nervous Activity and Psychophysiology Saint Petersburg199034 Russia Vellore Institute of Technology School of Electrical Engineering Vellore632014 India

Cross-cultural study that avoids bias in the formulation of emotion recognition models is indispensable to address the challenges in facial emotion classification for children, for it being a relatively difficult task. With evidences from literature for the hybrid feature extraction approaches to improve the image classification accuracy, this research work focuses on developing a hybrid framework for emotion recognition from facial images of Tamil and Russian children. The dataset is audio video recording of 28 Tamil speaking and 64 Russian speaking children. The data is collected in a controlled environment and labelled by experts. Traditional features like Grey Level Cooccurrence Matrix (GLCM) and facial landmark are extracted and are fused with You Look Only Once (YOLO V5) features. While facial landmarks and GLCM provide useful information about the facial expressions and texture of the image, YOLO V5 being a single-stage object detector makes the hybrid model super-fast and achieve high accuracy in detecting small objects and in low-light settings. The different classifiers that includes KNN, SVM, Random Forest, XGBoost, and Multilayer Perceptron, is employed yielding the accuracy result of 93%, 86%, 89%, 88%,and 90%. The use of majority voting ensemble of heterogeneous classifiers for the final prediction strengthens the model further, yielding an accuracy as high as 96% for the custom cross-cultural dataset, consisting of facial images of Russian and Indian children. This is consistent with the results obtained for Indian and Russian datasets. Further, the ablation study unveils the effect of feature fusion in boosting the performance and the dominance of YOLO V5 features over the other two. © 2025 IEEE.

关键词： Feature Selection

来源：评论

学校读者我要写书评

暂无评论

Comparative Analysis of Classifiers in a Plant Recommendation System based on Environmental Factors

Comparative Analysis of Classifiers in a Plant Recommendatio...

引用

2023 International Conference on Intelligent Data Communication Technologies and Internet of Things, IDCIoT 2023

作者： Mittal, Rohan Mandava, Sreenitya Shetty, Tanmay S. Patel, Harshita Vellore Institute of Technology School of Computer Science and Engineering Vellore India Vellore Institute of Technology School of Information Technology and Engineering Vellore India

ISBN: (纸本)9781665474511

With the growing demand for reforestation and a sustainable neighborhood, everyone has begun to grow their own plants. However, the survival of a plant depends on many factors. A common problem faced by general customers is that their purchased plants, in gardens or balconies, fail to live long. This might happen because of many reasons, but the most recurrent one is the plant not adapting to the environmental conditions. Thus, personalizing the plant preferences is essential for users, so that they can buy the plants with high confidence of them surviving long. Here, this research work intends to develop an application with various filtering options, to determine the environmental conditions of the location, and the quality of lifestyle the plants can be provided with. To do so, we performed a comparison of the popular classification algorithms and found that the Random Forest Classifier served our purpose, successfully training an AI Model for predicting plants suiting the given conditions. © 2023 IEEE.

关键词： Support vector machines

来源：评论

学校读者我要写书评

暂无评论

Automatic text inpainting and quality elevation in video sequences

引用

Multimedia Tools and Applications 2024年 1-34页

作者： Palivela, Lakshmi Harika Dharmalingam, Vivekanandan Gayathri, D. Bala School of Computer Science and Engineering Vellore Institute of Technology Chennai Vandalur-Kelambakkam Road Chennai600127 India Department of Information Technology Madras Institute of Technology Anna University Chennai India Department of Computer Science and Engineering B.S.Abdur Rahman Crescent Institute Of Science and Technology Chennai India

Scene text removal is a recent development in computer vision that replaces text patches in natural images with the appropriate background. Text removal is a difficult process leading to faulty areas of text containing text strokes with their hazy backgrounds. Text in the real world uses a variety of font kinds, some of which are difficult to localize due to their chaotic shapes, varied shading degrees, and orientation *** text erasing may include the subtasks of text detection as well as text inpainting. Both subtasks require a large amount of data to be successful;but, existing approaches were limited by insufficient real-world data for scene-text elimination. Eventhough the existing works produced considerable performance improvement in scene text removal, they often leave many text remains like text strokes, thus producinglow-quality visual outcomes. Therefore, this paper proposes an automatic text inpainting and video quality elevation model by using the Improved Convolutional Network-based ***, the video samples are collected from the diverse datasets and then converted into frames. Next, the frames are deblurred using an enhanced Convolutional Neural Network (CNN) model that has three convolutional layers for accurately localizing the texts in frames. Subsequently, the texts are detected by utilizing the CLARA-based VGG-16 network. Afterward, the text strokes are removed using a convolutional Encoder and decoder network to eliminate the presence of text on complex backgrounds and textures. Here, the coordinates of text in the deblurred frames are used to crop out the text stroke regions. So, the texts are in-painted, and then, the text in-painted regions are pasted back to their original positions in the frames. Furthermore, the video quality is elevated with the help of the DenseNet-centric Enhancement network. The experimental outcomes demonstrate that the proposed model effectively removed scene texts and enhanced the video qu

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Detecting potholes with a machine learning algorithm for road safety 3

Detecting potholes with a machine learning algorithm for roa...

引用

3rd IEEE North Karnataka Subsection Flagship International Conference, NKCon 2024

作者： Singh, Yuvraj Roy, Sritama Vellore Institute of Technology School of Computer Science and Engineering Chennai India Vellore Institute of Technology School of Electronics Engineering Chennai600127 India

ISBN: (纸本)9798350364569

An important part of a country's infrastructure are its roads, which provide a vast transportation network necessary for the movement of people and goods. Road transport is the most cost-effective means of transport in India because it is widely available in densely populated areas. On the other hand, potholes continue to be a major problem that cause delays in transit and, regrettably, fatalities. A system and corresponding algorithms have been developed to detect the probability of potholes in order to tackle this problem. With the least amount of work, Single Shot Detection (SSD) technology is essential to accomplishing the goal. SSD's fast object identification and localisation in photos, along with its real-time processing capabilities, make it possible to detect potholes on road surfaces with efficiency. By utilising SSD technology, the proposed system offers a dependable means of reducing the detrimental effects of potholes on traffic safety and the efficiency of transportation. © 2024 IEEE.

关键词： Motor transportation

来源：评论

学校读者我要写书评

暂无评论

Drought and Salinity Stress Classification in Soyabean Crops: Comparative Analysis of Machine Learning Models

Drought and Salinity Stress Classification in Soyabean Crops...

引用

2024 International Conference on Signal Processing, Computation, Electronics, Power and Telecommunication, IConSCEPT 2024

作者： Madhumitaa, P.S. Ragavi, Chagamreddy Kiranmayi, Chinthamani Venkatesan, M. Prabhavathy, P. National Institute of Technology Department of Computer Science & Engineering Puducherry India Vellore Institute of Technology Department of Information Technology Vellore India

ISBN: (纸本)9798331540685

Soyabean cultivation encounters significant obstacles attributed to abiotic stresses like drought and salinity, especially in areas characterized by tropical climates. In this paper, we seek to fill gaps identified in previous research efforts, particularly the lack of comprehensive analysis regarding genotypic factors and the underutilization of morphological characteristics in machine learning algorithms for Soyabean genotype selection. We propose a novel approach to bridge these gaps by employing six morphological variables in machine learning models. Through the process of training the models using specific parameters, plants are categorized into either the Control group or the Stress Group, which comprises of genotypes facing drought or salinity stress. Various models such as Random Forest Classifier (RFC), Support Vector Machine (SVM), Gradient Boosting (GB), k-Nearest Neighbors (kNN) and Naive Bayes algorithms are included, offering an in-depth comparative analysis of the performance and feature importance under each model used. However, the Support Vector Machine was identified as the most accurate algorithm with an accuracy of 96.79% in stress classification of Soyabean genotypes. Incorporating morphological parameters in machine learning models enhances the accuracy and efficiency of Stress Classification process, increasing the sustainability of Soyabean production and its ability to withstand environmental adversities. © 2024 IEEE.

关键词： Abiotic

来源：评论

学校读者我要写书评

暂无评论

Smart IoT-Based Emotion Sensing Wearable: Revolutionizing Emotional Well-Being 2

Smart IoT-Based Emotion Sensing Wearable: Revolutionizing Em...

引用

2nd International Conference on Recent Trends in Microelectronics, Automation, Computing, and Communications Systems, ICMACC 2024

作者： Kannapiran, Pritika Roy, Sritama School of Computer Science and Engineering Vellore Institute of Technology Chennai India School of Electronics Engineering Vellore Institute of Technology Chennai600127 India

ISBN: (纸本)9798350366570

In the fast-paced world of today, mental health has become a critical issue, with emotional stability being a key component of general health. Traditional methods of detecting and managing emotions often fall short due to their reliance on verbal and facial cues, leaving individuals with conditions such as Alexithymia or Autism Spectrum Disorder (ASD) at a disadvantage. This paper introduces a novel solution: the Smart IoT-Based Emotion Sensing Wearable, is designed to revolutionize emotional sensing and management. The primary goal of the research is to create a wearable gadget that can track vital signs and instantly identify different emotions. Using the Internet of Things (IoT), machine learning, and cloud computing, this wearable device bridges the gap for people with special needs and improves emotional understanding for all users by providing real-time emotion recognition and targeted help. © 2024 IEEE.

关键词： Emotion Recognition

来源：评论

学校读者我要写书评

暂无评论

Single-level Discrete Two Dimensional Wavelet Transform Based Multiscale Deep Learning Framework for Two-Wheeler Helmet Classification

EAI Endorsed Transactions on Industrial Networks and Intelli...

引用

EAI Endorsed Transactions on Industrial Networks and Intelligent Systems 2025年第2期12卷 1-13页

作者： Annadurai, Amrutha Prusty, Manas Ranjan Pandey, Trilok Nath Patra, Subhra Rani School of Computer Science and Engineering Vellore Institute of Technology Chennai India Centre for Cyber Physical Systems Vellore Institute of Technology Chennai India Information Systems and Operations Management University of Texas at Arlington Texas 76019 United States

INTRODUCTION: A robust method is proposed in this paper to detect helmet usage in two-wheeler riders to enhance road safety. OBJECTIVES: This involves a custom made dataset that contains 1000 images captured under diverse real-world scenarios, including variations in helmet size, colour, and lighting conditions. This dataset has two classes namely with helmet and without helmet. METHODS: The proposed helmet classification approach utilizes the Multi-Scale Deep Convolutional Neural Network (CNN) framework cascaded with Long Short-Term Memory (LSTM) network. Initially the Multi-Scale Deep CNN extracts modes by applying Single-level Discrete 2D Wavelet Transform (dwt2) to decompose the original images. In particular, four different modes are used for segmenting a single image namely approximation, horizontal detail, vertical detail and diagonal detail. After feeding the segmented images into a Multi-Scale Deep CNN model, it is cascaded with an LSTM network. RESULTS: The proposed model achieved accuracies of 99.20% and 95.99% using both 5-Fold Cross-Validation (CV) and Hold-out CV methods, respectively. CONCLUSION: This result was better than the CNN-LSTM, dwt2-LSTM and a tailor made CNN model. © 2025 A. Annadurai et al.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：